d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
8fba8672e6
freebsd-skq/sys/netinet/sctputil.c

7132 lines
193 KiB
C
Raw Normal View History

Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*-
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
* Copyright (c) 2001-2008, by Cisco Systems, Inc. All rights reserved.
Use consistent text at the begining of the files. MFC after: 3 days
2012-05-23 11:26:28 +00:00
* Copyright (c) 2008-2012, by Randall Stewart. All rights reserved.
* Copyright (c) 2008-2012, by Michael Tuexen. All rights reserved.
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* a) Redistributions of source code must retain the above copyright notice,
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
* this list of conditions and the following disclaimer.
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
*
* b) Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
* the documentation and/or other materials provided with the distribution.
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
*
* c) Neither the name of Cisco Systems, Inc. nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
* THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
* THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <netinet/sctp_os.h>
#include <netinet/sctp_pcb.h>
#include <netinet/sctputil.h>
#include <netinet/sctp_var.h>
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
#include <netinet/sctp_sysctl.h>
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#ifdef INET6
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
#include <netinet6/sctp6_var.h>
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#endif
#include <netinet/sctp_header.h>
#include <netinet/sctp_output.h>
#include <netinet/sctp_uio.h>
#include <netinet/sctp_timer.h>
#include <netinet/sctp_indata.h>/* for sctp_deliver_data() */
#include <netinet/sctp_auth.h>
#include <netinet/sctp_asconf.h>
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
#include <netinet/sctp_bsd_addr.h>
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
#include <netinet/udp.h>
#include <netinet/udp_var.h>
#include <sys/proc.h>
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Name change SCTP_KTR_SUBSYS -> KTR_SCTP
2007-06-15 20:54:12 +00:00
#ifndef KTR_SCTP
#define KTR_SCTP KTR_SUBSYS
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Add const to several constants. Thanks to Nicholas Nethercote for providing the patch via https://bugzilla.mozilla.org/show_bug.cgi?id=1255655 MFC after: 1 week
2016-03-23 13:28:04 +00:00
extern const struct sctp_cc_functions sctp_cc_functions[];
extern const struct sctp_ss_functions sctp_ss_functions[];
Cleanup the management of CC functions. MFC after: 3 months.
2011-01-19 22:10:35 +00:00
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
void
Small cleanups. No functional change. MFC after: 10 days
2012-09-14 18:32:20 +00:00
sctp_sblog(struct sockbuf *sb, struct sctp_tcb *stcb, int from, int incr)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
sctp_clog.x.sb.stcb = stcb;
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
sctp_clog.x.sb.so_sbcc = sb->sb_cc;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb)
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
sctp_clog.x.sb.stcb_sbcc = stcb->asoc.sb_cc;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
else
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.sb.stcb_sbcc = 0;
sctp_clog.x.sb.incr = incr;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_SB,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_log_closing(struct sctp_inpcb *inp, struct sctp_tcb *stcb, int16_t loc)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
sctp_clog.x.close.inp = (void *)inp;
sctp_clog.x.close.sctp_flags = inp->sctp_flags;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.close.stcb = (void *)stcb;
sctp_clog.x.close.state = (uint16_t) stcb->asoc.state;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.close.stcb = 0;
sctp_clog.x.close.state = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.close.loc = loc;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_CLOSE,
0,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
rto_logging(struct sctp_nets *net, int from)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
- Define changes in sctp.h - Bug in CA that does not get us incrementing the PBA properly which made us more conservative. - comment updated in sctp_input.c - memsets added before we log - added arg to hmac id's MFC after: 2 weeks
2008-05-20 09:51:36 +00:00
memset(&sctp_clog, 0, sizeof(sctp_clog));
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.rto.net = (void *)net;
* Cleanup the code computing the retransmission timeout. * Fix an initialization bug for the scaled variance of the RTO. MFC after: 3 months.
2011-02-24 22:36:40 +00:00
sctp_clog.x.rto.rtt = net->rtt / 1000;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_RTT,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
Turns out we would reset the TSN seq counter during a colliding INIT. This if fine except when we have data outstanding... we basically reset it to the previous value it was.. so then we end up assigning the same TSN to two different data chunks. This patch: 1) Finds a missing lock for when we change the stream numbers during COOKIE and INIT-ACK processing.. we were NOT locking the send_buffer.. which COULD cause problems (found by inspection looking for <2>) 2) Fixes a case during a colliding INIT where we incorrectly reset the sending Sequence thus in some cases duplicately assigning a TSN. 3) Additional enhancments to logging so we can see strm/tsn in the receiver AND new tracking to watch what the sender is doing with TSN and STRM seq's. Approved by: gnn
2006-11-11 15:59:01 +00:00
sctp_log_strm_del_alt(struct sctp_tcb *stcb, uint32_t tsn, uint16_t sseq, uint16_t stream, int from)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
sctp_clog.x.strlog.stcb = stcb;
sctp_clog.x.strlog.n_tsn = tsn;
sctp_clog.x.strlog.n_sseq = sseq;
sctp_clog.x.strlog.e_tsn = 0;
sctp_clog.x.strlog.e_sseq = 0;
sctp_clog.x.strlog.strm = stream;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_STRM,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_log_nagle_event(struct sctp_tcb *stcb, int action)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
sctp_clog.x.nagle.stcb = (void *)stcb;
sctp_clog.x.nagle.total_flight = stcb->asoc.total_flight;
sctp_clog.x.nagle.total_in_queue = stcb->asoc.total_output_queue_size;
sctp_clog.x.nagle.count_in_queue = stcb->asoc.chunks_on_out_queue;
sctp_clog.x.nagle.count_in_flight = stcb->asoc.total_flight_count;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_NAGLE,
action,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_log_sack(uint32_t old_cumack, uint32_t cumack, uint32_t tsn, uint16_t gaps, uint16_t dups, int from)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
sctp_clog.x.sack.cumack = cumack;
sctp_clog.x.sack.oldcumack = old_cumack;
sctp_clog.x.sack.tsn = tsn;
sctp_clog.x.sack.numGaps = gaps;
sctp_clog.x.sack.numDups = dups;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_SACK,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_log_map(uint32_t map, uint32_t cum, uint32_t high, int from)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
- Define changes in sctp.h - Bug in CA that does not get us incrementing the PBA properly which made us more conservative. - comment updated in sctp_input.c - memsets added before we log - added arg to hmac id's MFC after: 2 weeks
2008-05-20 09:51:36 +00:00
memset(&sctp_clog, 0, sizeof(sctp_clog));
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.map.base = map;
sctp_clog.x.map.cum = cum;
sctp_clog.x.map.high = high;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_MAP,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
Small cleanups. No functional change. MFC after: 10 days
2012-09-14 18:32:20 +00:00
sctp_log_fr(uint32_t biggest_tsn, uint32_t biggest_new_tsn, uint32_t tsn, int from)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
- Define changes in sctp.h - Bug in CA that does not get us incrementing the PBA properly which made us more conservative. - comment updated in sctp_input.c - memsets added before we log - added arg to hmac id's MFC after: 2 weeks
2008-05-20 09:51:36 +00:00
memset(&sctp_clog, 0, sizeof(sctp_clog));
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.fr.largest_tsn = biggest_tsn;
sctp_clog.x.fr.largest_new_tsn = biggest_new_tsn;
sctp_clog.x.fr.tsn = tsn;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_FR,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Minimize the usage of SCTP_BUF_IS_EXTENDED. This should help Robert...
2015-01-10 20:49:57 +00:00
#ifdef SCTP_MBUF_LOGGING
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
void
sctp_log_mb(struct mbuf *m, int from)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
sctp_clog.x.mb.mp = m;
sctp_clog.x.mb.mbuf_flags = (uint8_t) (SCTP_BUF_GET_FLAGS(m));
sctp_clog.x.mb.size = (uint16_t) (SCTP_BUF_LEN(m));
sctp_clog.x.mb.data = SCTP_BUF_AT(m, 0);
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (SCTP_BUF_IS_EXTENDED(m)) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.mb.ext = SCTP_BUF_EXTEND_BASE(m);
sctp_clog.x.mb.refcnt = (uint8_t) (SCTP_BUF_EXTEND_REFCNT(m));
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.mb.ext = 0;
sctp_clog.x.mb.refcnt = 0;
}
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_MBUF,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Minimize the usage of SCTP_BUF_IS_EXTENDED. This should help Robert...
2015-01-10 20:49:57 +00:00
void
sctp_log_mbc(struct mbuf *m, int from)
{
struct mbuf *mat;
for (mat = m; mat; mat = SCTP_BUF_NEXT(mat)) {
sctp_log_mb(mat, from);
}
}
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
void
Small cleanups. No functional change. MFC after: 10 days
2012-09-14 18:32:20 +00:00
sctp_log_strm_del(struct sctp_queued_to_read *control, struct sctp_queued_to_read *poschk, int from)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (control == NULL) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("Gak log of NULL?\n");
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
}
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.strlog.stcb = control->stcb;
sctp_clog.x.strlog.n_tsn = control->sinfo_tsn;
sctp_clog.x.strlog.n_sseq = control->sinfo_ssn;
sctp_clog.x.strlog.strm = control->sinfo_stream;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (poschk != NULL) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.strlog.e_tsn = poschk->sinfo_tsn;
sctp_clog.x.strlog.e_sseq = poschk->sinfo_ssn;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.strlog.e_tsn = 0;
sctp_clog.x.strlog.e_sseq = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_STRM,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_log_cwnd(struct sctp_tcb *stcb, struct sctp_nets *net, int augment, uint8_t from)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.net = net;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb->asoc.send_queue_cnt > 255)
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.cnt_in_send = 255;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
else
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.cnt_in_send = stcb->asoc.send_queue_cnt;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb->asoc.stream_queue_cnt > 255)
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.cnt_in_str = 255;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
else
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.cnt_in_str = stcb->asoc.stream_queue_cnt;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (net) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.cwnd_new_value = net->cwnd;
sctp_clog.x.cwnd.inflight = net->flight_size;
sctp_clog.x.cwnd.pseudo_cumack = net->pseudo_cumack;
sctp_clog.x.cwnd.meets_pseudo_cumack = net->new_pseudo_cumack;
sctp_clog.x.cwnd.need_new_pseudo_cumack = net->find_pseudo_cumack;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (SCTP_CWNDLOG_PRESEND == from) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.meets_pseudo_cumack = stcb->asoc.peers_rwnd;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.cwnd_augment = augment;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_CWND,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_log_lock(struct sctp_inpcb *inp, struct sctp_tcb *stcb, uint8_t from)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
- Define changes in sctp.h - Bug in CA that does not get us incrementing the PBA properly which made us more conservative. - comment updated in sctp_input.c - memsets added before we log - added arg to hmac id's MFC after: 2 weeks
2008-05-20 09:51:36 +00:00
memset(&sctp_clog, 0, sizeof(sctp_clog));
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
if (inp) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.lock.sock = (void *)inp->sctp_socket;
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
} else {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.lock.sock = (void *)NULL;
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
}
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.lock.inp = (void *)inp;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.lock.tcb_lock = mtx_owned(&stcb->tcb_mtx);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.lock.tcb_lock = SCTP_LOCK_UNKNOWN;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (inp) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.lock.inp_lock = mtx_owned(&inp->inp_mtx);
sctp_clog.x.lock.create_lock = mtx_owned(&inp->inp_create_mtx);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.lock.inp_lock = SCTP_LOCK_UNKNOWN;
sctp_clog.x.lock.create_lock = SCTP_LOCK_UNKNOWN;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
sctp_clog.x.lock.info_lock = rw_wowned(&SCTP_BASE_INFO(ipi_ep_mtx));
Fix some CLANG warnings. One clang warning is left due to the fact that its bogus.. nam->sa_family will not change from AF_INET6 to AF_INET (but clang thinks it does ;-D)
2010-09-05 13:41:45 +00:00
if (inp && (inp->sctp_socket)) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.lock.sock_lock = mtx_owned(&(inp->sctp_socket->so_rcv.sb_mtx));
sctp_clog.x.lock.sockrcvbuf_lock = mtx_owned(&(inp->sctp_socket->so_rcv.sb_mtx));
sctp_clog.x.lock.socksndbuf_lock = mtx_owned(&(inp->sctp_socket->so_snd.sb_mtx));
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.lock.sock_lock = SCTP_LOCK_UNKNOWN;
sctp_clog.x.lock.sockrcvbuf_lock = SCTP_LOCK_UNKNOWN;
sctp_clog.x.lock.socksndbuf_lock = SCTP_LOCK_UNKNOWN;
}
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_LOCK_EVENT,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_log_maxburst(struct sctp_tcb *stcb, struct sctp_nets *net, int error, int burst, uint8_t from)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
- Define changes in sctp.h - Bug in CA that does not get us incrementing the PBA properly which made us more conservative. - comment updated in sctp_input.c - memsets added before we log - added arg to hmac id's MFC after: 2 weeks
2008-05-20 09:51:36 +00:00
memset(&sctp_clog, 0, sizeof(sctp_clog));
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.net = net;
sctp_clog.x.cwnd.cwnd_new_value = error;
sctp_clog.x.cwnd.inflight = net->flight_size;
sctp_clog.x.cwnd.cwnd_augment = burst;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb->asoc.send_queue_cnt > 255)
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.cnt_in_send = 255;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
else
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.cnt_in_send = stcb->asoc.send_queue_cnt;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb->asoc.stream_queue_cnt > 255)
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.cnt_in_str = 255;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
else
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.cwnd.cnt_in_str = stcb->asoc.stream_queue_cnt;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_MAXBURST,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_log_rwnd(uint8_t from, uint32_t peers_rwnd, uint32_t snd_size, uint32_t overhead)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
sctp_clog.x.rwnd.rwnd = peers_rwnd;
sctp_clog.x.rwnd.send_size = snd_size;
sctp_clog.x.rwnd.overhead = overhead;
sctp_clog.x.rwnd.new_rwnd = 0;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_RWND,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_log_rwnd_set(uint8_t from, uint32_t peers_rwnd, uint32_t flight_size, uint32_t overhead, uint32_t a_rwndval)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
sctp_clog.x.rwnd.rwnd = peers_rwnd;
sctp_clog.x.rwnd.send_size = flight_size;
sctp_clog.x.rwnd.overhead = overhead;
sctp_clog.x.rwnd.new_rwnd = a_rwndval;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_RWND,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Minimize the usage of SCTP_BUF_IS_EXTENDED. This should help Robert...
2015-01-10 20:49:57 +00:00
#ifdef SCTP_MBCNT_LOGGING
static void
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_log_mbcnt(uint8_t from, uint32_t total_oq, uint32_t book, uint32_t total_mbcnt_q, uint32_t mbcnt)
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
sctp_clog.x.mbcnt.total_queue_size = total_oq;
sctp_clog.x.mbcnt.size_change = book;
sctp_clog.x.mbcnt.total_queue_mb_size = total_mbcnt_q;
sctp_clog.x.mbcnt.mbcnt_change = mbcnt;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_MBCNT,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Minimize the usage of SCTP_BUF_IS_EXTENDED. This should help Robert...
2015-01-10 20:49:57 +00:00
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
void
sctp_misc_ints(uint8_t from, uint32_t a, uint32_t b, uint32_t c, uint32_t d)
{
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_MISC_EVENT,
from,
a, b, c, d);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
sctp_wakeup_log(struct sctp_tcb *stcb, uint32_t wake_cnt, int from)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
sctp_clog.x.wake.stcb = (void *)stcb;
sctp_clog.x.wake.wake_cnt = wake_cnt;
sctp_clog.x.wake.flight = stcb->asoc.total_flight_count;
sctp_clog.x.wake.send_q = stcb->asoc.send_queue_cnt;
sctp_clog.x.wake.sent_q = stcb->asoc.sent_queue_cnt;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb->asoc.stream_queue_cnt < 0xff)
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.wake.stream_qcnt = (uint8_t) stcb->asoc.stream_queue_cnt;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
else
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.wake.stream_qcnt = 0xff;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb->asoc.chunks_on_out_queue < 0xff)
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.wake.chunks_on_oque = (uint8_t) stcb->asoc.chunks_on_out_queue;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
else
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.wake.chunks_on_oque = 0xff;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.wake.sctpflags = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* set in the defered mode stuff */
if (stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_DONT_WAKE)
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.wake.sctpflags |= 1;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_WAKEOUTPUT)
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.wake.sctpflags |= 2;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_WAKEINPUT)
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.wake.sctpflags |= 4;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* what about the sb */
if (stcb->sctp_socket) {
struct socket *so = stcb->sctp_socket;
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.wake.sbflags = (uint8_t) ((so->so_snd.sb_flags & 0x00ff));
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_clog.x.wake.sbflags = 0xff;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_WAKE,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
sctp_log_block(uint8_t from, struct sctp_association *asoc, size_t sendlen)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
struct sctp_cwnd_log sctp_clog;
sctp_clog.x.blk.onsb = asoc->total_output_queue_size;
sctp_clog.x.blk.send_sent_qcnt = (uint16_t) (asoc->send_queue_cnt + asoc->sent_queue_cnt);
sctp_clog.x.blk.peer_rwnd = asoc->peers_rwnd;
sctp_clog.x.blk.stream_qcnt = (uint16_t) asoc->stream_queue_cnt;
sctp_clog.x.blk.chunks_on_oque = (uint16_t) asoc->chunks_on_out_queue;
sctp_clog.x.blk.flight_size = (uint16_t) (asoc->total_flight / 1024);
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
sctp_clog.x.blk.sndlen = (uint32_t) sendlen;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_CTR6(KTR_SCTP, "SCTP:%d[%d]:%x-%x-%x-%x",
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
SCTP_LOG_EVENT_BLOCK,
from,
sctp_clog.x.misc.log1,
sctp_clog.x.misc.log2,
sctp_clog.x.misc.log3,
sctp_clog.x.misc.log4);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
int
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
sctp_fill_stat_log(void *optval SCTP_UNUSED, size_t *optsize SCTP_UNUSED)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
/* May need to fix this if ktrdump does not work */
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (0);
}
#ifdef SCTP_AUDITING_ENABLED
uint8_t sctp_audit_data[SCTP_AUDIT_SIZE][2];
static int sctp_audit_indx = 0;
static
void
sctp_print_audit_report(void)
{
int i;
int cnt;
cnt = 0;
for (i = sctp_audit_indx; i < SCTP_AUDIT_SIZE; i++) {
if ((sctp_audit_data[i][0] == 0xe0) &&
(sctp_audit_data[i][1] == 0x01)) {
cnt = 0;
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("\n");
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else if (sctp_audit_data[i][0] == 0xf0) {
cnt = 0;
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("\n");
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else if ((sctp_audit_data[i][0] == 0xc0) &&
(sctp_audit_data[i][1] == 0x01)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("\n");
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
cnt = 0;
}
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("%2.2x%2.2x ", (uint32_t) sctp_audit_data[i][0],
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
(uint32_t) sctp_audit_data[i][1]);
cnt++;
if ((cnt % 14) == 0)
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("\n");
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
for (i = 0; i < sctp_audit_indx; i++) {
if ((sctp_audit_data[i][0] == 0xe0) &&
(sctp_audit_data[i][1] == 0x01)) {
cnt = 0;
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("\n");
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else if (sctp_audit_data[i][0] == 0xf0) {
cnt = 0;
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("\n");
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else if ((sctp_audit_data[i][0] == 0xc0) &&
(sctp_audit_data[i][1] == 0x01)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("\n");
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
cnt = 0;
}
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("%2.2x%2.2x ", (uint32_t) sctp_audit_data[i][0],
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
(uint32_t) sctp_audit_data[i][1]);
cnt++;
if ((cnt % 14) == 0)
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("\n");
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("\n");
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_auditing(int from, struct sctp_inpcb *inp, struct sctp_tcb *stcb,
struct sctp_nets *net)
{
int resend_cnt, tot_out, rep, tot_book_cnt;
struct sctp_nets *lnet;
struct sctp_tmit_chunk *chk;
sctp_audit_data[sctp_audit_indx][0] = 0xAA;
sctp_audit_data[sctp_audit_indx][1] = 0x000000ff & from;
sctp_audit_indx++;
if (sctp_audit_indx >= SCTP_AUDIT_SIZE) {
sctp_audit_indx = 0;
}
if (inp == NULL) {
sctp_audit_data[sctp_audit_indx][0] = 0xAF;
sctp_audit_data[sctp_audit_indx][1] = 0x01;
sctp_audit_indx++;
if (sctp_audit_indx >= SCTP_AUDIT_SIZE) {
sctp_audit_indx = 0;
}
return;
}
if (stcb == NULL) {
sctp_audit_data[sctp_audit_indx][0] = 0xAF;
sctp_audit_data[sctp_audit_indx][1] = 0x02;
sctp_audit_indx++;
if (sctp_audit_indx >= SCTP_AUDIT_SIZE) {
sctp_audit_indx = 0;
}
return;
}
sctp_audit_data[sctp_audit_indx][0] = 0xA1;
sctp_audit_data[sctp_audit_indx][1] =
(0x000000ff & stcb->asoc.sent_queue_retran_cnt);
sctp_audit_indx++;
if (sctp_audit_indx >= SCTP_AUDIT_SIZE) {
sctp_audit_indx = 0;
}
rep = 0;
tot_book_cnt = 0;
resend_cnt = tot_out = 0;
TAILQ_FOREACH(chk, &stcb->asoc.sent_queue, sctp_next) {
if (chk->sent == SCTP_DATAGRAM_RESEND) {
resend_cnt++;
} else if (chk->sent < SCTP_DATAGRAM_RESEND) {
tot_out += chk->book_size;
tot_book_cnt++;
}
}
if (resend_cnt != stcb->asoc.sent_queue_retran_cnt) {
sctp_audit_data[sctp_audit_indx][0] = 0xAF;
sctp_audit_data[sctp_audit_indx][1] = 0xA1;
sctp_audit_indx++;
if (sctp_audit_indx >= SCTP_AUDIT_SIZE) {
sctp_audit_indx = 0;
}
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("resend_cnt:%d asoc-tot:%d\n",
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
resend_cnt, stcb->asoc.sent_queue_retran_cnt);
rep = 1;
stcb->asoc.sent_queue_retran_cnt = resend_cnt;
sctp_audit_data[sctp_audit_indx][0] = 0xA2;
sctp_audit_data[sctp_audit_indx][1] =
(0x000000ff & stcb->asoc.sent_queue_retran_cnt);
sctp_audit_indx++;
if (sctp_audit_indx >= SCTP_AUDIT_SIZE) {
sctp_audit_indx = 0;
}
}
if (tot_out != stcb->asoc.total_flight) {
sctp_audit_data[sctp_audit_indx][0] = 0xAF;
sctp_audit_data[sctp_audit_indx][1] = 0xA2;
sctp_audit_indx++;
if (sctp_audit_indx >= SCTP_AUDIT_SIZE) {
sctp_audit_indx = 0;
}
rep = 1;
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("tot_flt:%d asoc_tot:%d\n", tot_out,
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
(int)stcb->asoc.total_flight);
stcb->asoc.total_flight = tot_out;
}
if (tot_book_cnt != stcb->asoc.total_flight_count) {
sctp_audit_data[sctp_audit_indx][0] = 0xAF;
sctp_audit_data[sctp_audit_indx][1] = 0xA5;
sctp_audit_indx++;
if (sctp_audit_indx >= SCTP_AUDIT_SIZE) {
sctp_audit_indx = 0;
}
rep = 1;
* Fix compilation when using SCTP_AUDITING_ENABLED. * Fix delaying of SACK by taking out old optimization code which does not optimize anymore. * Fix fast retransmission of chunks abandoned by the "number of retransmissions" policy. MFC after: 3 days.
2010-04-23 08:19:47 +00:00
SCTP_PRINTF("tot_flt_book:%d\n", tot_book_cnt);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
stcb->asoc.total_flight_count = tot_book_cnt;
}
tot_out = 0;
TAILQ_FOREACH(lnet, &stcb->asoc.nets, sctp_next) {
tot_out += lnet->flight_size;
}
if (tot_out != stcb->asoc.total_flight) {
sctp_audit_data[sctp_audit_indx][0] = 0xAF;
sctp_audit_data[sctp_audit_indx][1] = 0xA3;
sctp_audit_indx++;
if (sctp_audit_indx >= SCTP_AUDIT_SIZE) {
sctp_audit_indx = 0;
}
rep = 1;
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("real flight:%d net total was %d\n",
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
stcb->asoc.total_flight, tot_out);
/* now corrective action */
TAILQ_FOREACH(lnet, &stcb->asoc.nets, sctp_next) {
tot_out = 0;
TAILQ_FOREACH(chk, &stcb->asoc.sent_queue, sctp_next) {
if ((chk->whoTo == lnet) &&
(chk->sent < SCTP_DATAGRAM_RESEND)) {
tot_out += chk->book_size;
}
}
if (lnet->flight_size != tot_out) {
* Fix compilation when using SCTP_AUDITING_ENABLED. * Fix delaying of SACK by taking out old optimization code which does not optimize anymore. * Fix fast retransmission of chunks abandoned by the "number of retransmissions" policy. MFC after: 3 days.
2010-04-23 08:19:47 +00:00
SCTP_PRINTF("net:%p flight was %d corrected to %d\n",
Using %p in a format string requires a void *. MFC after: 10 days
2012-09-05 18:52:01 +00:00
(void *)lnet, lnet->flight_size,
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
tot_out);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
lnet->flight_size = tot_out;
}
}
}
if (rep) {
sctp_print_audit_report();
}
}
void
sctp_audit_log(uint8_t ev, uint8_t fd)
{
sctp_audit_data[sctp_audit_indx][0] = ev;
sctp_audit_data[sctp_audit_indx][1] = fd;
sctp_audit_indx++;
if (sctp_audit_indx >= SCTP_AUDIT_SIZE) {
sctp_audit_indx = 0;
}
}
#endif
Not only stop all timers when entering the SHUTDOWN_SENT state, but also when entering the SHUTDOWN_ACK_SEND state. MFC after: 3 days.
2010-11-07 14:39:40 +00:00
/*
* sctp_stop_timers_for_shutdown() should be called
* when entering the SHUTDOWN_SENT or SHUTDOWN_ACK_SENT
* state to make sure that all timers are stopped.
*/
void
sctp_stop_timers_for_shutdown(struct sctp_tcb *stcb)
{
struct sctp_association *asoc;
struct sctp_nets *net;
asoc = &stcb->asoc;
(void)SCTP_OS_TIMER_STOP(&asoc->dack_timer.timer);
(void)SCTP_OS_TIMER_STOP(&asoc->strreset_timer.timer);
(void)SCTP_OS_TIMER_STOP(&asoc->asconf_timer.timer);
(void)SCTP_OS_TIMER_STOP(&asoc->autoclose_timer.timer);
(void)SCTP_OS_TIMER_STOP(&asoc->delayed_event_timer.timer);
TAILQ_FOREACH(net, &asoc->nets, sctp_next) {
(void)SCTP_OS_TIMER_STOP(&net->pmtu_timer.timer);
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
(void)SCTP_OS_TIMER_STOP(&net->hb_timer.timer);
Not only stop all timers when entering the SHUTDOWN_SENT state, but also when entering the SHUTDOWN_ACK_SEND state. MFC after: 3 days.
2010-11-07 14:39:40 +00:00
}
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* a list of sizes based on typical mtu's, used only if next hop size not
* returned.
*/
Do not have the MTU table twice in the code. Therefore move the function from the timer code to util, rename it appropriately and also fix a bug in sctp_get_prev_mtu(), where calling it with a value existing in the MTU table did not return a smaller one. MFC after: 3 days.
2010-11-07 18:50:35 +00:00
static uint32_t sctp_mtu_sizes[] = {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
68,
296,
508,
512,
544,
576,
1006,
1492,
1500,
1536,
2002,
2048,
4352,
4464,
8166,
17914,
32000,
65535
};
Do not have the MTU table twice in the code. Therefore move the function from the timer code to util, rename it appropriately and also fix a bug in sctp_get_prev_mtu(), where calling it with a value existing in the MTU table did not return a smaller one. MFC after: 3 days.
2010-11-07 18:50:35 +00:00
/*
* Return the largest MTU smaller than val. If there is no
* entry, just return val.
*/
uint32_t
sctp_get_prev_mtu(uint32_t val)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
Do not have the MTU table twice in the code. Therefore move the function from the timer code to util, rename it appropriately and also fix a bug in sctp_get_prev_mtu(), where calling it with a value existing in the MTU table did not return a smaller one. MFC after: 3 days.
2010-11-07 18:50:35 +00:00
uint32_t i;
if (val <= sctp_mtu_sizes[0]) {
return (val);
}
for (i = 1; i < (sizeof(sctp_mtu_sizes) / sizeof(uint32_t)); i++) {
if (val <= sctp_mtu_sizes[i]) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
}
}
Do not have the MTU table twice in the code. Therefore move the function from the timer code to util, rename it appropriately and also fix a bug in sctp_get_prev_mtu(), where calling it with a value existing in the MTU table did not return a smaller one. MFC after: 3 days.
2010-11-07 18:50:35 +00:00
return (sctp_mtu_sizes[i - 1]);
}
/*
* Return the smallest MTU larger than val. If there is no
* entry, just return val.
*/
uint32_t
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
sctp_get_next_mtu(uint32_t val)
Do not have the MTU table twice in the code. Therefore move the function from the timer code to util, rename it appropriately and also fix a bug in sctp_get_prev_mtu(), where calling it with a value existing in the MTU table did not return a smaller one. MFC after: 3 days.
2010-11-07 18:50:35 +00:00
{
/* select another MTU that is just bigger than this one */
uint32_t i;
for (i = 0; i < (sizeof(sctp_mtu_sizes) / sizeof(uint32_t)); i++) {
if (val < sctp_mtu_sizes[i]) {
return (sctp_mtu_sizes[i]);
}
}
return (val);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_fill_random_store(struct sctp_pcb *m)
{
/*
* Here we use the MD5/SHA-1 to hash with our good randomNumbers and
* our counter. The result becomes our good random numbers and we
* then setup to give these out. Note that we do no locking to
* protect this. This is ok, since if competing folks call this we
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
* will get more gobbled gook in the random store which is what we
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
* want. There is a danger that two guys will use the same random
* numbers, but thats ok too since that is random as well :->
*/
m->store_at = 0;
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
(void)sctp_hmac(SCTP_HMAC, (uint8_t *) m->random_numbers,
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sizeof(m->random_numbers), (uint8_t *) & m->random_counter,
sizeof(m->random_counter), (uint8_t *) m->random_store);
m->random_counter++;
}
uint32_t
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
sctp_select_initial_TSN(struct sctp_pcb *inp)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
/*
* A true implementation should use random selection process to get
* the initial stream sequence number, using RFC1750 as a good
* guideline
*/
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
uint32_t x, *xp;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
uint8_t *p;
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
int store_at, new_store;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
if (inp->initial_sequence_debug != 0) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
uint32_t ret;
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
ret = inp->initial_sequence_debug;
inp->initial_sequence_debug++;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (ret);
}
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
retry:
store_at = inp->store_at;
new_store = store_at + sizeof(uint32_t);
if (new_store >= (SCTP_SIGNATURE_SIZE - 3)) {
new_store = 0;
}
if (!atomic_cmpset_int(&inp->store_at, store_at, new_store)) {
goto retry;
}
if (new_store == 0) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Refill the random store */
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
sctp_fill_random_store(inp);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
p = &inp->random_store[store_at];
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
xp = (uint32_t *) p;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
x = *xp;
return (x);
}
uint32_t
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
sctp_select_a_tag(struct sctp_inpcb *inp, uint16_t lport, uint16_t rport, int check)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
uint32_t x;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
struct timeval now;
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
if (check) {
(void)SCTP_GETTIME_TIMEVAL(&now);
}
for (;;) {
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
x = sctp_select_initial_TSN(&inp->sctp_ep);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (x == 0) {
/* we never use 0 */
continue;
}
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
if (!check || sctp_is_vtag_good(x, lport, rport, &now)) {
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
return (x);
}
Fix the exporting of SCTP association states to userland. Without this, associations in SHUTDOWN-PENDING were never reported correctly. MFC after: 3 weeks
2015-08-29 09:14:32 +00:00
int32_t
sctp_map_assoc_state(int kernel_state)
{
int32_t user_state;
if (kernel_state & SCTP_STATE_WAS_ABORTED) {
user_state = SCTP_CLOSED;
} else if (kernel_state & SCTP_STATE_SHUTDOWN_PENDING) {
user_state = SCTP_SHUTDOWN_PENDING;
} else {
switch (kernel_state & SCTP_STATE_MASK) {
case SCTP_STATE_EMPTY:
user_state = SCTP_CLOSED;
break;
case SCTP_STATE_INUSE:
user_state = SCTP_CLOSED;
break;
case SCTP_STATE_COOKIE_WAIT:
user_state = SCTP_COOKIE_WAIT;
break;
case SCTP_STATE_COOKIE_ECHOED:
user_state = SCTP_COOKIE_ECHOED;
break;
case SCTP_STATE_OPEN:
user_state = SCTP_ESTABLISHED;
break;
case SCTP_STATE_SHUTDOWN_SENT:
user_state = SCTP_SHUTDOWN_SENT;
break;
case SCTP_STATE_SHUTDOWN_RECEIVED:
user_state = SCTP_SHUTDOWN_RECEIVED;
break;
case SCTP_STATE_SHUTDOWN_ACK_SENT:
user_state = SCTP_SHUTDOWN_ACK_SENT;
break;
default:
user_state = SCTP_CLOSED;
break;
}
}
return (user_state);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
int
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
sctp_init_asoc(struct sctp_inpcb *inp, struct sctp_tcb *stcb,
Fix the allocation of outgoing streams: * When processing a cookie, use the number of streams announced in the INIT-ACK. * When sending an INIT-ACK for an existing association, use the value from the association, not from the end-point. MFC after: 1 week
2015-12-06 16:17:57 +00:00
uint32_t override_tag, uint32_t vrf_id, uint16_t o_strms)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
- Fix a memory overwrite when the mapping array is expanded, size of expansion was not taken int consideration. - Fix so vtag hash is 1 bigger so that it modulo's out correctly, avoids a panic when restart with right modulo happens. - do not dereference stcb when control->do_not_ref_stcb is set - Fix up packet logging to not often use a lock and also to add to options. - Fix some logging option duplication in the sctputil.h
2007-05-30 17:39:45 +00:00
struct sctp_association *asoc;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* Anything set to zero is taken care of by the allocation routine's
* bzero
*/
/*
* Up front select what scoping to apply on addresses I tell my peer
* Not sure what to do with these right now, we will need to come up
* with a way to set them. We may need to pass them through from the
* caller in the sctp_aloc_assoc() function.
*/
int i;
Add support for the SCTP_PR_STREAM_STATUS and SCTP_PR_ASSOC_STATUS socket options. This includes managing the correspoing stat counters. Add the SCTP_DETAILED_STR_STATS kernel option to control per policy counters on every stream. The default is off and only an aggregated counter is available. This is sufficient for the RTCWeb usecase. MFC after: 1 week
2014-08-13 15:50:16 +00:00
#if defined(SCTP_DETAILED_STR_STATS)
int j;
#endif
- Fix a memory overwrite when the mapping array is expanded, size of expansion was not taken int consideration. - Fix so vtag hash is 1 bigger so that it modulo's out correctly, avoids a panic when restart with right modulo happens. - do not dereference stcb when control->do_not_ref_stcb is set - Fix up packet logging to not often use a lock and also to add to options. - Fix some logging option duplication in the sctputil.h
2007-05-30 17:39:45 +00:00
asoc = &stcb->asoc;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* init all variables to a known value. */
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_SET_STATE(&stcb->asoc, SCTP_STATE_INUSE);
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->max_burst = inp->sctp_ep.max_burst;
asoc->fr_max_burst = inp->sctp_ep.fr_max_burst;
asoc->heart_beat_delay = TICKS_TO_MSEC(inp->sctp_ep.sctp_timeoutticks[SCTP_TIMER_HEARTBEAT]);
asoc->cookie_life = inp->sctp_ep.def_cookie_life;
asoc->sctp_cmt_on_off = inp->sctp_cmt_on_off;
Cleanup the ECN configuration handling and provide an SCTP socket option for controlling ECN on future associations and get the status on current associations. A simialar pattern will be used for controlling SCTP extensions in upcoming commits.
2014-08-02 17:35:13 +00:00
asoc->ecn_supported = inp->ecn_supported;
Add support for the SCTP_PR_SUPPORTED socket option as specified in http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-prpolicies Add also a sysctl controlling the default of the end-points. MFC after: 1 week
2014-08-02 21:36:40 +00:00
asoc->prsctp_supported = inp->prsctp_supported;
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
asoc->idata_supported = inp->idata_supported;
Add support for the SCTP_AUTH_SUPPORTED and SCTP_ASCONF_SUPPORTED socket options. Add also a sysctl to control the support of ASCONF. MFC after: 1 week
2014-08-12 11:30:16 +00:00
asoc->auth_supported = inp->auth_supported;
asoc->asconf_supported = inp->asconf_supported;
Add support for the SCTP_RECONFIG_SUPPORTED and the corresponding sysctl controlling the negotiation of the RE-CONFIG extension. MFC after: 3 days
2014-08-04 20:07:35 +00:00
asoc->reconfig_supported = inp->reconfig_supported;
Add SCTP socket option SCTP_NRSACK_SUPPORTED to control the NRSACK extension. The default will still be off, since it it not an RFC (yet). Changing the sysctl name will be in a separate commit. MFC after: 1 week
2014-08-03 14:10:10 +00:00
asoc->nrsack_supported = inp->nrsack_supported;
Add support for the SCTP_PKTDROP_SUPPORTED socket option and the corresponding sysctl variable. The default is off, since the specification is not an RFC yet. MFC after: 1 week
2014-08-03 18:12:55 +00:00
asoc->pktdrop_supported = inp->pktdrop_supported;
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
asoc->idata_supported = inp->idata_supported;
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
asoc->sctp_cmt_pf = (uint8_t) 0;
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->sctp_frag_point = inp->sctp_frag_point;
asoc->sctp_features = inp->sctp_features;
asoc->default_dscp = inp->sctp_ep.default_dscp;
Add a SCTP socket option to limit the cwnd for each path. MFC after: 1 month
2015-03-10 19:49:25 +00:00
asoc->max_cwnd = inp->max_cwnd;
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
#ifdef INET6
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
if (inp->sctp_ep.default_flowlabel) {
asoc->default_flowlabel = inp->sctp_ep.default_flowlabel;
Fix the handling of the flowlabel and DSCP value in the SCTP_PEER_ADDR_PARAMS socket option. Honor the net.inet6.ip6.auto_flowlabel sysctl setting. Approved by: re (bz) MFC after: 1 month.
2011-09-14 08:15:21 +00:00
} else {
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
if (inp->ip_inp.inp.inp_flags & IN6P_AUTOFLOWLABEL) {
asoc->default_flowlabel = sctp_select_initial_TSN(&inp->sctp_ep);
Fix the handling of the flowlabel and DSCP value in the SCTP_PEER_ADDR_PARAMS socket option. Honor the net.inet6.ip6.auto_flowlabel sysctl setting. Approved by: re (bz) MFC after: 1 month.
2011-09-14 08:15:21 +00:00
asoc->default_flowlabel &= 0x000fffff;
asoc->default_flowlabel |= 0x80000000;
} else {
asoc->default_flowlabel = 0;
}
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#endif
- Found a problem in non-blocking sends. When sending, once the locks are all unlocked to do the copy's in, its possible that other events could then raise the number of bytes outstanding pushing it so not all the message would fit. This would then cause us to send only part of the message. This fix makes it so we keep a "reserved" amount that can be kept in mind when making calculations to send. - rcv msg args with a NULL/NULL for to/tolen will return an error incorrectly for the 1-2-1 model. - We were not doing 0 len return correctly and not setting cantrcv more correctly. Previouly we "fixed" this area by taking out the socantrcv since we then could not get the data out. The correct rix is to still flag the socket but alow a by-pass route to continue to read until all data is consumed. MFC after: 1 week
2007-12-04 14:41:48 +00:00
asoc->sb_send_resv = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (override_tag) {
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
asoc->my_vtag = override_tag;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->my_vtag = sctp_select_a_tag(inp, stcb->sctp_ep->sctp_lport, stcb->rport, 1);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
This patch fixes a LOR that happens during INIT-ACK collision. We were calling select_a_tag() inside sctp_send_initate_ack(). During collision cases we have a stcb and thus a SCTP_LOCK. When we call select_a_tag it (below it) locks the INFO lock. We now 1) pre-select the nonce-tie-tags in sctputil.c during setup of a tcb. 2) In the other case where we have to select tags, we unlock after incr the ref cnt (so assoc won't go away0 and then do the tag selection followed by a relock and decr the refcnt. Approved by: gnn
2006-11-10 13:34:55 +00:00
/* Get the nonce tags */
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->my_vtag_nonce = sctp_select_a_tag(inp, stcb->sctp_ep->sctp_lport, stcb->rport, 0);
asoc->peer_vtag_nonce = sctp_select_a_tag(inp, stcb->sctp_ep->sctp_lport, stcb->rport, 0);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
asoc->vrf_id = vrf_id;
This patch fixes a LOR that happens during INIT-ACK collision. We were calling select_a_tag() inside sctp_send_initate_ack(). During collision cases we have a stcb and thus a SCTP_LOCK. When we call select_a_tag it (below it) locks the INFO lock. We now 1) pre-select the nonce-tie-tags in sctputil.c during setup of a tcb. 2) In the other case where we have to select tags, we unlock after incr the ref cnt (so assoc won't go away0 and then do the tag selection followed by a relock and decr the refcnt. Approved by: gnn
2006-11-10 13:34:55 +00:00
- added pre-checks to the bindx call. - use proper tick gathering macro instead of ticks directly. - Placed reasonable boundaries on sets that a user can do that are converted to ticks from ms. - Fix CMT_PF to always check to be sure CMT is on. - Fix ticks use of CMT_PF. - put back code to allow asconfs to be queued while INITs are in flight and before the assoc is established. - During window probes, an ack'd packet might be left with the window probe mark on it causing it to be retransmitted. Change so that the flight decrease macro clears the window_probe mark. - Additional logging flight size/reading and ASOC LOG. This is only enabled if you manually insert things into opt_sctp.h since its a set of debug code only. - Found an interesting SMP race in the way data was appended which could cause a reader to lose a part of a message, had to reorder when we marked the message was complete to after the data was appended. - bug in ADD-IP for the subset bound socket case when the peer has only one address - fix ASCONF implicit success/error handling case - proper support of jails in Freebsd 6> - copy out the timeval for the 64 bit sparc world on cookie-echo alignment error crashes without this). Approved by: re(Ken Smith)
2007-07-17 20:58:26 +00:00
#ifdef SCTP_ASOCLOG_OF_TSNS
asoc->tsn_in_at = 0;
asoc->tsn_out_at = 0;
asoc->tsn_in_wrapped = 0;
asoc->tsn_out_wrapped = 0;
asoc->cumack_log_at = 0;
- fix sctp_ifn initial refcount issue (prevents deletion) - fix a bug during cookie collision that prevented an association from coming up in a specific restart case. - Fix it so the shutdown-pending flag gets removed (this is more for correctness then needed) when we enter shutdown-sent or shutdown-ack-sent states. - Fix a bug that caused the receiver to sometimes NOT send a SACK when a duplicate TSN arrived. Without this fix it was possible for the association to fall down if the - Deleted primary destination is also stored when SCTP_MOBILITY_BASE. (Previously, it is stored when only SCTP_MOBILITY_FASTHANDOFF) - Fix a locking issue where we might call send_initiate_ack() and incorrectly state the lock held/not held. Also fix it so that when we release the lock the inp cannot be deleted on us. - Add the debug option that can cause the stack to panic instead of aborting an assoc. This does not and should never show up in options but is useful for debugging unexpected aborts. - Add cumack_log sent to track sending cumack information for the debug case where we are running a special log per assoc. - Added extra () aroudn sctp_sbspace macro to avoid compile warnings. MFC after: 1 week
2007-10-16 14:05:51 +00:00
asoc->cumack_log_atsnt = 0;
- added pre-checks to the bindx call. - use proper tick gathering macro instead of ticks directly. - Placed reasonable boundaries on sets that a user can do that are converted to ticks from ms. - Fix CMT_PF to always check to be sure CMT is on. - Fix ticks use of CMT_PF. - put back code to allow asconfs to be queued while INITs are in flight and before the assoc is established. - During window probes, an ack'd packet might be left with the window probe mark on it causing it to be retransmitted. Change so that the flight decrease macro clears the window_probe mark. - Additional logging flight size/reading and ASOC LOG. This is only enabled if you manually insert things into opt_sctp.h since its a set of debug code only. - Found an interesting SMP race in the way data was appended which could cause a reader to lose a part of a message, had to reorder when we marked the message was complete to after the data was appended. - bug in ADD-IP for the subset bound socket case when the peer has only one address - fix ASCONF implicit success/error handling case - proper support of jails in Freebsd 6> - copy out the timeval for the 64 bit sparc world on cookie-echo alignment error crashes without this). Approved by: re(Ken Smith)
2007-07-17 20:58:26 +00:00
#endif
#ifdef SCTP_FS_SPEC_LOG
asoc->fs_index = 0;
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
asoc->refcnt = 0;
asoc->assoc_up_sent = 0;
asoc->asconf_seq_out = asoc->str_reset_seq_out = asoc->init_seq_number = asoc->sending_seq =
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
sctp_select_initial_TSN(&inp->sctp_ep);
- Adds support for the multi-asconf (From Kozuka-san) - Adds some prepwork (Not all yet) for vimage in particular support the delete the sctppcbinfo.xx structs. There is still a leak in here if it were to be called plus we stil need the regrouping (From Me and Michael Tuexen) - Adds support for UDP tunneling. For BSD there is no socket yet setup so its disabled, but major argument changes are in here to emcompass the passing of the port number (zero when you don't have a udp tunnel, the default for BSD). Will add some hooks in UDP here shortly (discussed with Robert) that will allow easy tunneling. (Mainly from Peter Lei and Michael Tuexen with some BSD work from me :-D) - Some ease for windows, evidently leave is reserved by their compile move label leave: -> out: MFC after: 1 week
2008-05-20 13:47:46 +00:00
asoc->asconf_seq_out_acked = asoc->asconf_seq_out - 1;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* we are optimisitic here */
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
asoc->peer_supports_nat = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
asoc->sent_queue_retran_cnt = 0;
/* for CMT */
Many bug fixes (from the IETF hack-fest): - PR-SCTP had major issues when skipping through a multi-part message. o Did not look at socket buffer. o Did not properly handle the reassmebly queue. o The MARKED segments could interfere and un-skip a chunk causing a problem with the proper FWD-TSN. o No FR of FWD-TSN's was being done. - NR-Sack code was basically disabled. It needed fixes that never got into the real code. - CMT code had issues when the two paths were NOT the same b/w. We found a few small bugs, but also the critcal one here was not dividing the rwnd amongst the paths. Obtained from: Michael Tuexen and myself at the IETF hack-fest ;-)
2009-04-04 11:43:32 +00:00
asoc->last_net_cmt_send_started = NULL;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* This will need to be adjusted */
asoc->last_acked_seq = asoc->init_seq_number - 1;
asoc->advanced_peer_ack_point = asoc->last_acked_seq;
asoc->asconf_seq_in = asoc->last_acked_seq;
/* here we are different, we hold the next one we expect */
asoc->str_reset_seq_in = asoc->last_acked_seq + 1;
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->initial_init_rto_max = inp->sctp_ep.initial_init_rto_max;
asoc->initial_rto = inp->sctp_ep.initial_rto;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->max_init_times = inp->sctp_ep.max_init_times;
asoc->max_send_times = inp->sctp_ep.max_send_times;
asoc->def_net_failure = inp->sctp_ep.def_net_failure;
asoc->def_net_pf_threshold = inp->sctp_ep.def_net_pf_threshold;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
asoc->free_chunk_cnt = 0;
asoc->iam_blocking = 0;
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->context = inp->sctp_context;
asoc->local_strreset_support = inp->local_strreset_support;
asoc->def_send = inp->def_send;
asoc->delayed_ack = TICKS_TO_MSEC(inp->sctp_ep.sctp_timeoutticks[SCTP_TIMER_RECV]);
asoc->sack_freq = inp->sctp_ep.sctp_sack_freq;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
asoc->pr_sctp_cnt = 0;
asoc->total_output_queue_size = 0;
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
if (inp->sctp_flags & SCTP_PCB_FLAGS_BOUND_V6) {
asoc->scope.ipv6_addr_legal = 1;
if (SCTP_IPV6_V6ONLY(inp) == 0) {
asoc->scope.ipv4_addr_legal = 1;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->scope.ipv4_addr_legal = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
} else {
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->scope.ipv6_addr_legal = 0;
asoc->scope.ipv4_addr_legal = 1;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->my_rwnd = max(SCTP_SB_LIMIT_RCV(inp->sctp_socket), SCTP_MINIMAL_RWND);
asoc->peers_rwnd = SCTP_SB_LIMIT_RCV(inp->sctp_socket);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->smallest_mtu = inp->sctp_frag_point;
asoc->minrto = inp->sctp_ep.sctp_minrto;
asoc->maxrto = inp->sctp_ep.sctp_maxrto;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
asoc->locked_on_sending = NULL;
asoc->stream_locked_on = 0;
asoc->ecn_echo_cnt_onq = 0;
asoc->stream_locked = 0;
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
asoc->send_sack = 1;
LIST_INIT(&asoc->sctp_restricted_addrs);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
TAILQ_INIT(&asoc->nets);
TAILQ_INIT(&asoc->pending_reply_queue);
- During shutdown pending, when the last sack came in and the last message on the send stream was "null" but still there, a state we allow, we could get hung and not clean it up and wait for the shutdown guard timer to clear the association without a graceful close. Fix this so that that we properly clean up. - Added support for Multiple ASCONF per new RFC. We only (so far) accept input of these and cannot yet generate a multi-asconf. - Sysctl'd support for experimental Fast Handover feature. Always disabled unless sysctl or socket option changes to enable. - Error case in add-ip where the peer supports AUTH and ADD-IP but does NOT require AUTH of ASCONF/ASCONF-ACK. We need to ABORT in this case. - According to the Kyoto summit of socket api developers (Solaris, Linux, BSD). We need to have: o non-eeor mode messages be atomic - Fixed o Allow implicit setup of an assoc in 1-2-1 model if using the sctp_**() send calls - Fixed o Get rid of HAVE_XXX declarations - Done o add a sctp_pr_policy in hole in sndrcvinfo structure - Done o add a PR_SCTP_POLICY_VALID type flag - yet to-do in a future patch! - Optimize sctp6 calls to reuse code in sctp_usrreq. Also optimize when we close sending out the data and disabling Nagle. - Change key concatenation order to match the auth RFC - When sending OOTB shutdown_complete always do csum. - Don't send PKT-DROP to a PKT-DROP - For abort chunks just always checksums same for shutdown-complete. - inpcb_free front state had a bug where in queue data could wedge an assoc. We need to just abandon ones in front states (free_assoc). - If a peer sends us a 64k abort, we would try to assemble a response packet which may be larger than 64k. This then would be dropped by IP. Instead make a "minimum" size for us 64k-2k (we want at least 2k for our initack). If we receive such an init discard it early without all the processing. - When we peel off we must increment the tcb ref count to keep it from being freed from underneath us. - handling fwd-tsn had bugs that caused memory overwrites when given faulty data, fixed so can't happen and we also stop at the first bad stream no. - Fixed so comm-up generates the adaption indication. - peeloff did not get the hmac params copied. - fix it so we lock the addr list when doing src-addr selection (in future we need to use a multi-reader/one writer lock here) - During lowlevel output, we could end up with a _l_addr set to null if the iterator is calling the output routine. This means we would possibly crash when we gather the MTU info. Fix so we only do the gather where we have a src address cached. - we need to be sure to set abort flag on conn state when we receive an abort. - peeloff could leak a socket. Moved code so the close will find the socket if the peeloff fails (uipc_syscalls.c) Approved by: re@freebsd.org(Ken Smith)
2007-08-27 05:19:48 +00:00
TAILQ_INIT(&asoc->asconf_ack_sent);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Setup to fill the hb random cache at first HB */
asoc->hb_random_idx = 4;
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->sctp_autoclose_ticks = inp->sctp_ep.auto_close_time;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
stcb->asoc.congestion_control_module = inp->sctp_ep.sctp_default_cc_module;
stcb->asoc.cc_functions = sctp_cc_functions[inp->sctp_ep.sctp_default_cc_module];
- Modular congestion control, with RFC2581 being the default. - CMT_PF states added (w/sysctl to turn the PF version on) - sctp_input.c had a missing incr of cookie case when the auth was bad. This meant a free was called without an increment to refcnt, added increment like rest of code. - There was a case, unlikely, when the scope of the destination changed (this is a TSNH case). In that case, it would not free the alloc'ed asoc (in sctp_input.c). - When listed addresses found a colliding cookie/Init, then the collided upon tcb was not unlocked in sctp_pcb.c - Add error checking on arguments of sctp_sendx(3) to prevent it from referencing a NULL pointer. - Fix an error return of sctp_sendx(3), it was returing ENOMEM not -1. - Get assoc id was changed to use the sanctified socket api method for getting a assoc id (PEER_ADDR_INFO instead of PEER_ADDR_PARAMS). - Fix it so a peeled off socket will get a proper error return if it trys to send to a different address then it is connected to. - Fix so that select_a_stream can avoid an endless loop that could hang a caller. - time_entered (state set time) was not being set in all cases to the time we went established. Approved by: re(ken smith)
2007-07-14 09:36:28 +00:00
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
stcb->asoc.stream_scheduling_module = inp->sctp_ep.sctp_default_ss_module;
stcb->asoc.ss_functions = sctp_ss_functions[inp->sctp_ep.sctp_default_ss_module];
Add stream scheduling support. This work is based on a patch received from Robin Seggelmann. MFC after: 3 months.
2011-01-23 19:36:28 +00:00
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* Now the stream parameters, here we allocate space for all streams
* that we request by default.
*/
Add the add-stream capability. Still needs more testing.. MFC after: 1 month
2009-02-20 15:03:54 +00:00
asoc->strm_realoutsize = asoc->streamoutcnt = asoc->pre_open_streams =
Fix the allocation of outgoing streams: * When processing a cookie, use the number of streams announced in the INIT-ACK. * When sending an INIT-ACK for an existing association, use the value from the association, not from the end-point. MFC after: 1 week
2015-12-06 16:17:57 +00:00
o_strms;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_MALLOC(asoc->strmout, struct sctp_stream_out *,
asoc->streamoutcnt * sizeof(struct sctp_stream_out),
- Fixes so we won't try to start a timer when we hold a wq lock for the iterator. Panda uses a silly recursive lock they hold through the timer. - Add poor mans wireshark compile option.. - Allocate and start using SCTP_M_XXX for all SCTP_MALLOC() calls. - sysctl now will get back the refcnt for viewing by onlookers. Reviewed by: gnn
2007-05-29 09:29:03 +00:00
SCTP_M_STRMO);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (asoc->strmout == NULL) {
/* big trouble no memory */
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(NULL, stcb, NULL, SCTP_FROM_SCTPUTIL, ENOMEM);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (ENOMEM);
}
for (i = 0; i < asoc->streamoutcnt; i++) {
/*
* inbound side must be set to 0xffff, also NOTE when we get
* the INIT-ACK back (for INIT sender) we MUST reduce the
* count (streamoutcnt) but first check if we sent to any of
* the upper streams that were dropped (if some were). Those
* that were dropped must be notified to the upper layer as
* failed to send.
*/
Move from early SSN assignment to late SSN assignment. This doesn't change functionality, but makes upcoming change much easier. Developed with rrs@ at the IETF 85. MFC after: 1 week
2012-11-05 20:55:17 +00:00
asoc->strmout[i].next_sequence_send = 0x0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
TAILQ_INIT(&asoc->strmout[i].outqueue);
Get the accounting working. We now have counters how many chunks for each SCTP outgoing stream are in the send and sent queue. While there, improve the naming of NR-SACK related constants recently introduced. MFC after: 1 week
2012-11-16 19:39:10 +00:00
asoc->strmout[i].chunks_on_queues = 0;
Add support for the SCTP_PR_STREAM_STATUS and SCTP_PR_ASSOC_STATUS socket options. This includes managing the correspoing stat counters. Add the SCTP_DETAILED_STR_STATS kernel option to control per policy counters on every stream. The default is off and only an aggregated counter is available. This is sufficient for the RTCWeb usecase. MFC after: 1 week
2014-08-13 15:50:16 +00:00
#if defined(SCTP_DETAILED_STR_STATS)
for (j = 0; j < SCTP_PR_SCTP_MAX + 1; j++) {
asoc->strmout[i].abandoned_sent[j] = 0;
asoc->strmout[i].abandoned_unsent[j] = 0;
}
#else
asoc->strmout[i].abandoned_sent[0] = 0;
asoc->strmout[i].abandoned_unsent[0] = 0;
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
asoc->strmout[i].stream_no = i;
asoc->strmout[i].last_msg_incomplete = 0;
Fix several problems with Stream Reset. 1) We were not handling (or sending) the IN_PROGRESS case if the other side (or our side) was not able to reset (awaiting more data). 2) We would improperly send a stream-reset when we should not. Not waiting until the TSN had been assigned when data was inqueue. Reviewed by: tuexen
2015-07-22 11:30:37 +00:00
asoc->strmout[i].state = SCTP_STREAM_OPENING;
Fix several bugs in the stream schedulers. From Robin Seggelmann. MFC after: 3 months.
2011-02-03 20:44:49 +00:00
asoc->ss_functions.sctp_ss_init_stream(&asoc->strmout[i], NULL);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Add stream scheduling support. This work is based on a patch received from Robin Seggelmann. MFC after: 3 months.
2011-01-23 19:36:28 +00:00
asoc->ss_functions.sctp_ss_init(stcb, asoc, 0);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Now the mapping array */
asoc->mapping_array_size = SCTP_INITIAL_MAPPING_ARRAY;
SCTP_MALLOC(asoc->mapping_array, uint8_t *, asoc->mapping_array_size,
- Fixes so we won't try to start a timer when we hold a wq lock for the iterator. Panda uses a silly recursive lock they hold through the timer. - Add poor mans wireshark compile option.. - Allocate and start using SCTP_M_XXX for all SCTP_MALLOC() calls. - sysctl now will get back the refcnt for viewing by onlookers. Reviewed by: gnn
2007-05-29 09:29:03 +00:00
SCTP_M_MAP);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (asoc->mapping_array == NULL) {
- Fixes so we won't try to start a timer when we hold a wq lock for the iterator. Panda uses a silly recursive lock they hold through the timer. - Add poor mans wireshark compile option.. - Allocate and start using SCTP_M_XXX for all SCTP_MALLOC() calls. - sysctl now will get back the refcnt for viewing by onlookers. Reviewed by: gnn
2007-05-29 09:29:03 +00:00
SCTP_FREE(asoc->strmout, SCTP_M_STRMO);
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(NULL, stcb, NULL, SCTP_FROM_SCTPUTIL, ENOMEM);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (ENOMEM);
}
memset(asoc->mapping_array, 0, asoc->mapping_array_size);
* Fix some race condition in SACK/NR-SACK processing. * Fix handling of mapping arrays when draining mbufs or processing FORWARD-TSN chunks. * Cleanup code (no duplicate code anymore for SACKs and NR-SACKs). Part of this code was developed together with rrs. MFC after: 2 weeks.
2010-04-03 15:40:14 +00:00
SCTP_MALLOC(asoc->nr_mapping_array, uint8_t *, asoc->mapping_array_size,
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
SCTP_M_MAP);
Fix a small memory leak from the nr-sack code - the mapping array was not being freed at term of association. Also get rid of the MICHAELS_EXP code.
2009-05-30 10:56:27 +00:00
if (asoc->nr_mapping_array == NULL) {
SCTP_FREE(asoc->strmout, SCTP_M_STRMO);
SCTP_FREE(asoc->mapping_array, SCTP_M_MAP);
SCTP_LTRACE_ERR_RET(NULL, stcb, NULL, SCTP_FROM_SCTPUTIL, ENOMEM);
return (ENOMEM);
}
* Fix some race condition in SACK/NR-SACK processing. * Fix handling of mapping arrays when draining mbufs or processing FORWARD-TSN chunks. * Cleanup code (no duplicate code anymore for SACKs and NR-SACKs). Part of this code was developed together with rrs. MFC after: 2 weeks.
2010-04-03 15:40:14 +00:00
memset(asoc->nr_mapping_array, 0, asoc->mapping_array_size);
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Now the init of the other outqueues */
TAILQ_INIT(&asoc->free_chunks);
TAILQ_INIT(&asoc->control_send_queue);
- Adds support for the multi-asconf (From Kozuka-san) - Adds some prepwork (Not all yet) for vimage in particular support the delete the sctppcbinfo.xx structs. There is still a leak in here if it were to be called plus we stil need the regrouping (From Me and Michael Tuexen) - Adds support for UDP tunneling. For BSD there is no socket yet setup so its disabled, but major argument changes are in here to emcompass the passing of the port number (zero when you don't have a udp tunnel, the default for BSD). Will add some hooks in UDP here shortly (discussed with Robert) that will allow easy tunneling. (Mainly from Peter Lei and Michael Tuexen with some BSD work from me :-D) - Some ease for windows, evidently leave is reserved by their compile move label leave: -> out: MFC after: 1 week
2008-05-20 13:47:46 +00:00
TAILQ_INIT(&asoc->asconf_send_queue);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
TAILQ_INIT(&asoc->send_queue);
TAILQ_INIT(&asoc->sent_queue);
TAILQ_INIT(&asoc->resetHead);
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->max_inbound_streams = inp->sctp_ep.max_open_streams_intome;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
TAILQ_INIT(&asoc->asconf_queue);
/* authentication fields */
asoc->authinfo.random = NULL;
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
asoc->authinfo.active_keyid = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
asoc->authinfo.assoc_key = NULL;
asoc->authinfo.assoc_keyid = 0;
asoc->authinfo.recv_key = NULL;
asoc->authinfo.recv_keyid = 0;
LIST_INIT(&asoc->shared_keys);
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
asoc->marked_retrans = 0;
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
asoc->port = inp->sctp_ep.port;
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
asoc->timoinit = 0;
asoc->timodata = 0;
asoc->timosack = 0;
asoc->timoshutdown = 0;
asoc->timoheartbeat = 0;
asoc->timocookie = 0;
asoc->timoshutdownack = 0;
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
(void)SCTP_GETTIME_TIMEVAL(&asoc->start_time);
asoc->discontinuity_time = asoc->start_time;
Add support for the SCTP_PR_STREAM_STATUS and SCTP_PR_ASSOC_STATUS socket options. This includes managing the correspoing stat counters. Add the SCTP_DETAILED_STR_STATS kernel option to control per policy counters on every stream. The default is off and only an aggregated counter is available. This is sufficient for the RTCWeb usecase. MFC after: 1 week
2014-08-13 15:50:16 +00:00
for (i = 0; i < SCTP_PR_SCTP_MAX + 1; i++) {
asoc->abandoned_unsent[i] = 0;
asoc->abandoned_sent[i] = 0;
}
- Fixes cstatic issues found by cisco sa tool (missing frees and such on error legs) - align sctp_sockstore to 64 bit boundary ..
2007-06-18 21:59:15 +00:00
/*
* sa_ignore MEMLEAK {memory is put in the assoc mapping array and
Fix for NR-Sack code. The code was NOT working properly when enabled. Basically most of the operations were incorrect causing bad sacks when you enabled nr-sack. The fixes range across 4 files and unifiy most of the processing so that we only test nr_sack flags to decide which type of sack to generate. Optimization left for this is to combine the sack generation code and make it capable of generating either sack thus shrinking out a routine. Reviewed by: tuexen@freebsd.org
2010-03-24 19:45:36 +00:00
* freed later when the association is freed.
- Fixes cstatic issues found by cisco sa tool (missing frees and such on error legs) - align sctp_sockstore to 64 bit boundary ..
2007-06-18 21:59:15 +00:00
*/
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (0);
}
Fixes a bug where SACKs in the face of mapping_array expansion would break. Basically once we expanded the array we no longer had both mapping arrays in sync which the sack processing code depends on. This would mean we were randomly referring to memory that was probably not there. This mostly just gave us bad sack results going back to the peer. If INVARIENTS was on of course we would hit the panic routine in the sack_check call. We also add a print routine for the place where one would panic in invarients so one can see what the main mapping array holds. Reviewed by: tuexen@freebsd.org MFC after: 2 weeks
2010-03-23 01:36:50 +00:00
void
sctp_print_mapping_array(struct sctp_association *asoc)
{
Fix a off-by-one bug in zeroing out the mapping arrays. Fix sctp_print_mapping_array(). MFC after: 1 week
2010-04-06 18:57:50 +00:00
unsigned int i, limit;
Fixes a bug where SACKs in the face of mapping_array expansion would break. Basically once we expanded the array we no longer had both mapping arrays in sync which the sack processing code depends on. This would mean we were randomly referring to memory that was probably not there. This mostly just gave us bad sack results going back to the peer. If INVARIENTS was on of course we would hit the panic routine in the sack_check call. We also add a print routine for the place where one would panic in invarients so one can see what the main mapping array holds. Reviewed by: tuexen@freebsd.org MFC after: 2 weeks
2010-03-23 01:36:50 +00:00
Use SCTP_PRINTF() instead of printf() in all SCTP sources. MFC after: 3 days
2012-05-04 09:27:00 +00:00
SCTP_PRINTF("Mapping array size: %d, baseTSN: %8.8x, cumAck: %8.8x, highestTSN: (%8.8x, %8.8x).\n",
Fixes a bug where SACKs in the face of mapping_array expansion would break. Basically once we expanded the array we no longer had both mapping arrays in sync which the sack processing code depends on. This would mean we were randomly referring to memory that was probably not there. This mostly just gave us bad sack results going back to the peer. If INVARIENTS was on of course we would hit the panic routine in the sack_check call. We also add a print routine for the place where one would panic in invarients so one can see what the main mapping array holds. Reviewed by: tuexen@freebsd.org MFC after: 2 weeks
2010-03-23 01:36:50 +00:00
asoc->mapping_array_size,
asoc->mapping_array_base_tsn,
asoc->cumulative_tsn,
Fix a off-by-one bug in zeroing out the mapping arrays. Fix sctp_print_mapping_array(). MFC after: 1 week
2010-04-06 18:57:50 +00:00
asoc->highest_tsn_inside_map,
asoc->highest_tsn_inside_nr_map);
for (limit = asoc->mapping_array_size; limit > 1; limit--) {
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
if (asoc->mapping_array[limit - 1] != 0) {
Fix for NR-Sack code. The code was NOT working properly when enabled. Basically most of the operations were incorrect causing bad sacks when you enabled nr-sack. The fixes range across 4 files and unifiy most of the processing so that we only test nr_sack flags to decide which type of sack to generate. Optimization left for this is to combine the sack generation code and make it capable of generating either sack thus shrinking out a routine. Reviewed by: tuexen@freebsd.org
2010-03-24 19:45:36 +00:00
break;
}
}
Use SCTP_PRINTF() instead of printf() in all SCTP sources. MFC after: 3 days
2012-05-04 09:27:00 +00:00
SCTP_PRINTF("Renegable mapping array (last %d entries are zero):\n", asoc->mapping_array_size - limit);
Fix for NR-Sack code. The code was NOT working properly when enabled. Basically most of the operations were incorrect causing bad sacks when you enabled nr-sack. The fixes range across 4 files and unifiy most of the processing so that we only test nr_sack flags to decide which type of sack to generate. Optimization left for this is to combine the sack generation code and make it capable of generating either sack thus shrinking out a routine. Reviewed by: tuexen@freebsd.org
2010-03-24 19:45:36 +00:00
for (i = 0; i < limit; i++) {
Use SCTP_PRINTF() instead of printf() in all SCTP sources. MFC after: 3 days
2012-05-04 09:27:00 +00:00
SCTP_PRINTF("%2.2x%c", asoc->mapping_array[i], ((i + 1) % 16) ? ' ' : '\n');
Fix for NR-Sack code. The code was NOT working properly when enabled. Basically most of the operations were incorrect causing bad sacks when you enabled nr-sack. The fixes range across 4 files and unifiy most of the processing so that we only test nr_sack flags to decide which type of sack to generate. Optimization left for this is to combine the sack generation code and make it capable of generating either sack thus shrinking out a routine. Reviewed by: tuexen@freebsd.org
2010-03-24 19:45:36 +00:00
}
Fix a off-by-one bug in zeroing out the mapping arrays. Fix sctp_print_mapping_array(). MFC after: 1 week
2010-04-06 18:57:50 +00:00
if (limit % 16)
Use SCTP_PRINTF() instead of printf() in all SCTP sources. MFC after: 3 days
2012-05-04 09:27:00 +00:00
SCTP_PRINTF("\n");
Fix a off-by-one bug in zeroing out the mapping arrays. Fix sctp_print_mapping_array(). MFC after: 1 week
2010-04-06 18:57:50 +00:00
for (limit = asoc->mapping_array_size; limit > 1; limit--) {
if (asoc->nr_mapping_array[limit - 1]) {
Fix for NR-Sack code. The code was NOT working properly when enabled. Basically most of the operations were incorrect causing bad sacks when you enabled nr-sack. The fixes range across 4 files and unifiy most of the processing so that we only test nr_sack flags to decide which type of sack to generate. Optimization left for this is to combine the sack generation code and make it capable of generating either sack thus shrinking out a routine. Reviewed by: tuexen@freebsd.org
2010-03-24 19:45:36 +00:00
break;
}
}
Use SCTP_PRINTF() instead of printf() in all SCTP sources. MFC after: 3 days
2012-05-04 09:27:00 +00:00
SCTP_PRINTF("Non renegable mapping array (last %d entries are zero):\n", asoc->mapping_array_size - limit);
Fix for NR-Sack code. The code was NOT working properly when enabled. Basically most of the operations were incorrect causing bad sacks when you enabled nr-sack. The fixes range across 4 files and unifiy most of the processing so that we only test nr_sack flags to decide which type of sack to generate. Optimization left for this is to combine the sack generation code and make it capable of generating either sack thus shrinking out a routine. Reviewed by: tuexen@freebsd.org
2010-03-24 19:45:36 +00:00
for (i = 0; i < limit; i++) {
Use SCTP_PRINTF() instead of printf() in all SCTP sources. MFC after: 3 days
2012-05-04 09:27:00 +00:00
SCTP_PRINTF("%2.2x%c", asoc->nr_mapping_array[i], ((i + 1) % 16) ? ' ' : '\n');
Fixes a bug where SACKs in the face of mapping_array expansion would break. Basically once we expanded the array we no longer had both mapping arrays in sync which the sack processing code depends on. This would mean we were randomly referring to memory that was probably not there. This mostly just gave us bad sack results going back to the peer. If INVARIENTS was on of course we would hit the panic routine in the sack_check call. We also add a print routine for the place where one would panic in invarients so one can see what the main mapping array holds. Reviewed by: tuexen@freebsd.org MFC after: 2 weeks
2010-03-23 01:36:50 +00:00
}
Fix a off-by-one bug in zeroing out the mapping arrays. Fix sctp_print_mapping_array(). MFC after: 1 week
2010-04-06 18:57:50 +00:00
if (limit % 16)
Use SCTP_PRINTF() instead of printf() in all SCTP sources. MFC after: 3 days
2012-05-04 09:27:00 +00:00
SCTP_PRINTF("\n");
Fixes a bug where SACKs in the face of mapping_array expansion would break. Basically once we expanded the array we no longer had both mapping arrays in sync which the sack processing code depends on. This would mean we were randomly referring to memory that was probably not there. This mostly just gave us bad sack results going back to the peer. If INVARIENTS was on of course we would hit the panic routine in the sack_check call. We also add a print routine for the place where one would panic in invarients so one can see what the main mapping array holds. Reviewed by: tuexen@freebsd.org MFC after: 2 weeks
2010-03-23 01:36:50 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
int
- Fix a memory overwrite when the mapping array is expanded, size of expansion was not taken int consideration. - Fix so vtag hash is 1 bigger so that it modulo's out correctly, avoids a panic when restart with right modulo happens. - do not dereference stcb when control->do_not_ref_stcb is set - Fix up packet logging to not often use a lock and also to add to options. - Fix some logging option duplication in the sctputil.h
2007-05-30 17:39:45 +00:00
sctp_expand_mapping_array(struct sctp_association *asoc, uint32_t needed)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
/* mapping array needs to grow */
* Fix some race condition in SACK/NR-SACK processing. * Fix handling of mapping arrays when draining mbufs or processing FORWARD-TSN chunks. * Cleanup code (no duplicate code anymore for SACKs and NR-SACKs). Part of this code was developed together with rrs. MFC after: 2 weeks.
2010-04-03 15:40:14 +00:00
uint8_t *new_array1, *new_array2;
- Fix a memory overwrite when the mapping array is expanded, size of expansion was not taken int consideration. - Fix so vtag hash is 1 bigger so that it modulo's out correctly, avoids a panic when restart with right modulo happens. - do not dereference stcb when control->do_not_ref_stcb is set - Fix up packet logging to not often use a lock and also to add to options. - Fix some logging option duplication in the sctputil.h
2007-05-30 17:39:45 +00:00
uint32_t new_size;
new_size = asoc->mapping_array_size + ((needed + 7) / 8 + SCTP_MAPPING_ARRAY_INCR);
* Fix some race condition in SACK/NR-SACK processing. * Fix handling of mapping arrays when draining mbufs or processing FORWARD-TSN chunks. * Cleanup code (no duplicate code anymore for SACKs and NR-SACKs). Part of this code was developed together with rrs. MFC after: 2 weeks.
2010-04-03 15:40:14 +00:00
SCTP_MALLOC(new_array1, uint8_t *, new_size, SCTP_M_MAP);
SCTP_MALLOC(new_array2, uint8_t *, new_size, SCTP_M_MAP);
if ((new_array1 == NULL) || (new_array2 == NULL)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* can't get more, forget it */
* Fix some race condition in SACK/NR-SACK processing. * Fix handling of mapping arrays when draining mbufs or processing FORWARD-TSN chunks. * Cleanup code (no duplicate code anymore for SACKs and NR-SACKs). Part of this code was developed together with rrs. MFC after: 2 weeks.
2010-04-03 15:40:14 +00:00
SCTP_PRINTF("No memory for expansion of SCTP mapping array %d\n", new_size);
if (new_array1) {
SCTP_FREE(new_array1, SCTP_M_MAP);
}
if (new_array2) {
SCTP_FREE(new_array2, SCTP_M_MAP);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (-1);
}
* Fix some race condition in SACK/NR-SACK processing. * Fix handling of mapping arrays when draining mbufs or processing FORWARD-TSN chunks. * Cleanup code (no duplicate code anymore for SACKs and NR-SACKs). Part of this code was developed together with rrs. MFC after: 2 weeks.
2010-04-03 15:40:14 +00:00
memset(new_array1, 0, new_size);
memset(new_array2, 0, new_size);
memcpy(new_array1, asoc->mapping_array, asoc->mapping_array_size);
memcpy(new_array2, asoc->nr_mapping_array, asoc->mapping_array_size);
- Fixes so we won't try to start a timer when we hold a wq lock for the iterator. Panda uses a silly recursive lock they hold through the timer. - Add poor mans wireshark compile option.. - Allocate and start using SCTP_M_XXX for all SCTP_MALLOC() calls. - sysctl now will get back the refcnt for viewing by onlookers. Reviewed by: gnn
2007-05-29 09:29:03 +00:00
SCTP_FREE(asoc->mapping_array, SCTP_M_MAP);
Fixes a bug where SACKs in the face of mapping_array expansion would break. Basically once we expanded the array we no longer had both mapping arrays in sync which the sack processing code depends on. This would mean we were randomly referring to memory that was probably not there. This mostly just gave us bad sack results going back to the peer. If INVARIENTS was on of course we would hit the panic routine in the sack_check call. We also add a print routine for the place where one would panic in invarients so one can see what the main mapping array holds. Reviewed by: tuexen@freebsd.org MFC after: 2 weeks
2010-03-23 01:36:50 +00:00
SCTP_FREE(asoc->nr_mapping_array, SCTP_M_MAP);
* Fix some race condition in SACK/NR-SACK processing. * Fix handling of mapping arrays when draining mbufs or processing FORWARD-TSN chunks. * Cleanup code (no duplicate code anymore for SACKs and NR-SACKs). Part of this code was developed together with rrs. MFC after: 2 weeks.
2010-04-03 15:40:14 +00:00
asoc->mapping_array = new_array1;
asoc->nr_mapping_array = new_array2;
asoc->mapping_array_size = new_size;
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
return (0);
}
Many bug fixes (from the IETF hack-fest): - PR-SCTP had major issues when skipping through a multi-part message. o Did not look at socket buffer. o Did not properly handle the reassmebly queue. o The MARKED segments could interfere and un-skip a chunk causing a problem with the proper FWD-TSN. o No FR of FWD-TSN's was being done. - NR-Sack code was basically disabled. It needed fixes that never got into the real code. - CMT code had issues when the two paths were NOT the same b/w. We found a few small bugs, but also the critcal one here was not dividing the rwnd amongst the paths. Obtained from: Michael Tuexen and myself at the IETF hack-fest ;-)
2009-04-04 11:43:32 +00:00
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
static void
sctp_iterator_work(struct sctp_iterator *it)
{
int iteration_count = 0;
int inp_skip = 0;
3 Fixes - a) There was a case where a ICMP message could cause us to return leaving a stuck lock on an stcb. b) The iterator needed some tweaks to fix its lock ordering. c) The ITERATOR_LOCK is no longer needed in the freeing of a stcb. Now that the timer based one is gone we don't have a multiple resume situation. Add to that that there was somewhere a path out of the freeing of an assoc that did NOT release the iterator_lock.. it was time to clean this old code up and in the process fix the lock bug. MFC after: 1 week
2010-06-11 03:54:00 +00:00
int first_in = 1;
struct sctp_inpcb *tinp;
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
3 Fixes - a) There was a case where a ICMP message could cause us to return leaving a stuck lock on an stcb. b) The iterator needed some tweaks to fix its lock ordering. c) The ITERATOR_LOCK is no longer needed in the freeing of a stcb. Now that the timer based one is gone we don't have a multiple resume situation. Add to that that there was somewhere a path out of the freeing of an assoc that did NOT release the iterator_lock.. it was time to clean this old code up and in the process fix the lock bug. MFC after: 1 week
2010-06-11 03:54:00 +00:00
SCTP_INP_INFO_RLOCK();
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_ITERATOR_LOCK();
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (it->inp) {
3 Fixes - a) There was a case where a ICMP message could cause us to return leaving a stuck lock on an stcb. b) The iterator needed some tweaks to fix its lock ordering. c) The ITERATOR_LOCK is no longer needed in the freeing of a stcb. Now that the timer based one is gone we don't have a multiple resume situation. Add to that that there was somewhere a path out of the freeing of an assoc that did NOT release the iterator_lock.. it was time to clean this old code up and in the process fix the lock bug. MFC after: 1 week
2010-06-11 03:54:00 +00:00
SCTP_INP_RLOCK(it->inp);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_INP_DECR_REF(it->inp);
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
if (it->inp == NULL) {
/* iterator is complete */
done_with_iterator:
SCTP_ITERATOR_UNLOCK();
3 Fixes - a) There was a case where a ICMP message could cause us to return leaving a stuck lock on an stcb. b) The iterator needed some tweaks to fix its lock ordering. c) The ITERATOR_LOCK is no longer needed in the freeing of a stcb. Now that the timer based one is gone we don't have a multiple resume situation. Add to that that there was somewhere a path out of the freeing of an assoc that did NOT release the iterator_lock.. it was time to clean this old code up and in the process fix the lock bug. MFC after: 1 week
2010-06-11 03:54:00 +00:00
SCTP_INP_INFO_RUNLOCK();
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
if (it->function_atend != NULL) {
(*it->function_atend) (it->pointer, it->val);
}
- Fixes so we won't try to start a timer when we hold a wq lock for the iterator. Panda uses a silly recursive lock they hold through the timer. - Add poor mans wireshark compile option.. - Allocate and start using SCTP_M_XXX for all SCTP_MALLOC() calls. - sysctl now will get back the refcnt for viewing by onlookers. Reviewed by: gnn
2007-05-29 09:29:03 +00:00
SCTP_FREE(it, SCTP_M_ITER);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
return;
}
select_a_new_ep:
3 Fixes - a) There was a case where a ICMP message could cause us to return leaving a stuck lock on an stcb. b) The iterator needed some tweaks to fix its lock ordering. c) The ITERATOR_LOCK is no longer needed in the freeing of a stcb. Now that the timer based one is gone we don't have a multiple resume situation. Add to that that there was somewhere a path out of the freeing of an assoc that did NOT release the iterator_lock.. it was time to clean this old code up and in the process fix the lock bug. MFC after: 1 week
2010-06-11 03:54:00 +00:00
if (first_in) {
first_in = 0;
} else {
SCTP_INP_RLOCK(it->inp);
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
while (((it->pcb_flags) &&
((it->inp->sctp_flags & it->pcb_flags) != it->pcb_flags)) ||
((it->pcb_features) &&
((it->inp->sctp_features & it->pcb_features) != it->pcb_features))) {
/* endpoint flags or features don't match, so keep looking */
if (it->iterator_flags & SCTP_ITERATOR_DO_SINGLE_INP) {
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
SCTP_INP_RUNLOCK(it->inp);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
goto done_with_iterator;
}
3 Fixes - a) There was a case where a ICMP message could cause us to return leaving a stuck lock on an stcb. b) The iterator needed some tweaks to fix its lock ordering. c) The ITERATOR_LOCK is no longer needed in the freeing of a stcb. Now that the timer based one is gone we don't have a multiple resume situation. Add to that that there was somewhere a path out of the freeing of an assoc that did NOT release the iterator_lock.. it was time to clean this old code up and in the process fix the lock bug. MFC after: 1 week
2010-06-11 03:54:00 +00:00
tinp = it->inp;
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
it->inp = LIST_NEXT(it->inp, sctp_list);
3 Fixes - a) There was a case where a ICMP message could cause us to return leaving a stuck lock on an stcb. b) The iterator needed some tweaks to fix its lock ordering. c) The ITERATOR_LOCK is no longer needed in the freeing of a stcb. Now that the timer based one is gone we don't have a multiple resume situation. Add to that that there was somewhere a path out of the freeing of an assoc that did NOT release the iterator_lock.. it was time to clean this old code up and in the process fix the lock bug. MFC after: 1 week
2010-06-11 03:54:00 +00:00
SCTP_INP_RUNLOCK(tinp);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
if (it->inp == NULL) {
goto done_with_iterator;
}
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
SCTP_INP_RLOCK(it->inp);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
}
/* now go through each assoc which is in the desired state */
if (it->done_current_ep == 0) {
if (it->function_inp != NULL)
inp_skip = (*it->function_inp) (it->inp, it->pointer, it->val);
it->done_current_ep = 1;
}
if (it->stcb == NULL) {
/* run the per instance function */
it->stcb = LIST_FIRST(&it->inp->sctp_asoc_list);
}
if ((inp_skip) || it->stcb == NULL) {
if (it->function_inp_end != NULL) {
inp_skip = (*it->function_inp_end) (it->inp,
it->pointer,
it->val);
}
SCTP_INP_RUNLOCK(it->inp);
goto no_stcb;
}
while (it->stcb) {
SCTP_TCB_LOCK(it->stcb);
if (it->asoc_state && ((it->stcb->asoc.state & it->asoc_state) != it->asoc_state)) {
/* not in the right state... keep looking */
SCTP_TCB_UNLOCK(it->stcb);
goto next_assoc;
}
/* see if we have limited out the iterator loop */
iteration_count++;
if (iteration_count > SCTP_ITERATOR_MAX_AT_ONCE) {
/* Pause to let others grab the lock */
atomic_add_int(&it->stcb->asoc.refcnt, 1);
SCTP_TCB_UNLOCK(it->stcb);
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_INP_INCR_REF(it->inp);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_INP_RUNLOCK(it->inp);
SCTP_ITERATOR_UNLOCK();
3 Fixes - a) There was a case where a ICMP message could cause us to return leaving a stuck lock on an stcb. b) The iterator needed some tweaks to fix its lock ordering. c) The ITERATOR_LOCK is no longer needed in the freeing of a stcb. Now that the timer based one is gone we don't have a multiple resume situation. Add to that that there was somewhere a path out of the freeing of an assoc that did NOT release the iterator_lock.. it was time to clean this old code up and in the process fix the lock bug. MFC after: 1 week
2010-06-11 03:54:00 +00:00
SCTP_INP_INFO_RUNLOCK();
SCTP_INP_INFO_RLOCK();
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_ITERATOR_LOCK();
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
if (sctp_it_ctl.iterator_flags) {
/* We won't be staying here */
SCTP_INP_DECR_REF(it->inp);
atomic_add_int(&it->stcb->asoc.refcnt, -1);
if (sctp_it_ctl.iterator_flags &
SCTP_ITERATOR_STOP_CUR_IT) {
sctp_it_ctl.iterator_flags &= ~SCTP_ITERATOR_STOP_CUR_IT;
goto done_with_iterator;
}
if (sctp_it_ctl.iterator_flags &
SCTP_ITERATOR_STOP_CUR_INP) {
sctp_it_ctl.iterator_flags &= ~SCTP_ITERATOR_STOP_CUR_INP;
goto no_stcb;
}
/* If we reach here huh? */
Use SCTP_PRINTF() instead of printf() in all SCTP sources. MFC after: 3 days
2012-05-04 09:27:00 +00:00
SCTP_PRINTF("Unknown it ctl flag %x\n",
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
sctp_it_ctl.iterator_flags);
sctp_it_ctl.iterator_flags = 0;
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_INP_RLOCK(it->inp);
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_INP_DECR_REF(it->inp);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_TCB_LOCK(it->stcb);
atomic_add_int(&it->stcb->asoc.refcnt, -1);
iteration_count = 0;
}
/* run function on this one */
(*it->function_assoc) (it->inp, it->stcb, it->pointer, it->val);
/*
* we lie here, it really needs to have its own type but
* first I must verify that this won't effect things :-0
*/
if (it->no_chunk_output == 0)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
sctp_chunk_output(it->inp, it->stcb, SCTP_OUTPUT_FROM_T3, SCTP_SO_NOT_LOCKED);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_TCB_UNLOCK(it->stcb);
next_assoc:
it->stcb = LIST_NEXT(it->stcb, sctp_tcblist);
if (it->stcb == NULL) {
/* Run last function */
if (it->function_inp_end != NULL) {
inp_skip = (*it->function_inp_end) (it->inp,
it->pointer,
it->val);
}
}
}
SCTP_INP_RUNLOCK(it->inp);
no_stcb:
/* done with all assocs on this endpoint, move on to next endpoint */
it->done_current_ep = 0;
if (it->iterator_flags & SCTP_ITERATOR_DO_SINGLE_INP) {
it->inp = NULL;
} else {
it->inp = LIST_NEXT(it->inp, sctp_list);
}
if (it->inp == NULL) {
goto done_with_iterator;
}
goto select_a_new_ep;
}
void
sctp_iterator_worker(void)
{
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
struct sctp_iterator *it, *nit;
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
/* This function is called with the WQ lock in place */
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
sctp_it_ctl.iterator_running = 1;
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
TAILQ_FOREACH_SAFE(it, &sctp_it_ctl.iteratorhead, sctp_nxt_itr, nit) {
sctp_it_ctl.cur_it = it;
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
/* now lets work on this one */
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
TAILQ_REMOVE(&sctp_it_ctl.iteratorhead, it, sctp_nxt_itr);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_IPI_ITERATOR_WQ_UNLOCK();
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
CURVNET_SET(it->vn);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
sctp_iterator_work(it);
Tunes and fixes the new DC-CC to seem to hit the right mix. Still may need some tweaks but it appears to almost not give away too much to an RFC2581 flow, but can really minimize the amount of buffers used in the net. MFC after: 3 months
2011-03-08 11:58:25 +00:00
sctp_it_ctl.cur_it = NULL;
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
CURVNET_RESTORE();
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_IPI_ITERATOR_WQ_LOCK();
- Fixed 1-2-1 model to not worry about associd in sockopts - Fixed RTOinfo for bounding. - Fixed connect() to return ECONNREFUSED when an ABORT is received. - Added comments to direct Static Analysis not to look at some things it does not understand (comments are /* sa_ignore XXXXX */) - Bind when colliding was broken, missing not_found = 1 before checking to see if the port was in use caused endless bind loop. - Cookie life needs to be in milliseconds to conform to socket api. - Cookie life is not supposed to change if its 0, On the assoc level set we changed it to 0 opps. - Two more static analysis issues identified by the cisco tool. Null checks needed. - An issue for sendfile(). Need to validate the correct input argument. - When sending failed due to a no route to host, we leaked the mbuf chain failing to call m_freem(). - Fix #ifdef issue for getting hash block len when HAVE_SHA2 is NOT defined Reviewed by: gnn
2007-05-17 12:16:24 +00:00
/* sa_ignore FREED_MEMORY */
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
}
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
sctp_it_ctl.iterator_running = 0;
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
return;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
static void
sctp_handle_addr_wq(void)
{
/* deal with the ADDR wq from the rtsock calls */
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
struct sctp_laddr *wi, *nwi;
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
struct sctp_asconf_iterator *asc;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_MALLOC(asc, struct sctp_asconf_iterator *,
- Fixes so we won't try to start a timer when we hold a wq lock for the iterator. Panda uses a silly recursive lock they hold through the timer. - Add poor mans wireshark compile option.. - Allocate and start using SCTP_M_XXX for all SCTP_MALLOC() calls. - sysctl now will get back the refcnt for viewing by onlookers. Reviewed by: gnn
2007-05-29 09:29:03 +00:00
sizeof(struct sctp_asconf_iterator), SCTP_M_ASC_IT);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
if (asc == NULL) {
/* Try later, no memory */
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_timer_start(SCTP_TIMER_TYPE_ADDR_WQ,
(struct sctp_inpcb *)NULL,
(struct sctp_tcb *)NULL,
(struct sctp_nets *)NULL);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
LIST_INIT(&asc->list_of_work);
asc->cnt = 0;
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
SCTP_WQ_ADDR_LOCK();
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
LIST_FOREACH_SAFE(wi, &SCTP_BASE_INFO(addr_wq), sctp_nxt_addr, nwi) {
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
LIST_REMOVE(wi, sctp_nxt_addr);
LIST_INSERT_HEAD(&asc->list_of_work, wi, sctp_nxt_addr);
asc->cnt++;
}
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
SCTP_WQ_ADDR_UNLOCK();
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
if (asc->cnt == 0) {
- Fixes so we won't try to start a timer when we hold a wq lock for the iterator. Panda uses a silly recursive lock they hold through the timer. - Add poor mans wireshark compile option.. - Allocate and start using SCTP_M_XXX for all SCTP_MALLOC() calls. - sysctl now will get back the refcnt for viewing by onlookers. Reviewed by: gnn
2007-05-29 09:29:03 +00:00
SCTP_FREE(asc, SCTP_M_ASC_IT);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
} else {
Improve the teardown of the SCTP stack. Obtained from: bz@ MFC after: 1 week
2016-02-16 19:36:25 +00:00
int ret;
ret = sctp_initiate_iterator(sctp_asconf_iterator_ep,
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
sctp_asconf_iterator_stcb,
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
NULL, /* No ep end for boundall */
SCTP_PCB_FLAGS_BOUNDALL,
SCTP_PCB_ANY_FEATURES,
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
SCTP_ASOC_ANY_STATE,
(void *)asc, 0,
sctp_asconf_iterator_end, NULL, 0);
Improve the teardown of the SCTP stack. Obtained from: bz@ MFC after: 1 week
2016-02-16 19:36:25 +00:00
if (ret) {
SCTP_PRINTF("Failed to initiate iterator for handle_addr_wq\n");
Whitespace changes.
2016-02-16 20:33:18 +00:00
/*
* Freeing if we are stopping or put back on the
* addr_wq.
*/
Improve the teardown of the SCTP stack. Obtained from: bz@ MFC after: 1 week
2016-02-16 19:36:25 +00:00
if (SCTP_BASE_VAR(sctp_pcb_initialized) == 0) {
sctp_asconf_iterator_end(asc, 0);
} else {
SCTP_WQ_ADDR_LOCK();
LIST_FOREACH(wi, &asc->list_of_work, sctp_nxt_addr) {
LIST_INSERT_HEAD(&SCTP_BASE_INFO(addr_wq), wi, sctp_nxt_addr);
}
SCTP_WQ_ADDR_UNLOCK();
SCTP_FREE(asc, SCTP_M_ASC_IT);
}
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
void
sctp_timeout_handler(void *t)
{
struct sctp_inpcb *inp;
struct sctp_tcb *stcb;
struct sctp_nets *net;
struct sctp_timer *tmr;
Provide consistent error causes whenever an ABORT chunk is sent. MFC after: 1 week
2015-07-27 22:35:54 +00:00
struct mbuf *op_err;
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
struct socket *so;
#endif
Address some compiler warnings. No functional change. MFC after: 3 days
2015-05-28 14:24:21 +00:00
int did_output;
Store the timer type for logging, because the timer can be freed during processing the timerout. MFC after: 3 days
2016-01-13 14:28:12 +00:00
int type;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
tmr = (struct sctp_timer *)t;
inp = (struct sctp_inpcb *)tmr->ep;
stcb = (struct sctp_tcb *)tmr->tcb;
net = (struct sctp_nets *)tmr->net;
Get SCTP working in combination with VIMAGE. Contains code from bz. Approved by: rrs (mentor) MFC after: 1 month.
2009-09-19 14:02:16 +00:00
CURVNET_SET((struct vnet *)tmr->vnet);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
did_output = 1;
#ifdef SCTP_AUDITING_ENABLED
sctp_audit_log(0xF0, (uint8_t) tmr->type);
sctp_auditing(3, inp, stcb, net);
#endif
/* sanity checks... */
if (tmr->self != (void *)tmr) {
/*
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
* SCTP_PRINTF("Stale SCTP timer fired (%p), ignoring...\n",
Using %p in a format string requires a void *. MFC after: 10 days
2012-09-05 18:52:01 +00:00
* (void *)tmr);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
*/
Get SCTP working in combination with VIMAGE. Contains code from bz. Approved by: rrs (mentor) MFC after: 1 month.
2009-09-19 14:02:16 +00:00
CURVNET_RESTORE();
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
}
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
tmr->stopped_from = 0xa001;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (!SCTP_IS_TIMER_TYPE_VALID(tmr->type)) {
/*
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
* SCTP_PRINTF("SCTP timer fired with invalid type: 0x%x\n",
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
* tmr->type);
*/
Get SCTP working in combination with VIMAGE. Contains code from bz. Approved by: rrs (mentor) MFC after: 1 month.
2009-09-19 14:02:16 +00:00
CURVNET_RESTORE();
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
}
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
tmr->stopped_from = 0xa002;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if ((tmr->type != SCTP_TIMER_TYPE_ADDR_WQ) && (inp == NULL)) {
Get SCTP working in combination with VIMAGE. Contains code from bz. Approved by: rrs (mentor) MFC after: 1 month.
2009-09-19 14:02:16 +00:00
CURVNET_RESTORE();
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
}
/* if this is an iterator timeout, get the struct and clear inp */
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
tmr->stopped_from = 0xa003;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (inp) {
SCTP_INP_INCR_REF(inp);
Use NULL instead of 0. MFC after: 1 month.
2012-01-06 20:20:59 +00:00
if ((inp->sctp_socket == NULL) &&
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
((tmr->type != SCTP_TIMER_TYPE_INPKILL) &&
* Fix a bug where PR-SCTP settings are ignore when using implicit association setup. * Fix a bug where message with illegal stream ids are not deleted. * Fix a crash when reporting back unsent messages from the send_queue. * Fix a bug related to INIT retransmission when the socket is already closed. * Fix a bug where associations were stalled when partial delivery API was enabled. * Fix a bug where the receive buffer size was smaller than the partial_delivery_point. Approved by: re, rrs (mentor) MFC after: One day.
2009-08-15 21:10:52 +00:00
(tmr->type != SCTP_TIMER_TYPE_INIT) &&
-Improvement: Add '\n' on debug output in sctp_lower_sosend(). -Improvement: panic() on INVARIANTS kernels if memory allocation fails for a tagblock in sctp_add_vtag_to_timewait(). -Bugfix: Protect code in sctp_is_in_timewait() by SCTP_INP_INFO_WLOCK/SCTP_INP_INFO_WUNLOCK. -Cleanup: Get rid of unused variable now in sctp_init_asoc(). -Bugfix: Reuse the correct vtag in sctp_add_vtag_to_timewait(). -Cleanup: Get rid of unused constant SCTP_TIME_WAIT_SHORT in sctp_constants.h. -Improvement: Use all hash buckets of the vtag hash table. -Cleanup: Get rid of then unused constant SCTP_STACK_VTAG_HASH_SIZE_A. -Bugfix: Handle SHUTDOWN;SACK packet correctly. -Bugfix: Last TSN in a gap ack block was not being "ack'd" in the internal scoreboard. Obtained from: (with help from Michael Tuexen)
2008-11-12 14:16:39 +00:00
(tmr->type != SCTP_TIMER_TYPE_SEND) &&
(tmr->type != SCTP_TIMER_TYPE_RECV) &&
(tmr->type != SCTP_TIMER_TYPE_HEARTBEAT) &&
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
(tmr->type != SCTP_TIMER_TYPE_SHUTDOWN) &&
(tmr->type != SCTP_TIMER_TYPE_SHUTDOWNACK) &&
(tmr->type != SCTP_TIMER_TYPE_SHUTDOWNGUARD) &&
(tmr->type != SCTP_TIMER_TYPE_ASOCKILL))
) {
SCTP_INP_DECR_REF(inp);
Get SCTP working in combination with VIMAGE. Contains code from bz. Approved by: rrs (mentor) MFC after: 1 month.
2009-09-19 14:02:16 +00:00
CURVNET_RESTORE();
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
}
}
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
tmr->stopped_from = 0xa004;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb) {
- fix source address selection when picking an acceptable address - name change of prefered -> preferred - CMT fast recover code added. - Comment fixes in CMT. - We were not giving a reason of cant_start_asoc per socket api if we failed to get init/or/cookie to bring up an assoc. Change so we don't just give a generic "comm lost" but look at actual states of dying assoc. - change "crc32" arguments to "crc32c" to silence strict/noisy compiler warnings when crc32() is also declared - A few minor tweaks to get the portable stuff truely portable for sctp6_usrreq.c :-D - one-2-one style vrf match problem. - window recovery would leave chks marked for retran during window probes on the sent queue. This would then cause an out-of-order problem and assure that the flight size "problem" would occur. - Solves a flight size logging issue that caused rwnd overruns, flight size off as well as false retransmissions.g - Macroize the up and down of flight size. - Fix a ECNE bug in its counting. - The strict_sacks options was causing aborts when window probing was active, fix to make strict sacks a bit smarter about what the next unsent TSN is. - Fixes a one-2-one wakeup bug found by Martin Kulas. - If-defed out form, Andre's copy routines pending his commit of at least m_last().. need to adjust for 6.2 as well.. since m_last won't exist. Reviewed by: gnn
2007-04-14 09:44:09 +00:00
atomic_add_int(&stcb->asoc.refcnt, 1);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb->asoc.state == 0) {
- fix source address selection when picking an acceptable address - name change of prefered -> preferred - CMT fast recover code added. - Comment fixes in CMT. - We were not giving a reason of cant_start_asoc per socket api if we failed to get init/or/cookie to bring up an assoc. Change so we don't just give a generic "comm lost" but look at actual states of dying assoc. - change "crc32" arguments to "crc32c" to silence strict/noisy compiler warnings when crc32() is also declared - A few minor tweaks to get the portable stuff truely portable for sctp6_usrreq.c :-D - one-2-one style vrf match problem. - window recovery would leave chks marked for retran during window probes on the sent queue. This would then cause an out-of-order problem and assure that the flight size "problem" would occur. - Solves a flight size logging issue that caused rwnd overruns, flight size off as well as false retransmissions.g - Macroize the up and down of flight size. - Fix a ECNE bug in its counting. - The strict_sacks options was causing aborts when window probing was active, fix to make strict sacks a bit smarter about what the next unsent TSN is. - Fixes a one-2-one wakeup bug found by Martin Kulas. - If-defed out form, Andre's copy routines pending his commit of at least m_last().. need to adjust for 6.2 as well.. since m_last won't exist. Reviewed by: gnn
2007-04-14 09:44:09 +00:00
atomic_add_int(&stcb->asoc.refcnt, -1);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (inp) {
SCTP_INP_DECR_REF(inp);
}
Get SCTP working in combination with VIMAGE. Contains code from bz. Approved by: rrs (mentor) MFC after: 1 month.
2009-09-19 14:02:16 +00:00
CURVNET_RESTORE();
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
}
}
Store the timer type for logging, because the timer can be freed during processing the timerout. MFC after: 3 days
2016-01-13 14:28:12 +00:00
type = tmr->type;
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
tmr->stopped_from = 0xa005;
Store the timer type for logging, because the timer can be freed during processing the timerout. MFC after: 3 days
2016-01-13 14:28:12 +00:00
SCTPDBG(SCTP_DEBUG_TIMER1, "Timer type %d goes off\n", type);
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (!SCTP_OS_TIMER_ACTIVE(&tmr->timer)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (inp) {
SCTP_INP_DECR_REF(inp);
}
- Fixes so we won't try to start a timer when we hold a wq lock for the iterator. Panda uses a silly recursive lock they hold through the timer. - Add poor mans wireshark compile option.. - Allocate and start using SCTP_M_XXX for all SCTP_MALLOC() calls. - sysctl now will get back the refcnt for viewing by onlookers. Reviewed by: gnn
2007-05-29 09:29:03 +00:00
if (stcb) {
atomic_add_int(&stcb->asoc.refcnt, -1);
}
Get SCTP working in combination with VIMAGE. Contains code from bz. Approved by: rrs (mentor) MFC after: 1 month.
2009-09-19 14:02:16 +00:00
CURVNET_RESTORE();
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
}
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
tmr->stopped_from = 0xa006;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb) {
SCTP_TCB_LOCK(stcb);
Tons of fixes to get all the 64bit issues removed. This also moves two 16 bit int's to become 32 bit values so we do not have to use atomic_add_16. Most of the changes are %p, casts and other various nasty's that were in the orignal code base. With this commit my machine will now do a build universe.. however I as yet have not tested on a 64bit machine .. it may not work :-(
2006-11-05 13:25:18 +00:00
atomic_add_int(&stcb->asoc.refcnt, -1);
Store the timer type for logging, because the timer can be freed during processing the timerout. MFC after: 3 days
2016-01-13 14:28:12 +00:00
if ((type != SCTP_TIMER_TYPE_ASOCKILL) &&
- Modular congestion control, with RFC2581 being the default. - CMT_PF states added (w/sysctl to turn the PF version on) - sctp_input.c had a missing incr of cookie case when the auth was bad. This meant a free was called without an increment to refcnt, added increment like rest of code. - There was a case, unlikely, when the scope of the destination changed (this is a TSNH case). In that case, it would not free the alloc'ed asoc (in sctp_input.c). - When listed addresses found a colliding cookie/Init, then the collided upon tcb was not unlocked in sctp_pcb.c - Add error checking on arguments of sctp_sendx(3) to prevent it from referencing a NULL pointer. - Fix an error return of sctp_sendx(3), it was returing ENOMEM not -1. - Get assoc id was changed to use the sanctified socket api method for getting a assoc id (PEER_ADDR_INFO instead of PEER_ADDR_PARAMS). - Fix it so a peeled off socket will get a proper error return if it trys to send to a different address then it is connected to. - Fix so that select_a_stream can avoid an endless loop that could hang a caller. - time_entered (state set time) was not being set in all cases to the time we went established. Approved by: re(ken smith)
2007-07-14 09:36:28 +00:00
((stcb->asoc.state == 0) ||
(stcb->asoc.state & SCTP_STATE_ABOUT_TO_BE_FREED))) {
SCTP_TCB_UNLOCK(stcb);
if (inp) {
SCTP_INP_DECR_REF(inp);
}
Get SCTP working in combination with VIMAGE. Contains code from bz. Approved by: rrs (mentor) MFC after: 1 month.
2009-09-19 14:02:16 +00:00
CURVNET_RESTORE();
- Modular congestion control, with RFC2581 being the default. - CMT_PF states added (w/sysctl to turn the PF version on) - sctp_input.c had a missing incr of cookie case when the auth was bad. This meant a free was called without an increment to refcnt, added increment like rest of code. - There was a case, unlikely, when the scope of the destination changed (this is a TSNH case). In that case, it would not free the alloc'ed asoc (in sctp_input.c). - When listed addresses found a colliding cookie/Init, then the collided upon tcb was not unlocked in sctp_pcb.c - Add error checking on arguments of sctp_sendx(3) to prevent it from referencing a NULL pointer. - Fix an error return of sctp_sendx(3), it was returing ENOMEM not -1. - Get assoc id was changed to use the sanctified socket api method for getting a assoc id (PEER_ADDR_INFO instead of PEER_ADDR_PARAMS). - Fix it so a peeled off socket will get a proper error return if it trys to send to a different address then it is connected to. - Fix so that select_a_stream can avoid an endless loop that could hang a caller. - time_entered (state set time) was not being set in all cases to the time we went established. Approved by: re(ken smith)
2007-07-14 09:36:28 +00:00
return;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
/* record in stopped what t-o occured */
Store the timer type for logging, because the timer can be freed during processing the timerout. MFC after: 3 days
2016-01-13 14:28:12 +00:00
tmr->stopped_from = type;
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* mark as being serviced now */
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
if (SCTP_OS_TIMER_PENDING(&tmr->timer)) {
/*
* Callout has been rescheduled.
*/
goto get_out;
}
if (!SCTP_OS_TIMER_ACTIVE(&tmr->timer)) {
/*
* Not active, so no action.
*/
goto get_out;
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_OS_TIMER_DEACTIVATE(&tmr->timer);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* call the handler for the appropriate timer type */
Store the timer type for logging, because the timer can be freed during processing the timerout. MFC after: 3 days
2016-01-13 14:28:12 +00:00
switch (type) {
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
case SCTP_TIMER_TYPE_ZERO_COPY:
- Fixes cstatic issues found by cisco sa tool (missing frees and such on error legs) - align sctp_sockstore to 64 bit boundary ..
2007-06-18 21:59:15 +00:00
if (inp == NULL) {
break;
}
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
if (sctp_is_feature_on(inp, SCTP_PCB_FLAGS_ZERO_COPY_ACTIVE)) {
SCTP_ZERO_COPY_EVENT(inp, inp->sctp_socket);
}
break;
- Take out the broken table-id concept. Panda Routers have a M-VRF concept that is NOT well thought out for a multi-homed transport protocol. So the useless table-id entries passed around need to be removed. - Add a event timer for the zero copy api. - Fix a bug in sctp_timer.c when searching for an alternate with the largest ssthresh (the compare was wrong).
2007-06-01 11:19:54 +00:00
case SCTP_TIMER_TYPE_ZCOPY_SENDQ:
- Fixes cstatic issues found by cisco sa tool (missing frees and such on error legs) - align sctp_sockstore to 64 bit boundary ..
2007-06-18 21:59:15 +00:00
if (inp == NULL) {
break;
}
- Take out the broken table-id concept. Panda Routers have a M-VRF concept that is NOT well thought out for a multi-homed transport protocol. So the useless table-id entries passed around need to be removed. - Add a event timer for the zero copy api. - Fix a bug in sctp_timer.c when searching for an alternate with the largest ssthresh (the compare was wrong).
2007-06-01 11:19:54 +00:00
if (sctp_is_feature_on(inp, SCTP_PCB_FLAGS_ZERO_COPY_ACTIVE)) {
SCTP_ZERO_COPY_SENDQ_EVENT(inp, inp->sctp_socket);
}
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
case SCTP_TIMER_TYPE_ADDR_WQ:
sctp_handle_addr_wq();
break;
case SCTP_TIMER_TYPE_SEND:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_STAT_INCR(sctps_timodata);
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
stcb->asoc.timodata++;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
stcb->asoc.num_send_timers_up--;
if (stcb->asoc.num_send_timers_up < 0) {
stcb->asoc.num_send_timers_up = 0;
}
- Modular congestion control, with RFC2581 being the default. - CMT_PF states added (w/sysctl to turn the PF version on) - sctp_input.c had a missing incr of cookie case when the auth was bad. This meant a free was called without an increment to refcnt, added increment like rest of code. - There was a case, unlikely, when the scope of the destination changed (this is a TSNH case). In that case, it would not free the alloc'ed asoc (in sctp_input.c). - When listed addresses found a colliding cookie/Init, then the collided upon tcb was not unlocked in sctp_pcb.c - Add error checking on arguments of sctp_sendx(3) to prevent it from referencing a NULL pointer. - Fix an error return of sctp_sendx(3), it was returing ENOMEM not -1. - Get assoc id was changed to use the sanctified socket api method for getting a assoc id (PEER_ADDR_INFO instead of PEER_ADDR_PARAMS). - Fix it so a peeled off socket will get a proper error return if it trys to send to a different address then it is connected to. - Fix so that select_a_stream can avoid an endless loop that could hang a caller. - time_entered (state set time) was not being set in all cases to the time we went established. Approved by: re(ken smith)
2007-07-14 09:36:28 +00:00
SCTP_TCB_LOCK_ASSERT(stcb);
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
if (sctp_t3rxt_timer(inp, stcb, net)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* no need to unlock on tcb its gone */
goto out_decr;
}
- Modular congestion control, with RFC2581 being the default. - CMT_PF states added (w/sysctl to turn the PF version on) - sctp_input.c had a missing incr of cookie case when the auth was bad. This meant a free was called without an increment to refcnt, added increment like rest of code. - There was a case, unlikely, when the scope of the destination changed (this is a TSNH case). In that case, it would not free the alloc'ed asoc (in sctp_input.c). - When listed addresses found a colliding cookie/Init, then the collided upon tcb was not unlocked in sctp_pcb.c - Add error checking on arguments of sctp_sendx(3) to prevent it from referencing a NULL pointer. - Fix an error return of sctp_sendx(3), it was returing ENOMEM not -1. - Get assoc id was changed to use the sanctified socket api method for getting a assoc id (PEER_ADDR_INFO instead of PEER_ADDR_PARAMS). - Fix it so a peeled off socket will get a proper error return if it trys to send to a different address then it is connected to. - Fix so that select_a_stream can avoid an endless loop that could hang a caller. - time_entered (state set time) was not being set in all cases to the time we went established. Approved by: re(ken smith)
2007-07-14 09:36:28 +00:00
SCTP_TCB_LOCK_ASSERT(stcb);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#ifdef SCTP_AUDITING_ENABLED
sctp_auditing(4, inp, stcb, net);
#endif
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
sctp_chunk_output(inp, stcb, SCTP_OUTPUT_FROM_T3, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if ((stcb->asoc.num_send_timers_up == 0) &&
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
(stcb->asoc.sent_queue_cnt > 0)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
struct sctp_tmit_chunk *chk;
/*
* safeguard. If there on some on the sent queue
* somewhere but no timers running something is
* wrong... so we start a timer on the first chunk
* on the send queue on whatever net it is sent to.
*/
chk = TAILQ_FIRST(&stcb->asoc.sent_queue);
sctp_timer_start(SCTP_TIMER_TYPE_SEND, inp, stcb,
chk->whoTo);
}
break;
case SCTP_TIMER_TYPE_INIT:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_STAT_INCR(sctps_timoinit);
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
stcb->asoc.timoinit++;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (sctp_t1init_timer(inp, stcb, net)) {
/* no need to unlock on tcb its gone */
goto out_decr;
}
/* We do output but not here */
did_output = 0;
break;
case SCTP_TIMER_TYPE_RECV:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
SCTP_STAT_INCR(sctps_timosack);
stcb->asoc.timosack++;
sctp_send_sack(stcb, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#ifdef SCTP_AUDITING_ENABLED
sctp_auditing(4, inp, stcb, net);
#endif
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
sctp_chunk_output(inp, stcb, SCTP_OUTPUT_FROM_SACK_TMR, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_TIMER_TYPE_SHUTDOWN:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (sctp_shutdown_timer(inp, stcb, net)) {
/* no need to unlock on tcb its gone */
goto out_decr;
}
SCTP_STAT_INCR(sctps_timoshutdown);
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
stcb->asoc.timoshutdown++;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#ifdef SCTP_AUDITING_ENABLED
sctp_auditing(4, inp, stcb, net);
#endif
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
sctp_chunk_output(inp, stcb, SCTP_OUTPUT_FROM_SHUT_TMR, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_TIMER_TYPE_HEARTBEAT:
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
if ((stcb == NULL) || (inp == NULL) || (net == NULL)) {
break;
}
SCTP_STAT_INCR(sctps_timoheartbeat);
stcb->asoc.timoheartbeat++;
if (sctp_heartbeat_timer(inp, stcb, net)) {
/* no need to unlock on tcb its gone */
goto out_decr;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#ifdef SCTP_AUDITING_ENABLED
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
sctp_auditing(4, inp, stcb, net);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#endif
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
if (!(net->dest_state & SCTP_ADDR_NOHB)) {
Update the inp stored in a HB-timer when moving an stcb to a new inp. Use only this stored inp when processing a HB timeout. This fixes a bug which results in a crash. MFC after: 3 days.
2011-10-09 14:12:17 +00:00
sctp_timer_start(SCTP_TIMER_TYPE_HEARTBEAT, inp, stcb, net);
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
sctp_chunk_output(inp, stcb, SCTP_OUTPUT_FROM_HB_TMR, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
break;
case SCTP_TIMER_TYPE_COOKIE:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (sctp_cookie_timer(inp, stcb, net)) {
/* no need to unlock on tcb its gone */
goto out_decr;
}
SCTP_STAT_INCR(sctps_timocookie);
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
stcb->asoc.timocookie++;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#ifdef SCTP_AUDITING_ENABLED
sctp_auditing(4, inp, stcb, net);
#endif
/*
* We consider T3 and Cookie timer pretty much the same with
* respect to where from in chunk_output.
*/
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
sctp_chunk_output(inp, stcb, SCTP_OUTPUT_FROM_T3, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_TIMER_TYPE_NEWCOOKIE:
{
struct timeval tv;
int i, secret;
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (inp == NULL) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_STAT_INCR(sctps_timosecret);
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
(void)SCTP_GETTIME_TIMEVAL(&tv);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_INP_WLOCK(inp);
inp->sctp_ep.time_of_secret_change = tv.tv_sec;
inp->sctp_ep.last_secret_number =
inp->sctp_ep.current_secret_number;
inp->sctp_ep.current_secret_number++;
if (inp->sctp_ep.current_secret_number >=
SCTP_HOW_MANY_SECRETS) {
inp->sctp_ep.current_secret_number = 0;
}
secret = (int)inp->sctp_ep.current_secret_number;
for (i = 0; i < SCTP_NUMBER_OF_SECRETS; i++) {
inp->sctp_ep.secret_key[secret][i] =
sctp_select_initial_TSN(&inp->sctp_ep);
}
SCTP_INP_WUNLOCK(inp);
sctp_timer_start(SCTP_TIMER_TYPE_NEWCOOKIE, inp, stcb, net);
}
did_output = 0;
break;
case SCTP_TIMER_TYPE_PATHMTURAISE:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_STAT_INCR(sctps_timopathmtu);
sctp_pathmtu_timer(inp, stcb, net);
did_output = 0;
break;
case SCTP_TIMER_TYPE_SHUTDOWNACK:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (sctp_shutdownack_timer(inp, stcb, net)) {
/* no need to unlock on tcb its gone */
goto out_decr;
}
SCTP_STAT_INCR(sctps_timoshutdownack);
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
stcb->asoc.timoshutdownack++;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#ifdef SCTP_AUDITING_ENABLED
sctp_auditing(4, inp, stcb, net);
#endif
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
sctp_chunk_output(inp, stcb, SCTP_OUTPUT_FROM_SHUT_ACK_TMR, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_TIMER_TYPE_SHUTDOWNGUARD:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_STAT_INCR(sctps_timoshutdownguard);
Provide consistent error causes whenever an ABORT chunk is sent. MFC after: 1 week
2015-07-27 22:35:54 +00:00
op_err = sctp_generate_cause(SCTP_BASE_SYSCTL(sctp_diag_info_code),
"Shutdown guard timer expired");
sctp_abort_an_association(inp, stcb, op_err, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* no need to unlock on tcb its gone */
goto out_decr;
case SCTP_TIMER_TYPE_STRRESET:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (sctp_strreset_timer(inp, stcb, net)) {
/* no need to unlock on tcb its gone */
goto out_decr;
}
SCTP_STAT_INCR(sctps_timostrmrst);
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
sctp_chunk_output(inp, stcb, SCTP_OUTPUT_FROM_STRRST_TMR, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_TIMER_TYPE_ASCONF:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (sctp_asconf_timer(inp, stcb, net)) {
/* no need to unlock on tcb its gone */
goto out_decr;
}
SCTP_STAT_INCR(sctps_timoasconf);
#ifdef SCTP_AUDITING_ENABLED
sctp_auditing(4, inp, stcb, net);
#endif
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
sctp_chunk_output(inp, stcb, SCTP_OUTPUT_FROM_ASCONF_TMR, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
case SCTP_TIMER_TYPE_PRIM_DELETED:
if ((stcb == NULL) || (inp == NULL)) {
break;
}
- Incorrect error EAGAIN returned for invalid send on a locked stream (using EEOR mode). Changed to EINVAL (in sctp_output.c) - Static analysis comments added - fix in mobility code to return a value (static analysis found). - sctp6_notify function made visible instead of static (this is needed for Panda). Approved by: re@freebsd.org (B Mah)
2007-09-13 10:36:43 +00:00
sctp_delete_prim_timer(inp, stcb, net);
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
SCTP_STAT_INCR(sctps_timodelprim);
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
case SCTP_TIMER_TYPE_AUTOCLOSE:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_STAT_INCR(sctps_timoautoclose);
sctp_autoclose_timer(inp, stcb, net);
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
sctp_chunk_output(inp, stcb, SCTP_OUTPUT_FROM_AUTOCLOSE_TMR, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
did_output = 0;
break;
case SCTP_TIMER_TYPE_ASOCKILL:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((stcb == NULL) || (inp == NULL)) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_STAT_INCR(sctps_timoassockill);
/* Can we free it yet? */
SCTP_INP_DECR_REF(inp);
Some more debug info cleanup. MFC after: 3 days
2015-05-28 16:39:22 +00:00
sctp_timer_stop(SCTP_TIMER_TYPE_ASOCKILL, inp, stcb, NULL,
SCTP_FROM_SCTPUTIL + SCTP_LOC_1);
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
so = SCTP_INP_SO(inp);
atomic_add_int(&stcb->asoc.refcnt, 1);
SCTP_TCB_UNLOCK(stcb);
SCTP_SOCKET_LOCK(so, 1);
SCTP_TCB_LOCK(stcb);
atomic_subtract_int(&stcb->asoc.refcnt, 1);
#endif
Some more debug info cleanup. MFC after: 3 days
2015-05-28 16:39:22 +00:00
(void)sctp_free_assoc(inp, stcb, SCTP_NORMAL_PROC,
SCTP_FROM_SCTPUTIL + SCTP_LOC_2);
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
SCTP_SOCKET_UNLOCK(so, 1);
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* free asoc, always unlocks (or destroy's) so prevent
* duplicate unlock or unlock of a free mtx :-0
*/
stcb = NULL;
goto out_no_decr;
case SCTP_TIMER_TYPE_INPKILL:
SCTP_STAT_INCR(sctps_timoinpkill);
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (inp == NULL) {
break;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* special case, take away our increment since WE are the
* killer
*/
SCTP_INP_DECR_REF(inp);
Some more debug info cleanup. MFC after: 3 days
2015-05-28 16:39:22 +00:00
sctp_timer_stop(SCTP_TIMER_TYPE_INPKILL, inp, NULL, NULL,
SCTP_FROM_SCTPUTIL + SCTP_LOC_3);
- Get rid of the sctp_inpcb_free() "magic numbers", now they are sensible defines that tell what you are directing the function to do.
2007-05-08 15:53:03 +00:00
sctp_inpcb_free(inp, SCTP_FREE_SHOULD_USE_ABORT,
Ok, yet another bug in killing off all the hundreds of apitesters.. Basically we end up with attempting to destroy a lock thats contended on. A cookie echo arrives at the same time that the close is happening. The close gets the lock but the cookie echo has already passed the check for the gone flag and is then locked waiting on the create lock.. when we go to destroy it bam. For now we do the timer destroy for all calls to close.. We can probably optimize this later so that we check whats being contended on and if there is contention then do the timer thing. but this is probably safest since the inp has been removed from all lists and references and only the timer can find it.. once the locks are released all other places will instantly see the GONE flag and bail (thats what the change in sctp_input is one place that was lacking the bail code). MFC after: 1 week
2010-06-06 19:24:32 +00:00
SCTP_CALLED_FROM_INPKILL_TIMER);
- sctputil.c - If debug is on, the INPKILL timer can deref a freed value. Change so that we save off a type field for display and NULL inp just for good measure. - sctp_output.c - Fix it so in sending to the loopback we use the src address of the inbound INIT. We don't want to do this for non local addresses since otherwise we might be ingressed filtered so we need to use the best src address and list the address sent to. Obtained from: time bug - Neil Wilson MFC after: 1 week
2008-05-21 16:51:21 +00:00
inp = NULL;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
goto out_no_decr;
default:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTPDBG(SCTP_DEBUG_TIMER1, "sctp_timeout_handler:unknown timer %d\n",
Store the timer type for logging, because the timer can be freed during processing the timerout. MFC after: 3 days
2016-01-13 14:28:12 +00:00
type);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#ifdef SCTP_AUDITING_ENABLED
Store the timer type for logging, because the timer can be freed during processing the timerout. MFC after: 3 days
2016-01-13 14:28:12 +00:00
sctp_audit_log(0xF1, (uint8_t) type);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (inp)
sctp_auditing(5, inp, stcb, net);
#endif
if ((did_output) && stcb) {
/*
* Now we need to clean up the control chunk chain if an
* ECNE is on it. It must be marked as UNSENT again so next
* call will continue to send it until such time that we get
* a CWR, to remove it. It is, however, less likely that we
* will find a ecn echo on the chain though.
*/
sctp_fix_ecn_echo(&stcb->asoc);
}
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
get_out:
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb) {
SCTP_TCB_UNLOCK(stcb);
}
out_decr:
if (inp) {
SCTP_INP_DECR_REF(inp);
}
out_no_decr:
Store the timer type for logging, because the timer can be freed during processing the timerout. MFC after: 3 days
2016-01-13 14:28:12 +00:00
SCTPDBG(SCTP_DEBUG_TIMER1, "Timer now complete (type = %d)\n", type);
Get SCTP working in combination with VIMAGE. Contains code from bz. Approved by: rrs (mentor) MFC after: 1 month.
2009-09-19 14:02:16 +00:00
CURVNET_RESTORE();
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
void
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_timer_start(int t_type, struct sctp_inpcb *inp, struct sctp_tcb *stcb,
struct sctp_nets *net)
{
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
uint32_t to_ticks;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
struct sctp_timer *tmr;
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if ((t_type != SCTP_TIMER_TYPE_ADDR_WQ) && (inp == NULL))
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
tmr = NULL;
if (stcb) {
SCTP_TCB_LOCK_ASSERT(stcb);
}
switch (t_type) {
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
case SCTP_TIMER_TYPE_ZERO_COPY:
tmr = &inp->sctp_ep.zero_copy_timer;
to_ticks = SCTP_ZERO_COPY_TICK_DELAY;
break;
- Take out the broken table-id concept. Panda Routers have a M-VRF concept that is NOT well thought out for a multi-homed transport protocol. So the useless table-id entries passed around need to be removed. - Add a event timer for the zero copy api. - Fix a bug in sctp_timer.c when searching for an alternate with the largest ssthresh (the compare was wrong).
2007-06-01 11:19:54 +00:00
case SCTP_TIMER_TYPE_ZCOPY_SENDQ:
tmr = &inp->sctp_ep.zero_copy_sendq_timer;
to_ticks = SCTP_ZERO_COPY_SENDQ_TICK_DELAY;
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
case SCTP_TIMER_TYPE_ADDR_WQ:
/* Only 1 tick away :-) */
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
tmr = &SCTP_BASE_INFO(addr_wq_timer);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
to_ticks = SCTP_ADDRESS_TICK_DELAY;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_TIMER_TYPE_SEND:
/* Here we use the RTO timer */
{
int rto_val;
if ((stcb == NULL) || (net == NULL)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &net->rxt_timer;
if (net->RTO == 0) {
rto_val = stcb->asoc.initial_rto;
} else {
rto_val = net->RTO;
}
to_ticks = MSEC_TO_TICKS(rto_val);
}
break;
case SCTP_TIMER_TYPE_INIT:
/*
* Here we use the INIT timer default usually about 1
* minute.
*/
if ((stcb == NULL) || (net == NULL)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &net->rxt_timer;
if (net->RTO == 0) {
to_ticks = MSEC_TO_TICKS(stcb->asoc.initial_rto);
} else {
to_ticks = MSEC_TO_TICKS(net->RTO);
}
break;
case SCTP_TIMER_TYPE_RECV:
/*
* Here we use the Delayed-Ack timer value from the inp
* ususually about 200ms.
*/
if (stcb == NULL) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &stcb->asoc.dack_timer;
to_ticks = MSEC_TO_TICKS(stcb->asoc.delayed_ack);
break;
case SCTP_TIMER_TYPE_SHUTDOWN:
/* Here we use the RTO of the destination. */
if ((stcb == NULL) || (net == NULL)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (net->RTO == 0) {
to_ticks = MSEC_TO_TICKS(stcb->asoc.initial_rto);
} else {
to_ticks = MSEC_TO_TICKS(net->RTO);
}
tmr = &net->rxt_timer;
break;
case SCTP_TIMER_TYPE_HEARTBEAT:
/*
* the net is used here so that we can add in the RTO. Even
* though we use a different timer. We also add the HB timer
* PLUS a random jitter.
*/
Don't check twice that inp is not NULL. Reported by: Coverity CID: 748671 MFC after: 3 days
2014-12-21 13:58:53 +00:00
if ((stcb == NULL) || (net == NULL)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
} else {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
uint32_t rndval;
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
uint32_t jitter;
if ((net->dest_state & SCTP_ADDR_NOHB) &&
!(net->dest_state & SCTP_ADDR_UNCONFIRMED)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
if (net->RTO == 0) {
to_ticks = stcb->asoc.initial_rto;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
to_ticks = net->RTO;
}
rndval = sctp_select_initial_TSN(&inp->sctp_ep);
jitter = rndval % to_ticks;
if (jitter >= (to_ticks >> 1)) {
to_ticks = to_ticks + (jitter - (to_ticks >> 1));
} else {
to_ticks = to_ticks - jitter;
}
if (!(net->dest_state & SCTP_ADDR_UNCONFIRMED) &&
!(net->dest_state & SCTP_ADDR_PF)) {
to_ticks += net->heart_beat_delay;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
/*
* Now we must convert the to_ticks that are now in
* ms to ticks.
*/
to_ticks = MSEC_TO_TICKS(to_ticks);
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
tmr = &net->hb_timer;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
break;
case SCTP_TIMER_TYPE_COOKIE:
/*
* Here we can use the RTO timer from the network since one
* RTT was compelete. If a retran happened then we will be
* using the RTO initial value.
*/
if ((stcb == NULL) || (net == NULL)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (net->RTO == 0) {
to_ticks = MSEC_TO_TICKS(stcb->asoc.initial_rto);
} else {
to_ticks = MSEC_TO_TICKS(net->RTO);
}
tmr = &net->rxt_timer;
break;
case SCTP_TIMER_TYPE_NEWCOOKIE:
/*
* nothing needed but the endpoint here ususually about 60
* minutes.
*/
tmr = &inp->sctp_ep.signature_change;
to_ticks = inp->sctp_ep.sctp_timeoutticks[SCTP_TIMER_SIGNATURE];
break;
case SCTP_TIMER_TYPE_ASOCKILL:
if (stcb == NULL) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &stcb->asoc.strreset_timer;
to_ticks = MSEC_TO_TICKS(SCTP_ASOC_KILL_TIMEOUT);
break;
case SCTP_TIMER_TYPE_INPKILL:
/*
* The inp is setup to die. We re-use the signature_chage
* timer since that has stopped and we are in the GONE
* state.
*/
tmr = &inp->sctp_ep.signature_change;
to_ticks = MSEC_TO_TICKS(SCTP_INP_KILL_TIMEOUT);
break;
case SCTP_TIMER_TYPE_PATHMTURAISE:
/*
* Here we use the value found in the EP for PMTU ususually
* about 10 minutes.
*/
Don't check twice that inp is not NULL. Reported by: Coverity CID: 748671 MFC after: 3 days
2014-12-21 13:58:53 +00:00
if ((stcb == NULL) || (net == NULL)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Fix the enabling/disabling of Heartbeats and path MTU discovery when using the SCTP_PEER_ADDR_PARAMS socket option. Approved by: re MFC after: 1 month.
2011-09-17 08:50:29 +00:00
if (net->dest_state & SCTP_ADDR_NO_PMTUD) {
return;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
to_ticks = inp->sctp_ep.sctp_timeoutticks[SCTP_TIMER_PMTU];
tmr = &net->pmtu_timer;
break;
case SCTP_TIMER_TYPE_SHUTDOWNACK:
/* Here we use the RTO of the destination */
if ((stcb == NULL) || (net == NULL)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (net->RTO == 0) {
to_ticks = MSEC_TO_TICKS(stcb->asoc.initial_rto);
} else {
to_ticks = MSEC_TO_TICKS(net->RTO);
}
tmr = &net->rxt_timer;
break;
case SCTP_TIMER_TYPE_SHUTDOWNGUARD:
/*
* Here we use the endpoints shutdown guard timer usually
* about 3 minutes.
*/
Don't check twice that inp is not NULL. Reported by: Coverity CID: 748671 MFC after: 3 days
2014-12-21 13:58:53 +00:00
if (stcb == NULL) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Use 5 times RTO.Max as the default for the shutdown guard timer as required by RFC 4960. The sysctl variable can be used to overwrite this. Discussed with: rrs MFC after: 1 week
2015-08-29 17:26:29 +00:00
if (inp->sctp_ep.sctp_timeoutticks[SCTP_TIMER_MAXSHUTDOWN] == 0) {
to_ticks = 5 * MSEC_TO_TICKS(stcb->asoc.maxrto);
} else {
to_ticks = inp->sctp_ep.sctp_timeoutticks[SCTP_TIMER_MAXSHUTDOWN];
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
tmr = &stcb->asoc.shut_guard_timer;
break;
case SCTP_TIMER_TYPE_STRRESET:
/*
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
* Here the timer comes from the stcb but its value is from
* the net's RTO.
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
*/
if ((stcb == NULL) || (net == NULL)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (net->RTO == 0) {
to_ticks = MSEC_TO_TICKS(stcb->asoc.initial_rto);
} else {
to_ticks = MSEC_TO_TICKS(net->RTO);
}
tmr = &stcb->asoc.strreset_timer;
break;
case SCTP_TIMER_TYPE_ASCONF:
/*
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
* Here the timer comes from the stcb but its value is from
* the net's RTO.
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
*/
if ((stcb == NULL) || (net == NULL)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (net->RTO == 0) {
to_ticks = MSEC_TO_TICKS(stcb->asoc.initial_rto);
} else {
to_ticks = MSEC_TO_TICKS(net->RTO);
}
tmr = &stcb->asoc.asconf_timer;
break;
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
case SCTP_TIMER_TYPE_PRIM_DELETED:
if ((stcb == NULL) || (net != NULL)) {
return;
}
to_ticks = MSEC_TO_TICKS(stcb->asoc.initial_rto);
tmr = &stcb->asoc.delete_prim_timer;
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
case SCTP_TIMER_TYPE_AUTOCLOSE:
if (stcb == NULL) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (stcb->asoc.sctp_autoclose_ticks == 0) {
/*
* Really an error since stcb is NOT set to
* autoclose
*/
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
to_ticks = stcb->asoc.sctp_autoclose_ticks;
tmr = &stcb->asoc.autoclose_timer;
break;
default:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTPDBG(SCTP_DEBUG_TIMER1, "%s: Unknown timer type %d\n",
Use __func__ instead of __FUNCTION__. This allows to compile the userland stack without errors using gcc5. Thanks to saghul for makeing me aware and providing the patch. MFC after: 1 week
2015-10-19 11:17:54 +00:00
__func__, t_type);
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if ((to_ticks <= 0) || (tmr == NULL)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTPDBG(SCTP_DEBUG_TIMER1, "%s: %d:software error to_ticks:%d tmr:%p not set ??\n",
Use __func__ instead of __FUNCTION__. This allows to compile the userland stack without errors using gcc5. Thanks to saghul for makeing me aware and providing the patch. MFC after: 1 week
2015-10-19 11:17:54 +00:00
__func__, t_type, to_ticks, (void *)tmr);
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (SCTP_OS_TIMER_PENDING(&tmr->timer)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* we do NOT allow you to have it already running. if it is
* we leave the current one up unchanged
*/
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
/* At this point we can proceed */
if (t_type == SCTP_TIMER_TYPE_SEND) {
stcb->asoc.num_send_timers_up++;
}
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
tmr->stopped_from = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
tmr->type = t_type;
tmr->ep = (void *)inp;
tmr->tcb = (void *)stcb;
tmr->net = (void *)net;
tmr->self = (void *)tmr;
Get SCTP working in combination with VIMAGE. Contains code from bz. Approved by: rrs (mentor) MFC after: 1 month.
2009-09-19 14:02:16 +00:00
tmr->vnet = (void *)curvnet;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
tmr->ticks = sctp_get_tick_count();
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
(void)SCTP_OS_TIMER_START(&tmr->timer, to_ticks, sctp_timeout_handler, tmr);
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
void
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_timer_stop(int t_type, struct sctp_inpcb *inp, struct sctp_tcb *stcb,
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
struct sctp_nets *net, uint32_t from)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
struct sctp_timer *tmr;
if ((t_type != SCTP_TIMER_TYPE_ADDR_WQ) &&
(inp == NULL))
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
tmr = NULL;
if (stcb) {
SCTP_TCB_LOCK_ASSERT(stcb);
}
switch (t_type) {
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
case SCTP_TIMER_TYPE_ZERO_COPY:
tmr = &inp->sctp_ep.zero_copy_timer;
break;
- Take out the broken table-id concept. Panda Routers have a M-VRF concept that is NOT well thought out for a multi-homed transport protocol. So the useless table-id entries passed around need to be removed. - Add a event timer for the zero copy api. - Fix a bug in sctp_timer.c when searching for an alternate with the largest ssthresh (the compare was wrong).
2007-06-01 11:19:54 +00:00
case SCTP_TIMER_TYPE_ZCOPY_SENDQ:
tmr = &inp->sctp_ep.zero_copy_sendq_timer;
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
case SCTP_TIMER_TYPE_ADDR_WQ:
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
tmr = &SCTP_BASE_INFO(addr_wq_timer);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_TIMER_TYPE_SEND:
if ((stcb == NULL) || (net == NULL)) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &net->rxt_timer;
break;
case SCTP_TIMER_TYPE_INIT:
if ((stcb == NULL) || (net == NULL)) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &net->rxt_timer;
break;
case SCTP_TIMER_TYPE_RECV:
if (stcb == NULL) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &stcb->asoc.dack_timer;
break;
case SCTP_TIMER_TYPE_SHUTDOWN:
if ((stcb == NULL) || (net == NULL)) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &net->rxt_timer;
break;
case SCTP_TIMER_TYPE_HEARTBEAT:
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
if ((stcb == NULL) || (net == NULL)) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
tmr = &net->hb_timer;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_TIMER_TYPE_COOKIE:
if ((stcb == NULL) || (net == NULL)) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &net->rxt_timer;
break;
case SCTP_TIMER_TYPE_NEWCOOKIE:
/* nothing needed but the endpoint here */
tmr = &inp->sctp_ep.signature_change;
/*
* We re-use the newcookie timer for the INP kill timer. We
* must assure that we do not kill it by accident.
*/
break;
case SCTP_TIMER_TYPE_ASOCKILL:
/*
* Stop the asoc kill timer.
*/
if (stcb == NULL) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &stcb->asoc.strreset_timer;
break;
case SCTP_TIMER_TYPE_INPKILL:
/*
* The inp is setup to die. We re-use the signature_chage
* timer since that has stopped and we are in the GONE
* state.
*/
tmr = &inp->sctp_ep.signature_change;
break;
case SCTP_TIMER_TYPE_PATHMTURAISE:
if ((stcb == NULL) || (net == NULL)) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &net->pmtu_timer;
break;
case SCTP_TIMER_TYPE_SHUTDOWNACK:
if ((stcb == NULL) || (net == NULL)) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &net->rxt_timer;
break;
case SCTP_TIMER_TYPE_SHUTDOWNGUARD:
if (stcb == NULL) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &stcb->asoc.shut_guard_timer;
break;
case SCTP_TIMER_TYPE_STRRESET:
if (stcb == NULL) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &stcb->asoc.strreset_timer;
break;
case SCTP_TIMER_TYPE_ASCONF:
if (stcb == NULL) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &stcb->asoc.asconf_timer;
break;
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
case SCTP_TIMER_TYPE_PRIM_DELETED:
if (stcb == NULL) {
return;
}
tmr = &stcb->asoc.delete_prim_timer;
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
case SCTP_TIMER_TYPE_AUTOCLOSE:
if (stcb == NULL) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
tmr = &stcb->asoc.autoclose_timer;
break;
default:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTPDBG(SCTP_DEBUG_TIMER1, "%s: Unknown timer type %d\n",
Use __func__ instead of __FUNCTION__. This allows to compile the userland stack without errors using gcc5. Thanks to saghul for makeing me aware and providing the patch. MFC after: 1 week
2015-10-19 11:17:54 +00:00
__func__, t_type);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (tmr == NULL) {
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if ((tmr->type != t_type) && tmr->type) {
/*
* Ok we have a timer that is under joint use. Cookie timer
* per chance with the SEND timer. We therefore are NOT
* running the timer that the caller wants stopped. So just
* return.
*/
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if ((t_type == SCTP_TIMER_TYPE_SEND) && (stcb != NULL)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
stcb->asoc.num_send_timers_up--;
if (stcb->asoc.num_send_timers_up < 0) {
stcb->asoc.num_send_timers_up = 0;
}
}
tmr->self = NULL;
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
tmr->stopped_from = from;
- Static analyisis fixes for cisco's commit (this is equivilant to the coverity tool.. may even be the same one.. not sure). - A bug in the way sctp_abort() and friends were setting the IP_CLOSE flag.. and NOT passing the last argument as a (,1)... so that things would get freed..
2007-05-08 14:32:53 +00:00
(void)SCTP_OS_TIMER_STOP(&tmr->timer);
return;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
uint32_t
sctp_calculate_len(struct mbuf *m)
{
uint32_t tlen = 0;
struct mbuf *at;
at = m;
while (at) {
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
tlen += SCTP_BUF_LEN(at);
at = SCTP_BUF_NEXT(at);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
return (tlen);
}
void
sctp_mtu_size_reset(struct sctp_inpcb *inp,
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
struct sctp_association *asoc, uint32_t mtu)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
/*
* Reset the P-MTU size on this association, this involves changing
* the asoc MTU, going through ANY chunk+overhead larger than mtu to
* allow the DF flag to be cleared.
*/
struct sctp_tmit_chunk *chk;
unsigned int eff_mtu, ovh;
asoc->smallest_mtu = mtu;
if (inp->sctp_flags & SCTP_PCB_FLAGS_BOUND_V6) {
ovh = SCTP_MIN_OVERHEAD;
} else {
ovh = SCTP_MIN_V4_OVERHEAD;
}
eff_mtu = mtu - ovh;
TAILQ_FOREACH(chk, &asoc->send_queue, sctp_next) {
if (chk->send_size > eff_mtu) {
chk->flags |= CHUNK_FLAGS_FRAGMENT_OK;
}
}
TAILQ_FOREACH(chk, &asoc->sent_queue, sctp_next) {
if (chk->send_size > eff_mtu) {
chk->flags |= CHUNK_FLAGS_FRAGMENT_OK;
}
}
}
/*
* given an association and starting time of the current RTT period return
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
* RTO in number of msecs net should point to the current network
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
*/
1) Allow a chunk to track the cwnd it was at when sent. 2) Add separate max-bursts for retransmit and hb. These are set to sysctlable values but not settable via the socket api. This makes sure we don't blast out HB's or fast-retransmits. 3) Determine on the first data transmission on a net if its local-lan (by being under or over a RTT). This can later be used to think about different algorithms based on locallan vs big-i (experimental) 4) The cwnd should NOT be allowed to grow when an ECNEcho is seen (TCP has this same bug). We fix this in SCTP so an ECNe being seen prevents an advance of cwnd. 5) CWR's should not be sent multiple times to the same network, instead just updating the TSN being transmitted if needed. MFC after: 1 Month
2011-02-02 11:13:23 +00:00
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
uint32_t
sctp_calculate_rto(struct sctp_tcb *stcb,
struct sctp_association *asoc,
struct sctp_nets *net,
- added pre-checks to the bindx call. - use proper tick gathering macro instead of ticks directly. - Placed reasonable boundaries on sets that a user can do that are converted to ticks from ms. - Fix CMT_PF to always check to be sure CMT is on. - Fix ticks use of CMT_PF. - put back code to allow asconfs to be queued while INITs are in flight and before the assoc is established. - During window probes, an ack'd packet might be left with the window probe mark on it causing it to be retransmitted. Change so that the flight decrease macro clears the window_probe mark. - Additional logging flight size/reading and ASOC LOG. This is only enabled if you manually insert things into opt_sctp.h since its a set of debug code only. - Found an interesting SMP race in the way data was appended which could cause a reader to lose a part of a message, had to reorder when we marked the message was complete to after the data was appended. - bug in ADD-IP for the subset bound socket case when the peer has only one address - fix ASCONF implicit success/error handling case - proper support of jails in Freebsd 6> - copy out the timeval for the 64 bit sparc world on cookie-echo alignment error crashes without this). Approved by: re(Ken Smith)
2007-07-17 20:58:26 +00:00
struct timeval *told,
Tunes and fixes the new DC-CC to seem to hit the right mix. Still may need some tweaks but it appears to almost not give away too much to an RFC2581 flow, but can really minimize the amount of buffers used in the net. MFC after: 3 months
2011-03-08 11:58:25 +00:00
int safe, int rtt_from_sack)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
- added pre-checks to the bindx call. - use proper tick gathering macro instead of ticks directly. - Placed reasonable boundaries on sets that a user can do that are converted to ticks from ms. - Fix CMT_PF to always check to be sure CMT is on. - Fix ticks use of CMT_PF. - put back code to allow asconfs to be queued while INITs are in flight and before the assoc is established. - During window probes, an ack'd packet might be left with the window probe mark on it causing it to be retransmitted. Change so that the flight decrease macro clears the window_probe mark. - Additional logging flight size/reading and ASOC LOG. This is only enabled if you manually insert things into opt_sctp.h since its a set of debug code only. - Found an interesting SMP race in the way data was appended which could cause a reader to lose a part of a message, had to reorder when we marked the message was complete to after the data was appended. - bug in ADD-IP for the subset bound socket case when the peer has only one address - fix ASCONF implicit success/error handling case - proper support of jails in Freebsd 6> - copy out the timeval for the 64 bit sparc world on cookie-echo alignment error crashes without this). Approved by: re(Ken Smith)
2007-07-17 20:58:26 +00:00
/*-
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
* given an association and the starting time of the current RTT
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
* period (in value1/value2) return RTO in number of msecs.
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
*/
* Cleanup the code computing the retransmission timeout. * Fix an initialization bug for the scaled variance of the RTO. MFC after: 3 months.
2011-02-24 22:36:40 +00:00
int32_t rtt; /* RTT in ms */
uint32_t new_rto;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
int first_measure = 0;
- added pre-checks to the bindx call. - use proper tick gathering macro instead of ticks directly. - Placed reasonable boundaries on sets that a user can do that are converted to ticks from ms. - Fix CMT_PF to always check to be sure CMT is on. - Fix ticks use of CMT_PF. - put back code to allow asconfs to be queued while INITs are in flight and before the assoc is established. - During window probes, an ack'd packet might be left with the window probe mark on it causing it to be retransmitted. Change so that the flight decrease macro clears the window_probe mark. - Additional logging flight size/reading and ASOC LOG. This is only enabled if you manually insert things into opt_sctp.h since its a set of debug code only. - Found an interesting SMP race in the way data was appended which could cause a reader to lose a part of a message, had to reorder when we marked the message was complete to after the data was appended. - bug in ADD-IP for the subset bound socket case when the peer has only one address - fix ASCONF implicit success/error handling case - proper support of jails in Freebsd 6> - copy out the timeval for the 64 bit sparc world on cookie-echo alignment error crashes without this). Approved by: re(Ken Smith)
2007-07-17 20:58:26 +00:00
struct timeval now, then, *old;
/* Copy it out for sparc64 */
if (safe == sctp_align_unsafe_makecopy) {
old = &then;
memcpy(&then, told, sizeof(struct timeval));
} else if (safe == sctp_align_safe_nocopy) {
old = told;
} else {
/* error */
SCTP_PRINTF("Huh, bad rto calc call\n");
return (0);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/************************/
/* 1. calculate new RTT */
/************************/
/* get the current time */
Improvements to CC modules: 1) Add four new points that allow you to get more information to cc algo's 2) Fix the case where user changes module on a existing TCB, in such a case, the initialization module needs to be called on all nets. 3) Move htcp_cc structure to a union that other modules can use. 4) Add 5th point for get/set socket options for cc_module specific options MFC after: 2 months
2011-02-26 15:23:46 +00:00
if (stcb->asoc.use_precise_time) {
(void)SCTP_GETPTIME_TIMEVAL(&now);
} else {
(void)SCTP_GETTIME_TIMEVAL(&now);
}
* Cleanup the code computing the retransmission timeout. * Fix an initialization bug for the scaled variance of the RTO. MFC after: 3 months.
2011-02-24 22:36:40 +00:00
timevalsub(&now, old);
/* store the current RTT in us */
Fix a typo which results in RTT to be off by a factor of 10, if the RTT is larger than 1 second. MFC after: 3 days
2012-09-02 12:37:30 +00:00
net->rtt = (uint64_t) 1000000 *(uint64_t) now.tv_sec +
Whitespace change. MFC after: 3 days
2012-09-04 22:40:49 +00:00
(uint64_t) now.tv_usec;
* Cleanup the code computing the retransmission timeout. * Fix an initialization bug for the scaled variance of the RTO. MFC after: 3 months.
2011-02-24 22:36:40 +00:00
Add a explict cast to silence a warning when building the userland stack on Windows. This issue was reported by Peter Kasting from Google. MFC after: 3 days
2014-09-16 14:39:24 +00:00
/* compute rtt in ms */
rtt = (int32_t) (net->rtt / 1000);
Tunes and fixes the new DC-CC to seem to hit the right mix. Still may need some tweaks but it appears to almost not give away too much to an RFC2581 flow, but can really minimize the amount of buffers used in the net. MFC after: 3 months
2011-03-08 11:58:25 +00:00
if ((asoc->cc_functions.sctp_rtt_calculated) && (rtt_from_sack == SCTP_RTT_FROM_DATA)) {
/*
* Tell the CC module that a new update has just occurred
* from a sack
*/
(*asoc->cc_functions.sctp_rtt_calculated) (stcb, net, &now);
}
/*
* Do we need to determine the lan? We do this only on sacks i.e.
* RTT being determined from data not non-data (HB/INIT->INITACK).
*/
if ((rtt_from_sack == SCTP_RTT_FROM_DATA) &&
* Cleanup the code computing the retransmission timeout. * Fix an initialization bug for the scaled variance of the RTO. MFC after: 3 months.
2011-02-24 22:36:40 +00:00
(net->lan_type == SCTP_LAN_UNKNOWN)) {
if (net->rtt > SCTP_LOCAL_LAN_RTT) {
1) Allow a chunk to track the cwnd it was at when sent. 2) Add separate max-bursts for retransmit and hb. These are set to sysctlable values but not settable via the socket api. This makes sure we don't blast out HB's or fast-retransmits. 3) Determine on the first data transmission on a net if its local-lan (by being under or over a RTT). This can later be used to think about different algorithms based on locallan vs big-i (experimental) 4) The cwnd should NOT be allowed to grow when an ECNEcho is seen (TCP has this same bug). We fix this in SCTP so an ECNe being seen prevents an advance of cwnd. 5) CWR's should not be sent multiple times to the same network, instead just updating the TSN being transmitted if needed. MFC after: 1 Month
2011-02-02 11:13:23 +00:00
net->lan_type = SCTP_LAN_INTERNET;
} else {
net->lan_type = SCTP_LAN_LOCAL;
}
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/***************************/
/* 2. update RTTVAR & SRTT */
/***************************/
* Cleanup the code computing the retransmission timeout. * Fix an initialization bug for the scaled variance of the RTO. MFC after: 3 months.
2011-02-24 22:36:40 +00:00
/*-
* Compute the scaled average lastsa and the
* scaled variance lastsv as described in van Jacobson
* Paper "Congestion Avoidance and Control", Annex A.
*
* (net->lastsa >> SCTP_RTT_SHIFT) is the srtt
* (net->lastsa >> SCTP_RTT_VAR_SHIFT) is the rttvar
*/
- Fixed cookie handling to calc an RTO when its an INIT collision case. - Fixed RTO calc to maintain a seperate variable to track if a RTO calc as been done, this allows the RTO var to be doubled during initial timeouts. - Reduces the amount of stack used by process control. - Use a constant for the peer chunk overhead. - Name change to spell candidate correctly.
2007-06-13 01:31:53 +00:00
if (net->RTO_measured) {
* Cleanup the code computing the retransmission timeout. * Fix an initialization bug for the scaled variance of the RTO. MFC after: 3 months.
2011-02-24 22:36:40 +00:00
rtt -= (net->lastsa >> SCTP_RTT_SHIFT);
net->lastsa += rtt;
if (rtt < 0) {
rtt = -rtt;
}
rtt -= (net->lastsv >> SCTP_RTT_VAR_SHIFT);
net->lastsv += rtt;
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_RTTVAR_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
rto_logging(net, SCTP_LOG_RTTVAR);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
/* First RTO measurment */
- Fixed cookie handling to calc an RTO when its an INIT collision case. - Fixed RTO calc to maintain a seperate variable to track if a RTO calc as been done, this allows the RTO var to be doubled during initial timeouts. - Reduces the amount of stack used by process control. - Use a constant for the peer chunk overhead. - Name change to spell candidate correctly.
2007-06-13 01:31:53 +00:00
net->RTO_measured = 1;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
first_measure = 1;
* Cleanup the code computing the retransmission timeout. * Fix an initialization bug for the scaled variance of the RTO. MFC after: 3 months.
2011-02-24 22:36:40 +00:00
net->lastsa = rtt << SCTP_RTT_SHIFT;
net->lastsv = (rtt / 2) << SCTP_RTT_VAR_SHIFT;
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_RTTVAR_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
rto_logging(net, SCTP_LOG_INITIAL_RTT);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
* Cleanup the code computing the retransmission timeout. * Fix an initialization bug for the scaled variance of the RTO. MFC after: 3 months.
2011-02-24 22:36:40 +00:00
if (net->lastsv == 0) {
net->lastsv = SCTP_CLOCK_GRANULARITY;
}
- RTO was not being initialized to 0, thus the rtt calculation algoritm would not go through the proper initialization. - The initialization was incorrect as well, causing problems in sat networks with > 1sec RTT - Get rid of magic numbers in RTT calculations.
2007-06-08 10:57:11 +00:00
new_rto = (net->lastsa >> SCTP_RTT_SHIFT) + net->lastsv;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if ((new_rto > SCTP_SAT_NETWORK_MIN) &&
(stcb->asoc.sat_network_lockout == 0)) {
stcb->asoc.sat_network = 1;
} else if ((!first_measure) && stcb->asoc.sat_network) {
stcb->asoc.sat_network = 0;
stcb->asoc.sat_network_lockout = 1;
}
/* bound it, per C6/C7 in Section 5.3.1 */
if (new_rto < stcb->asoc.minrto) {
new_rto = stcb->asoc.minrto;
}
if (new_rto > stcb->asoc.maxrto) {
new_rto = stcb->asoc.maxrto;
}
- Found bug in min split point bundling which caused incorrect, non-bundlable fragmentation. - Added min residual to better control split points for both how big a msg must be as well as how much needs to be left over. - With our new algo in place, we need to implicitly set "end of msg" on the sp-> structure otherwise we end up with "hung" associations. - Room reserved up front in IP header by pushing IP header to back of mbuf. - Fix so FR's peg count of retransmissions needed. - Fix so an unlucky chunk that never gets across will kill the assoc via the kill timer and send an abort too. - Fix bug in sctp_input which can result in a crash. - Do not strip off IP options anymore. - Clean up sctp_calculate_rto(). - Get rid of unused sysctl. - Fixed so we discard all M-Cast - Fixed so port check done AFTER checksum - Fixed bug in fragmentation code that prevented us from fragmenting a small complete message when we needed to. - Window probes were not marked back to unsent and flight adjusted when a sack came in with no window change or accepting of the probe data. We now fix this with having a mark on the net and the chunk so we can clear it out when the sack arrives forcing it to retran just like it was "new" this improves the handling of window probes, which were dropped by the receiver. - Tighten AUTH protocol error checks during INIT/INIT-ACK exchange
2007-03-31 11:47:30 +00:00
/* we are now returning the RTO */
return (new_rto);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
/*
* return a pointer to a contiguous piece of data from the given mbuf chain
* starting at 'off' for 'len' bytes. If the desired piece spans more than
* one mbuf, a copy is made at 'ptr'. caller must ensure that the buffer size
* is >= 'len' returns NULL if there there isn't 'len' bytes in the chain.
*/
- Matthew's changes to get inlines out, plus a few of my own to deal with the VRF inline function -> becomes a macro now. Submitted by: Matthew Jacobs
2007-06-16 00:33:47 +00:00
caddr_t
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_m_getptr(struct mbuf *m, int off, int len, uint8_t * in_ptr)
{
uint32_t count;
uint8_t *ptr;
ptr = in_ptr;
if ((off < 0) || (len <= 0))
return (NULL);
/* find the desired start location */
while ((m != NULL) && (off > 0)) {
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (off < SCTP_BUF_LEN(m))
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
off -= SCTP_BUF_LEN(m);
m = SCTP_BUF_NEXT(m);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (m == NULL)
return (NULL);
/* is the current mbuf large enough (eg. contiguous)? */
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if ((SCTP_BUF_LEN(m) - off) >= len) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (mtod(m, caddr_t)+off);
} else {
/* else, it spans more than one mbuf, so save a temp copy... */
while ((m != NULL) && (len > 0)) {
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
count = min(SCTP_BUF_LEN(m) - off, len);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
bcopy(mtod(m, caddr_t)+off, ptr, count);
len -= count;
ptr += count;
off = 0;
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
m = SCTP_BUF_NEXT(m);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if ((m == NULL) && (len > 0))
return (NULL);
else
return ((caddr_t)in_ptr);
}
}
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
struct sctp_paramhdr *
sctp_get_next_param(struct mbuf *m,
int offset,
struct sctp_paramhdr *pull,
int pull_limit)
{
/* This just provides a typed signature to Peter's Pull routine */
return ((struct sctp_paramhdr *)sctp_m_getptr(m, offset, pull_limit,
(uint8_t *) pull));
}
Cleanup sctp_send_initiate() and sctp_send_initiate_ack() to be in sync as much as possible. This simplifies upcoming changes.
2014-08-01 12:42:37 +00:00
struct mbuf *
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_add_pad_tombuf(struct mbuf *m, int padlen)
{
Cleanup sctp_send_initiate() and sctp_send_initiate_ack() to be in sync as much as possible. This simplifies upcoming changes.
2014-08-01 12:42:37 +00:00
struct mbuf *m_last;
caddr_t dp;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (padlen > 3) {
Cleanup sctp_send_initiate() and sctp_send_initiate_ack() to be in sync as much as possible. This simplifies upcoming changes.
2014-08-01 12:42:37 +00:00
return (NULL);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- More fixes for lock misses on the transfer of data to the sent_queue. Sometimes I wonder why any code ever works :-) - Fix the pad of the last mbuf routine, It was working improperly on non-4 byte aligned chunks which could cause memory overruns. MFC after: 1 week
2007-12-07 01:32:14 +00:00
if (padlen <= M_TRAILINGSPACE(m)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* The easy way. We hope the majority of the time we hit
* here :)
*/
Cleanup sctp_send_initiate() and sctp_send_initiate_ack() to be in sync as much as possible. This simplifies upcoming changes.
2014-08-01 12:42:37 +00:00
m_last = m;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
Cleanup sctp_send_initiate() and sctp_send_initiate_ack() to be in sync as much as possible. This simplifies upcoming changes.
2014-08-01 12:42:37 +00:00
/* Hard way we must grow the mbuf chain */
m_last = sctp_get_mbuf_for_msg(padlen, 0, M_NOWAIT, 1, MT_DATA);
if (m_last == NULL) {
return (NULL);
}
SCTP_BUF_LEN(m_last) = 0;
SCTP_BUF_NEXT(m_last) = NULL;
SCTP_BUF_NEXT(m) = m_last;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Cleanup sctp_send_initiate() and sctp_send_initiate_ack() to be in sync as much as possible. This simplifies upcoming changes.
2014-08-01 12:42:37 +00:00
dp = mtod(m_last, caddr_t)+SCTP_BUF_LEN(m_last);
SCTP_BUF_LEN(m_last) += padlen;
memset(dp, 0, padlen);
return (m_last);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Cleanup sctp_send_initiate() and sctp_send_initiate_ack() to be in sync as much as possible. This simplifies upcoming changes.
2014-08-01 12:42:37 +00:00
struct mbuf *
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_pad_lastmbuf(struct mbuf *m, int padval, struct mbuf *last_mbuf)
{
/* find the last mbuf in chain and pad it */
struct mbuf *m_at;
Cleanup sctp_send_initiate() and sctp_send_initiate_ack() to be in sync as much as possible. This simplifies upcoming changes.
2014-08-01 12:42:37 +00:00
if (last_mbuf != NULL) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (sctp_add_pad_tombuf(last_mbuf, padval));
} else {
Small cleanup. MFC after: 3 days
2012-06-12 09:18:28 +00:00
for (m_at = m; m_at; m_at = SCTP_BUF_NEXT(m_at)) {
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (SCTP_BUF_NEXT(m_at) == NULL) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (sctp_add_pad_tombuf(m_at, padval));
}
}
}
Cleanup sctp_send_initiate() and sctp_send_initiate_ack() to be in sync as much as possible. This simplifies upcoming changes.
2014-08-01 12:42:37 +00:00
return (NULL);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
static void
Only provide the supported features in the SCTP_ASSOC_CHANGE notif if the state is SCTP_COMM_UP or SCTP_RESTART. While there, do some cleanups. MFC after: 3 days
2012-05-11 18:07:36 +00:00
sctp_notify_assoc_change(uint16_t state, struct sctp_tcb *stcb,
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
uint16_t error, struct sctp_abort_chunk *abort, uint8_t from_peer, int so_locked
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
struct mbuf *m_notify;
struct sctp_assoc_change *sac;
struct sctp_queued_to_read *control;
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
unsigned int notif_len;
uint16_t abort_len;
Add support for the sac_info field in struct sctp_assoc_change as required by RFC 6458. MFC after: 3 days
2012-05-06 13:09:13 +00:00
unsigned int i;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
struct socket *so;
#endif
Add a NULL pointer check to silence the clang code analyzer. MFC after: 1 week
2015-09-04 09:22:16 +00:00
if (stcb == NULL) {
return;
}
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
if (sctp_stcb_is_feature_on(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVASSOCEVNT)) {
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
notif_len = (unsigned int)sizeof(struct sctp_assoc_change);
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
if (abort != NULL) {
Use htons()/ntohs() appropriately. These issues were reported by Andrew Galante. MFC after: 3 days
2013-11-07 16:37:12 +00:00
abort_len = ntohs(abort->ch.chunk_length);
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
} else {
abort_len = 0;
}
if ((state == SCTP_COMM_UP) || (state == SCTP_RESTART)) {
notif_len += SCTP_ASSOC_SUPPORTS_MAX;
} else if ((state == SCTP_COMM_LOST) || (state == SCTP_CANT_STR_ASSOC)) {
notif_len += abort_len;
}
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(notif_len, 0, M_NOWAIT, 1, MT_DATA);
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
if (m_notify == NULL) {
/* Retry with smaller value. */
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
notif_len = (unsigned int)sizeof(struct sctp_assoc_change);
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(notif_len, 0, M_NOWAIT, 1, MT_DATA);
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
if (m_notify == NULL) {
goto set_error;
}
}
SCTP_BUF_NEXT(m_notify) = NULL;
sac = mtod(m_notify, struct sctp_assoc_change *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(sac, 0, notif_len);
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
sac->sac_type = SCTP_ASSOC_CHANGE;
sac->sac_flags = 0;
sac->sac_length = sizeof(struct sctp_assoc_change);
sac->sac_state = state;
sac->sac_error = error;
/* XXX verify these stream counts */
sac->sac_outbound_streams = stcb->asoc.streamoutcnt;
sac->sac_inbound_streams = stcb->asoc.streamincnt;
sac->sac_assoc_id = sctp_get_associd(stcb);
if (notif_len > sizeof(struct sctp_assoc_change)) {
if ((state == SCTP_COMM_UP) || (state == SCTP_RESTART)) {
i = 0;
Add support for the SCTP_AUTH_SUPPORTED and SCTP_ASCONF_SUPPORTED socket options. Add also a sysctl to control the support of ASCONF. MFC after: 1 week
2014-08-12 11:30:16 +00:00
if (stcb->asoc.prsctp_supported == 1) {
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
sac->sac_info[i++] = SCTP_ASSOC_SUPPORTS_PR;
}
Add support for the SCTP_AUTH_SUPPORTED and SCTP_ASCONF_SUPPORTED socket options. Add also a sysctl to control the support of ASCONF. MFC after: 1 week
2014-08-12 11:30:16 +00:00
if (stcb->asoc.auth_supported == 1) {
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
sac->sac_info[i++] = SCTP_ASSOC_SUPPORTS_AUTH;
}
Add support for the SCTP_AUTH_SUPPORTED and SCTP_ASCONF_SUPPORTED socket options. Add also a sysctl to control the support of ASCONF. MFC after: 1 week
2014-08-12 11:30:16 +00:00
if (stcb->asoc.asconf_supported == 1) {
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
sac->sac_info[i++] = SCTP_ASSOC_SUPPORTS_ASCONF;
}
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
if (stcb->asoc.idata_supported == 1) {
sac->sac_info[i++] = SCTP_ASSOC_SUPPORTS_INTERLEAVING;
}
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
sac->sac_info[i++] = SCTP_ASSOC_SUPPORTS_MULTIBUF;
Add support for the SCTP_AUTH_SUPPORTED and SCTP_ASCONF_SUPPORTED socket options. Add also a sysctl to control the support of ASCONF. MFC after: 1 week
2014-08-12 11:30:16 +00:00
if (stcb->asoc.reconfig_supported == 1) {
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
sac->sac_info[i++] = SCTP_ASSOC_SUPPORTS_RE_CONFIG;
}
sac->sac_length += i;
} else if ((state == SCTP_COMM_LOST) || (state == SCTP_CANT_STR_ASSOC)) {
memcpy(sac->sac_info, abort, abort_len);
sac->sac_length += abort_len;
}
}
SCTP_BUF_LEN(m_notify) = sac->sac_length;
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
0, 0, stcb->asoc.context, 0, 0, 0,
m_notify);
if (control != NULL) {
control->length = SCTP_BUF_LEN(m_notify);
/* not that we need this */
control->tail_mbuf = m_notify;
control->spec_flags = M_NOTIFICATION;
sctp_add_to_readq(stcb->sctp_ep, stcb,
control,
&stcb->sctp_socket->so_rcv, 1, SCTP_READ_LOCK_NOT_HELD,
so_locked);
} else {
sctp_m_freem(m_notify);
}
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
* For 1-to-1 style sockets, we send up and error when an ABORT
* comes in.
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
*/
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
set_error:
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (((stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_TCPTYPE) ||
(stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_IN_TCPPOOL)) &&
Only provide the supported features in the SCTP_ASSOC_CHANGE notif if the state is SCTP_COMM_UP or SCTP_RESTART. While there, do some cleanups. MFC after: 3 days
2012-05-11 18:07:36 +00:00
((state == SCTP_COMM_LOST) || (state == SCTP_CANT_STR_ASSOC))) {
Fix a potential race in returning setting errno when an association goes down. Reported by Mozilla in https://bugzilla.mozilla.org/show_bug.cgi?id=845513 MFC after: 3 days
2013-02-27 19:51:47 +00:00
SOCK_LOCK(stcb->sctp_socket);
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
if (from_peer) {
if (SCTP_GET_STATE(&stcb->asoc) == SCTP_STATE_COOKIE_WAIT) {
SCTP_LTRACE_ERR_RET(NULL, stcb, NULL, SCTP_FROM_SCTPUTIL, ECONNREFUSED);
stcb->sctp_socket->so_error = ECONNREFUSED;
} else {
SCTP_LTRACE_ERR_RET(NULL, stcb, NULL, SCTP_FROM_SCTPUTIL, ECONNRESET);
stcb->sctp_socket->so_error = ECONNRESET;
}
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
} else {
Set errno to ETIMEDOUT if an SCTP association times out during setup. MFC after: 1 week
2013-05-17 22:26:05 +00:00
if ((SCTP_GET_STATE(&stcb->asoc) == SCTP_STATE_COOKIE_WAIT) ||
(SCTP_GET_STATE(&stcb->asoc) == SCTP_STATE_COOKIE_ECHOED)) {
SCTP_LTRACE_ERR_RET(NULL, stcb, NULL, SCTP_FROM_SCTPUTIL, ETIMEDOUT);
stcb->sctp_socket->so_error = ETIMEDOUT;
} else {
SCTP_LTRACE_ERR_RET(NULL, stcb, NULL, SCTP_FROM_SCTPUTIL, ECONNABORTED);
stcb->sctp_socket->so_error = ECONNABORTED;
}
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
}
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
}
/* Wake ANY sleepers */
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
so = SCTP_INP_SO(stcb->sctp_ep);
if (!so_locked) {
atomic_add_int(&stcb->asoc.refcnt, 1);
SCTP_TCB_UNLOCK(stcb);
SCTP_SOCKET_LOCK(so, 1);
SCTP_TCB_LOCK(stcb);
atomic_subtract_int(&stcb->asoc.refcnt, 1);
if (stcb->asoc.state & SCTP_STATE_CLOSED_SOCKET) {
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
SCTP_SOCKET_UNLOCK(so, 1);
Provide in the association change notification the received ABORT chunk if case of SCTP_COMM_LOST or SCTP_CANT_STR_ASSOC as required by RFC 6458. MFC after: 3 days
2012-05-12 20:11:35 +00:00
return;
}
}
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
#endif
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
if (((stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_TCPTYPE) ||
(stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_IN_TCPPOOL)) &&
((state == SCTP_COMM_LOST) || (state == SCTP_CANT_STR_ASSOC))) {
Fix a potential race in returning setting errno when an association goes down. Reported by Mozilla in https://bugzilla.mozilla.org/show_bug.cgi?id=845513 MFC after: 3 days
2013-02-27 19:51:47 +00:00
socantrcvmore_locked(stcb->sctp_socket);
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
}
sorwakeup(stcb->sctp_socket);
sowwakeup(stcb->sctp_socket);
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
if (!so_locked) {
SCTP_SOCKET_UNLOCK(so, 1);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets. MFC after: 3 days
2012-05-13 16:07:53 +00:00
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
static void
sctp_notify_peer_addr_change(struct sctp_tcb *stcb, uint32_t state,
Keep track on the socket lock state. This fixes a bug showing up on Mac OS X. MFC after: 3 days
2015-03-10 22:38:10 +00:00
struct sockaddr *sa, uint32_t error, int so_locked
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
struct mbuf *m_notify;
struct sctp_paddr_change *spc;
struct sctp_queued_to_read *control;
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
if ((stcb == NULL) ||
sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVPADDREVNT)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* event not enabled */
return;
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
}
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(sizeof(struct sctp_paddr_change), 0, M_NOWAIT, 1, MT_DATA);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (m_notify == NULL)
return;
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_LEN(m_notify) = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
spc = mtod(m_notify, struct sctp_paddr_change *);
Initialize notification strucuture. This was missed in an earlier commit MFC after: 3 days
2014-07-24 18:06:18 +00:00
memset(spc, 0, sizeof(struct sctp_paddr_change));
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
spc->spc_type = SCTP_PEER_ADDR_CHANGE;
spc->spc_flags = 0;
spc->spc_length = sizeof(struct sctp_paddr_change);
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
switch (sa->sa_family) {
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#ifdef INET
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
case AF_INET:
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#ifdef INET6
if (sctp_is_feature_on(stcb->sctp_ep, SCTP_PCB_FLAGS_NEEDS_MAPPED_V4)) {
in6_sin_2_v4mapsin6((struct sockaddr_in *)sa,
(struct sockaddr_in6 *)&spc->spc_aaddr);
} else {
memcpy(&spc->spc_aaddr, sa, sizeof(struct sockaddr_in));
}
#else
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
memcpy(&spc->spc_aaddr, sa, sizeof(struct sockaddr_in));
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#endif
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
break;
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#endif
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#ifdef INET6
case AF_INET6:
{
struct sockaddr_in6 *sin6;
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
memcpy(&spc->spc_aaddr, sa, sizeof(struct sockaddr_in6));
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
sin6 = (struct sockaddr_in6 *)&spc->spc_aaddr;
if (IN6_IS_SCOPE_LINKLOCAL(&sin6->sin6_addr)) {
if (sin6->sin6_scope_id == 0) {
/* recover scope_id for user */
(void)sa6_recoverscope(sin6);
} else {
/* clear embedded scope_id for user */
in6_clearscope(&sin6->sin6_addr);
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
break;
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#endif
default:
/* TSNH */
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
spc->spc_state = state;
spc->spc_error = error;
spc->spc_assoc_id = sctp_get_associd(stcb);
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_LEN(m_notify) = sizeof(struct sctp_paddr_change);
SCTP_BUF_NEXT(m_notify) = NULL;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* append to socket */
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
0, 0, stcb->asoc.context, 0, 0, 0,
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
m_notify);
if (control == NULL) {
/* no memory */
sctp_m_freem(m_notify);
return;
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
control->length = SCTP_BUF_LEN(m_notify);
control->spec_flags = M_NOTIFICATION;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* not that we need this */
control->tail_mbuf = m_notify;
sctp_add_to_readq(stcb->sctp_ep, stcb,
control,
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
&stcb->sctp_socket->so_rcv, 1,
SCTP_READ_LOCK_NOT_HELD,
Keep track on the socket lock state. This fixes a bug showing up on Mac OS X. MFC after: 3 days
2015-03-10 22:38:10 +00:00
so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
static void
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
sctp_notify_send_failed(struct sctp_tcb *stcb, uint8_t sent, uint32_t error,
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
struct sctp_tmit_chunk *chk, int so_locked
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
struct mbuf *m_notify;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
struct sctp_send_failed *ssf;
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
struct sctp_send_failed_event *ssfe;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
struct sctp_queued_to_read *control;
int length;
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
if ((stcb == NULL) ||
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
(sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVSENDFAILEVNT) &&
sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVNSENDFAILEVNT))) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* event not enabled */
return;
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
}
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
if (sctp_stcb_is_feature_on(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVNSENDFAILEVNT)) {
length = sizeof(struct sctp_send_failed_event);
} else {
length = sizeof(struct sctp_send_failed);
}
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(length, 0, M_NOWAIT, 1, MT_DATA);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (m_notify == NULL)
/* no space left */
return;
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_LEN(m_notify) = 0;
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
if (sctp_stcb_is_feature_on(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVNSENDFAILEVNT)) {
ssfe = mtod(m_notify, struct sctp_send_failed_event *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(ssfe, 0, length);
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
ssfe->ssfe_type = SCTP_SEND_FAILED_EVENT;
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
if (sent) {
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
ssfe->ssfe_flags = SCTP_DATA_SENT;
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
} else {
ssfe->ssfe_flags = SCTP_DATA_UNSENT;
}
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
length += chk->send_size;
length -= sizeof(struct sctp_data_chunk);
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
ssfe->ssfe_length = length;
ssfe->ssfe_error = error;
/* not exactly what the user sent in, but should be close :) */
ssfe->ssfe_info.snd_sid = chk->rec.data.stream_number;
ssfe->ssfe_info.snd_flags = chk->rec.data.rcv_flags;
ssfe->ssfe_info.snd_ppid = chk->rec.data.payloadtype;
ssfe->ssfe_info.snd_context = chk->rec.data.context;
ssfe->ssfe_info.snd_assoc_id = sctp_get_associd(stcb);
ssfe->ssfe_assoc_id = sctp_get_associd(stcb);
SCTP_BUF_LEN(m_notify) = sizeof(struct sctp_send_failed_event);
} else {
ssf = mtod(m_notify, struct sctp_send_failed *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(ssf, 0, length);
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
ssf->ssf_type = SCTP_SEND_FAILED;
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
if (sent) {
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
ssf->ssf_flags = SCTP_DATA_SENT;
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
} else {
ssf->ssf_flags = SCTP_DATA_UNSENT;
}
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
length += chk->send_size;
length -= sizeof(struct sctp_data_chunk);
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
ssf->ssf_length = length;
ssf->ssf_error = error;
/* not exactly what the user sent in, but should be close :) */
bzero(&ssf->ssf_info, sizeof(ssf->ssf_info));
ssf->ssf_info.sinfo_stream = chk->rec.data.stream_number;
ssf->ssf_info.sinfo_ssn = chk->rec.data.stream_seq;
ssf->ssf_info.sinfo_flags = chk->rec.data.rcv_flags;
ssf->ssf_info.sinfo_ppid = chk->rec.data.payloadtype;
ssf->ssf_info.sinfo_context = chk->rec.data.context;
ssf->ssf_info.sinfo_assoc_id = sctp_get_associd(stcb);
ssf->ssf_assoc_id = sctp_get_associd(stcb);
SCTP_BUF_LEN(m_notify) = sizeof(struct sctp_send_failed);
}
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
if (chk->data) {
/*
* trim off the sctp chunk header(it should be there)
*/
if (chk->send_size >= sizeof(struct sctp_data_chunk)) {
m_adj(chk->data, sizeof(struct sctp_data_chunk));
sctp_mbuf_crush(chk->data);
chk->send_size -= sizeof(struct sctp_data_chunk);
}
}
* Fix a bug where PR-SCTP settings are ignore when using implicit association setup. * Fix a bug where message with illegal stream ids are not deleted. * Fix a crash when reporting back unsent messages from the send_queue. * Fix a bug related to INIT retransmission when the socket is already closed. * Fix a bug where associations were stalled when partial delivery API was enabled. * Fix a bug where the receive buffer size was smaller than the partial_delivery_point. Approved by: re, rrs (mentor) MFC after: One day.
2009-08-15 21:10:52 +00:00
SCTP_BUF_NEXT(m_notify) = chk->data;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Steal off the mbuf */
chk->data = NULL;
/*
* For this case, we check the actual socket buffer, since the assoc
* is going away we don't want to overfill the socket buffer for a
* non-reader
*/
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (sctp_sbspace_failedmsgs(&stcb->sctp_socket->so_rcv) < SCTP_BUF_LEN(m_notify)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_m_freem(m_notify);
return;
}
/* append to socket */
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
0, 0, stcb->asoc.context, 0, 0, 0,
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
m_notify);
if (control == NULL) {
/* no memory */
sctp_m_freem(m_notify);
return;
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
control->spec_flags = M_NOTIFICATION;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_add_to_readq(stcb->sctp_ep, stcb,
control,
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
&stcb->sctp_socket->so_rcv, 1,
SCTP_READ_LOCK_NOT_HELD,
so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
static void
sctp_notify_send_failed2(struct sctp_tcb *stcb, uint32_t error,
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
struct sctp_stream_queue_pending *sp, int so_locked
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
struct mbuf *m_notify;
struct sctp_send_failed *ssf;
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
struct sctp_send_failed_event *ssfe;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
struct sctp_queued_to_read *control;
int length;
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
if ((stcb == NULL) ||
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
(sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVSENDFAILEVNT) &&
sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVNSENDFAILEVNT))) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* event not enabled */
return;
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
}
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
if (sctp_stcb_is_feature_on(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVNSENDFAILEVNT)) {
length = sizeof(struct sctp_send_failed_event);
} else {
length = sizeof(struct sctp_send_failed);
}
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(length, 0, M_NOWAIT, 1, MT_DATA);
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
if (m_notify == NULL) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* no space left */
return;
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_LEN(m_notify) = 0;
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
if (sctp_stcb_is_feature_on(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVNSENDFAILEVNT)) {
ssfe = mtod(m_notify, struct sctp_send_failed_event *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(ssfe, 0, length);
Provide the correct notification type (SCTP_SEND_FAILED_EVENT) for unsent messages. MFC after: 3 days
2012-07-14 21:25:14 +00:00
ssfe->ssfe_type = SCTP_SEND_FAILED_EVENT;
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
ssfe->ssfe_flags = SCTP_DATA_UNSENT;
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
length += sp->length;
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
ssfe->ssfe_length = length;
ssfe->ssfe_error = error;
/* not exactly what the user sent in, but should be close :) */
ssfe->ssfe_info.snd_sid = sp->stream;
if (sp->some_taken) {
ssfe->ssfe_info.snd_flags = SCTP_DATA_LAST_FRAG;
} else {
ssfe->ssfe_info.snd_flags = SCTP_DATA_NOT_FRAG;
}
ssfe->ssfe_info.snd_ppid = sp->ppid;
ssfe->ssfe_info.snd_context = sp->context;
ssfe->ssfe_info.snd_assoc_id = sctp_get_associd(stcb);
ssfe->ssfe_assoc_id = sctp_get_associd(stcb);
SCTP_BUF_LEN(m_notify) = sizeof(struct sctp_send_failed_event);
1) Adds the rest of the VIMAGE change macros 2) Adds some __UserSpace__ on some of the common defines that the user space code needs 3) Fixes a bug when we send up data to a user that failed. We need to a) trim off the data chunk headers, if present, and b) make sure the frag bit is communicated properly for the msgs coming off the stream queues... i.e. we see if some of the msg has been taken. Obtained from: jeli contributed the VIMAGE changes on this pass Thanks Julain!
2008-07-09 16:45:30 +00:00
} else {
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
ssf = mtod(m_notify, struct sctp_send_failed *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(ssf, 0, length);
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
ssf->ssf_type = SCTP_SEND_FAILED;
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
ssf->ssf_flags = SCTP_DATA_UNSENT;
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
length += sp->length;
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
ssf->ssf_length = length;
ssf->ssf_error = error;
/* not exactly what the user sent in, but should be close :) */
ssf->ssf_info.sinfo_stream = sp->stream;
Move from early SSN assignment to late SSN assignment. This doesn't change functionality, but makes upcoming change much easier. Developed with rrs@ at the IETF 85. MFC after: 1 week
2012-11-05 20:55:17 +00:00
ssf->ssf_info.sinfo_ssn = 0;
Add support for SCTP_SEND_FAILED_EVENT as required by RFC 6458. MFC after: 3 days
2012-05-06 11:02:53 +00:00
if (sp->some_taken) {
ssf->ssf_info.sinfo_flags = SCTP_DATA_LAST_FRAG;
} else {
ssf->ssf_info.sinfo_flags = SCTP_DATA_NOT_FRAG;
}
ssf->ssf_info.sinfo_ppid = sp->ppid;
ssf->ssf_info.sinfo_context = sp->context;
ssf->ssf_info.sinfo_assoc_id = sctp_get_associd(stcb);
ssf->ssf_assoc_id = sctp_get_associd(stcb);
SCTP_BUF_LEN(m_notify) = sizeof(struct sctp_send_failed);
1) Adds the rest of the VIMAGE change macros 2) Adds some __UserSpace__ on some of the common defines that the user space code needs 3) Fixes a bug when we send up data to a user that failed. We need to a) trim off the data chunk headers, if present, and b) make sure the frag bit is communicated properly for the msgs coming off the stream queues... i.e. we see if some of the msg has been taken. Obtained from: jeli contributed the VIMAGE changes on this pass Thanks Julain!
2008-07-09 16:45:30 +00:00
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_NEXT(m_notify) = sp->data;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Steal off the mbuf */
sp->data = NULL;
/*
* For this case, we check the actual socket buffer, since the assoc
* is going away we don't want to overfill the socket buffer for a
* non-reader
*/
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (sctp_sbspace_failedmsgs(&stcb->sctp_socket->so_rcv) < SCTP_BUF_LEN(m_notify)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_m_freem(m_notify);
return;
}
/* append to socket */
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
0, 0, stcb->asoc.context, 0, 0, 0,
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
m_notify);
if (control == NULL) {
/* no memory */
sctp_m_freem(m_notify);
return;
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
control->spec_flags = M_NOTIFICATION;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_add_to_readq(stcb->sctp_ep, stcb,
control,
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
&stcb->sctp_socket->so_rcv, 1, SCTP_READ_LOCK_NOT_HELD, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
static void
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
sctp_notify_adaptation_layer(struct sctp_tcb *stcb)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
struct mbuf *m_notify;
struct sctp_adaptation_event *sai;
struct sctp_queued_to_read *control;
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
if ((stcb == NULL) ||
sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_ADAPTATIONEVNT)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* event not enabled */
return;
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
}
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(sizeof(struct sctp_adaption_event), 0, M_NOWAIT, 1, MT_DATA);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (m_notify == NULL)
/* no space left */
return;
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_LEN(m_notify) = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sai = mtod(m_notify, struct sctp_adaptation_event *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(sai, 0, sizeof(struct sctp_adaptation_event));
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sai->sai_type = SCTP_ADAPTATION_INDICATION;
sai->sai_flags = 0;
sai->sai_length = sizeof(struct sctp_adaptation_event);
- During shutdown pending, when the last sack came in and the last message on the send stream was "null" but still there, a state we allow, we could get hung and not clean it up and wait for the shutdown guard timer to clear the association without a graceful close. Fix this so that that we properly clean up. - Added support for Multiple ASCONF per new RFC. We only (so far) accept input of these and cannot yet generate a multi-asconf. - Sysctl'd support for experimental Fast Handover feature. Always disabled unless sysctl or socket option changes to enable. - Error case in add-ip where the peer supports AUTH and ADD-IP but does NOT require AUTH of ASCONF/ASCONF-ACK. We need to ABORT in this case. - According to the Kyoto summit of socket api developers (Solaris, Linux, BSD). We need to have: o non-eeor mode messages be atomic - Fixed o Allow implicit setup of an assoc in 1-2-1 model if using the sctp_**() send calls - Fixed o Get rid of HAVE_XXX declarations - Done o add a sctp_pr_policy in hole in sndrcvinfo structure - Done o add a PR_SCTP_POLICY_VALID type flag - yet to-do in a future patch! - Optimize sctp6 calls to reuse code in sctp_usrreq. Also optimize when we close sending out the data and disabling Nagle. - Change key concatenation order to match the auth RFC - When sending OOTB shutdown_complete always do csum. - Don't send PKT-DROP to a PKT-DROP - For abort chunks just always checksums same for shutdown-complete. - inpcb_free front state had a bug where in queue data could wedge an assoc. We need to just abandon ones in front states (free_assoc). - If a peer sends us a 64k abort, we would try to assemble a response packet which may be larger than 64k. This then would be dropped by IP. Instead make a "minimum" size for us 64k-2k (we want at least 2k for our initack). If we receive such an init discard it early without all the processing. - When we peel off we must increment the tcb ref count to keep it from being freed from underneath us. - handling fwd-tsn had bugs that caused memory overwrites when given faulty data, fixed so can't happen and we also stop at the first bad stream no. - Fixed so comm-up generates the adaption indication. - peeloff did not get the hmac params copied. - fix it so we lock the addr list when doing src-addr selection (in future we need to use a multi-reader/one writer lock here) - During lowlevel output, we could end up with a _l_addr set to null if the iterator is calling the output routine. This means we would possibly crash when we gather the MTU info. Fix so we only do the gather where we have a src address cached. - we need to be sure to set abort flag on conn state when we receive an abort. - peeloff could leak a socket. Moved code so the close will find the socket if the peeloff fails (uipc_syscalls.c) Approved by: re@freebsd.org(Ken Smith)
2007-08-27 05:19:48 +00:00
sai->sai_adaptation_ind = stcb->asoc.peers_adaptation;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sai->sai_assoc_id = sctp_get_associd(stcb);
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_LEN(m_notify) = sizeof(struct sctp_adaptation_event);
SCTP_BUF_NEXT(m_notify) = NULL;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* append to socket */
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
0, 0, stcb->asoc.context, 0, 0, 0,
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
m_notify);
if (control == NULL) {
/* no memory */
sctp_m_freem(m_notify);
return;
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
control->length = SCTP_BUF_LEN(m_notify);
control->spec_flags = M_NOTIFICATION;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* not that we need this */
control->tail_mbuf = m_notify;
sctp_add_to_readq(stcb->sctp_ep, stcb,
control,
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
&stcb->sctp_socket->so_rcv, 1, SCTP_READ_LOCK_NOT_HELD, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
/* This always must be called with the read-queue LOCKED in the INP */
* Fix a bug where PR-SCTP settings are ignore when using implicit association setup. * Fix a bug where message with illegal stream ids are not deleted. * Fix a crash when reporting back unsent messages from the send_queue. * Fix a bug related to INIT retransmission when the socket is already closed. * Fix a bug where associations were stalled when partial delivery API was enabled. * Fix a bug where the receive buffer size was smaller than the partial_delivery_point. Approved by: re, rrs (mentor) MFC after: One day.
2009-08-15 21:10:52 +00:00
static void
- Remove extra comment for 7.0 (no GIANT here). - Remove unneeded WLOCK/UNLOCK of inp for getting TCB lock. - Fix panic that may occur when freeing an assoc that has partial delivery in progress (may dereference null socket pointer when queuing partial delivery aborted notification) - Some spacing and comment fixes. - Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. Approved by: re@freebsd.org (Bruce Mah)
2007-08-16 01:51:22 +00:00
sctp_notify_partial_delivery_indication(struct sctp_tcb *stcb, uint32_t error,
* Fix a bug where PR-SCTP settings are ignore when using implicit association setup. * Fix a bug where message with illegal stream ids are not deleted. * Fix a crash when reporting back unsent messages from the send_queue. * Fix a bug related to INIT retransmission when the socket is already closed. * Fix a bug where associations were stalled when partial delivery API was enabled. * Fix a bug where the receive buffer size was smaller than the partial_delivery_point. Approved by: re, rrs (mentor) MFC after: One day.
2009-08-15 21:10:52 +00:00
uint32_t val, int so_locked
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
struct mbuf *m_notify;
struct sctp_pdapi_event *pdapi;
struct sctp_queued_to_read *control;
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
struct sockbuf *sb;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
if ((stcb == NULL) ||
sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_PDAPIEVNT)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* event not enabled */
return;
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
}
Fix a rece condition in the shutdown handling. The race condition resulted in a panic. MFC after: 3 days
2010-06-18 09:01:44 +00:00
if (stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_SOCKET_CANT_READ) {
return;
}
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(sizeof(struct sctp_pdapi_event), 0, M_NOWAIT, 1, MT_DATA);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (m_notify == NULL)
/* no space left */
return;
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_LEN(m_notify) = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
pdapi = mtod(m_notify, struct sctp_pdapi_event *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(pdapi, 0, sizeof(struct sctp_pdapi_event));
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
pdapi->pdapi_type = SCTP_PARTIAL_DELIVERY_EVENT;
pdapi->pdapi_flags = 0;
pdapi->pdapi_length = sizeof(struct sctp_pdapi_event);
pdapi->pdapi_indication = error;
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
pdapi->pdapi_stream = (val >> 16);
pdapi->pdapi_seq = (val & 0x0000ffff);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
pdapi->pdapi_assoc_id = sctp_get_associd(stcb);
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_LEN(m_notify) = sizeof(struct sctp_pdapi_event);
SCTP_BUF_NEXT(m_notify) = NULL;
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
0, 0, stcb->asoc.context, 0, 0, 0,
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
m_notify);
if (control == NULL) {
/* no memory */
sctp_m_freem(m_notify);
return;
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
control->spec_flags = M_NOTIFICATION;
control->length = SCTP_BUF_LEN(m_notify);
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
/* not that we need this */
control->tail_mbuf = m_notify;
control->held_length = 0;
control->length = 0;
sb = &stcb->sctp_socket->so_rcv;
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(sb, control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBALLOC, SCTP_BUF_LEN(m_notify));
}
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
sctp_sballoc(stcb, sb, m_notify);
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(sb, control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBRESULT, 0);
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
atomic_add_int(&control->length, SCTP_BUF_LEN(m_notify));
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
control->end_added = 1;
if (stcb->asoc.control_pdapi)
TAILQ_INSERT_AFTER(&stcb->sctp_ep->read_queue, stcb->asoc.control_pdapi, control, next);
else {
/* we really should not see this case */
TAILQ_INSERT_TAIL(&stcb->sctp_ep->read_queue, control, next);
}
if (stcb->sctp_ep && stcb->sctp_socket) {
/* This should always be the case */
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
* Fix a bug where PR-SCTP settings are ignore when using implicit association setup. * Fix a bug where message with illegal stream ids are not deleted. * Fix a crash when reporting back unsent messages from the send_queue. * Fix a bug related to INIT retransmission when the socket is already closed. * Fix a bug where associations were stalled when partial delivery API was enabled. * Fix a bug where the receive buffer size was smaller than the partial_delivery_point. Approved by: re, rrs (mentor) MFC after: One day.
2009-08-15 21:10:52 +00:00
struct socket *so;
so = SCTP_INP_SO(stcb->sctp_ep);
if (!so_locked) {
atomic_add_int(&stcb->asoc.refcnt, 1);
SCTP_TCB_UNLOCK(stcb);
SCTP_SOCKET_LOCK(so, 1);
SCTP_TCB_LOCK(stcb);
atomic_subtract_int(&stcb->asoc.refcnt, 1);
if (stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE) {
SCTP_SOCKET_UNLOCK(so, 1);
return;
}
}
#endif
Tons of fixes to get all the 64bit issues removed. This also moves two 16 bit int's to become 32 bit values so we do not have to use atomic_add_16. Most of the changes are %p, casts and other various nasty's that were in the orignal code base. With this commit my machine will now do a build universe.. however I as yet have not tested on a 64bit machine .. it may not work :-(
2006-11-05 13:25:18 +00:00
sctp_sorwakeup(stcb->sctp_ep, stcb->sctp_socket);
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
* Fix a bug where PR-SCTP settings are ignore when using implicit association setup. * Fix a bug where message with illegal stream ids are not deleted. * Fix a crash when reporting back unsent messages from the send_queue. * Fix a bug related to INIT retransmission when the socket is already closed. * Fix a bug where associations were stalled when partial delivery API was enabled. * Fix a bug where the receive buffer size was smaller than the partial_delivery_point. Approved by: re, rrs (mentor) MFC after: One day.
2009-08-15 21:10:52 +00:00
if (!so_locked) {
SCTP_SOCKET_UNLOCK(so, 1);
}
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
static void
sctp_notify_shutdown_event(struct sctp_tcb *stcb)
{
struct mbuf *m_notify;
struct sctp_shutdown_event *sse;
struct sctp_queued_to_read *control;
/*
* For TCP model AND UDP connected sockets we will send an error up
* when an SHUTDOWN completes
*/
if ((stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_TCPTYPE) ||
(stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_IN_TCPPOOL)) {
/* mark socket closed for read/write and wakeup! */
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
struct socket *so;
so = SCTP_INP_SO(stcb->sctp_ep);
atomic_add_int(&stcb->asoc.refcnt, 1);
SCTP_TCB_UNLOCK(stcb);
SCTP_SOCKET_LOCK(so, 1);
SCTP_TCB_LOCK(stcb);
atomic_subtract_int(&stcb->asoc.refcnt, 1);
if (stcb->asoc.state & SCTP_STATE_CLOSED_SOCKET) {
SCTP_SOCKET_UNLOCK(so, 1);
return;
}
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
socantsendmore(stcb->sctp_socket);
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
SCTP_SOCKET_UNLOCK(so, 1);
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Add support for the newly added SCTP API. In particular add support for: * SCTP_SNDINFO, SCTP_PRINFO, SCTP_AUTHINFO, SCTP_DSTADDRV4, and SCTP_DSTADDRV6 cmsgs. * SCTP_NXTINFO and SCTP_RCVINFO cmgs. * SCTP_EVENT, SCTP_RECVRCVINFO, SCTP_RECVNXTINFO and SCTP_DEFAULT_SNDINFO socket option. * Special association ids (SCTP_FUTURE_ASSOC, ...) * sctp_recvv() and sctp_sendv() functions. MFC after: 1 month.
2011-06-15 23:50:27 +00:00
if (sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVSHUTDOWNEVNT)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* event not enabled */
return;
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
}
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(sizeof(struct sctp_shutdown_event), 0, M_NOWAIT, 1, MT_DATA);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (m_notify == NULL)
/* no space left */
return;
sse = mtod(m_notify, struct sctp_shutdown_event *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(sse, 0, sizeof(struct sctp_shutdown_event));
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sse->sse_type = SCTP_SHUTDOWN_EVENT;
sse->sse_flags = 0;
sse->sse_length = sizeof(struct sctp_shutdown_event);
sse->sse_assoc_id = sctp_get_associd(stcb);
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_LEN(m_notify) = sizeof(struct sctp_shutdown_event);
SCTP_BUF_NEXT(m_notify) = NULL;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* append to socket */
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
0, 0, stcb->asoc.context, 0, 0, 0,
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
m_notify);
if (control == NULL) {
/* no memory */
sctp_m_freem(m_notify);
return;
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
control->spec_flags = M_NOTIFICATION;
control->length = SCTP_BUF_LEN(m_notify);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* not that we need this */
control->tail_mbuf = m_notify;
sctp_add_to_readq(stcb->sctp_ep, stcb,
control,
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
&stcb->sctp_socket->so_rcv, 1, SCTP_READ_LOCK_NOT_HELD, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
static void
sctp_notify_sender_dry_event(struct sctp_tcb *stcb,
int so_locked
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
{
struct mbuf *m_notify;
struct sctp_sender_dry_event *event;
struct sctp_queued_to_read *control;
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
if ((stcb == NULL) ||
sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_DRYEVNT)) {
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
/* event not enabled */
return;
}
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(sizeof(struct sctp_sender_dry_event), 0, M_NOWAIT, 1, MT_DATA);
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
if (m_notify == NULL) {
/* no space left */
return;
}
SCTP_BUF_LEN(m_notify) = 0;
event = mtod(m_notify, struct sctp_sender_dry_event *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(event, 0, sizeof(struct sctp_sender_dry_event));
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
event->sender_dry_type = SCTP_SENDER_DRY_EVENT;
event->sender_dry_flags = 0;
event->sender_dry_length = sizeof(struct sctp_sender_dry_event);
event->sender_dry_assoc_id = sctp_get_associd(stcb);
SCTP_BUF_LEN(m_notify) = sizeof(struct sctp_sender_dry_event);
SCTP_BUF_NEXT(m_notify) = NULL;
/* append to socket */
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
0, 0, stcb->asoc.context, 0, 0, 0,
m_notify);
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
if (control == NULL) {
/* no memory */
sctp_m_freem(m_notify);
return;
}
control->length = SCTP_BUF_LEN(m_notify);
control->spec_flags = M_NOTIFICATION;
/* not that we need this */
control->tail_mbuf = m_notify;
sctp_add_to_readq(stcb->sctp_ep, stcb, control,
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
&stcb->sctp_socket->so_rcv, 1, SCTP_READ_LOCK_NOT_HELD, so_locked);
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
}
Add the add-stream capability. Still needs more testing.. MFC after: 1 month
2009-02-20 15:03:54 +00:00
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
void
sctp_notify_stream_reset_add(struct sctp_tcb *stcb, uint16_t numberin, uint16_t numberout, int flag)
Add the add-stream capability. Still needs more testing.. MFC after: 1 month
2009-02-20 15:03:54 +00:00
{
struct mbuf *m_notify;
struct sctp_queued_to_read *control;
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
struct sctp_stream_change_event *stradd;
Add the add-stream capability. Still needs more testing.. MFC after: 1 month
2009-02-20 15:03:54 +00:00
Address clang warnings. MFC after: 3 days
2012-05-06 14:50:54 +00:00
if ((stcb == NULL) ||
(sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_STREAM_CHANGEEVNT))) {
Add the add-stream capability. Still needs more testing.. MFC after: 1 month
2009-02-20 15:03:54 +00:00
/* event not enabled */
return;
}
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
if ((stcb->asoc.peer_req_out) && flag) {
/* Peer made the request, don't tell the local user */
stcb->asoc.peer_req_out = 0;
return;
}
stcb->asoc.peer_req_out = 0;
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
m_notify = sctp_get_mbuf_for_msg(sizeof(struct sctp_stream_change_event), 0, M_NOWAIT, 1, MT_DATA);
Add the add-stream capability. Still needs more testing.. MFC after: 1 month
2009-02-20 15:03:54 +00:00
if (m_notify == NULL)
/* no space left */
return;
SCTP_BUF_LEN(m_notify) = 0;
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
stradd = mtod(m_notify, struct sctp_stream_change_event *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(stradd, 0, sizeof(struct sctp_stream_change_event));
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
stradd->strchange_type = SCTP_STREAM_CHANGE_EVENT;
stradd->strchange_flags = flag;
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
stradd->strchange_length = sizeof(struct sctp_stream_change_event);
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
stradd->strchange_assoc_id = sctp_get_associd(stcb);
stradd->strchange_instrms = numberin;
stradd->strchange_outstrms = numberout;
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
SCTP_BUF_LEN(m_notify) = sizeof(struct sctp_stream_change_event);
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
SCTP_BUF_NEXT(m_notify) = NULL;
if (sctp_sbspace(&stcb->asoc, &stcb->sctp_socket->so_rcv) < SCTP_BUF_LEN(m_notify)) {
/* no space */
sctp_m_freem(m_notify);
return;
}
/* append to socket */
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
0, 0, stcb->asoc.context, 0, 0, 0,
m_notify);
if (control == NULL) {
/* no memory */
sctp_m_freem(m_notify);
return;
}
control->spec_flags = M_NOTIFICATION;
control->length = SCTP_BUF_LEN(m_notify);
/* not that we need this */
control->tail_mbuf = m_notify;
sctp_add_to_readq(stcb->sctp_ep, stcb,
control,
&stcb->sctp_socket->so_rcv, 1, SCTP_READ_LOCK_NOT_HELD, SCTP_SO_NOT_LOCKED);
}
void
sctp_notify_stream_reset_tsn(struct sctp_tcb *stcb, uint32_t sending_tsn, uint32_t recv_tsn, int flag)
{
struct mbuf *m_notify;
struct sctp_queued_to_read *control;
struct sctp_assoc_reset_event *strasoc;
Add the add-stream capability. Still needs more testing.. MFC after: 1 month
2009-02-20 15:03:54 +00:00
Address clang warnings. MFC after: 3 days
2012-05-06 14:50:54 +00:00
if ((stcb == NULL) ||
(sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_ASSOC_RESETEVNT))) {
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
/* event not enabled */
return;
}
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
m_notify = sctp_get_mbuf_for_msg(sizeof(struct sctp_assoc_reset_event), 0, M_NOWAIT, 1, MT_DATA);
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
if (m_notify == NULL)
/* no space left */
return;
SCTP_BUF_LEN(m_notify) = 0;
strasoc = mtod(m_notify, struct sctp_assoc_reset_event *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(strasoc, 0, sizeof(struct sctp_assoc_reset_event));
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
strasoc->assocreset_type = SCTP_ASSOC_RESET_EVENT;
strasoc->assocreset_flags = flag;
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
strasoc->assocreset_length = sizeof(struct sctp_assoc_reset_event);
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
strasoc->assocreset_assoc_id = sctp_get_associd(stcb);
strasoc->assocreset_local_tsn = sending_tsn;
strasoc->assocreset_remote_tsn = recv_tsn;
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
SCTP_BUF_LEN(m_notify) = sizeof(struct sctp_assoc_reset_event);
Add the add-stream capability. Still needs more testing.. MFC after: 1 month
2009-02-20 15:03:54 +00:00
SCTP_BUF_NEXT(m_notify) = NULL;
if (sctp_sbspace(&stcb->asoc, &stcb->sctp_socket->so_rcv) < SCTP_BUF_LEN(m_notify)) {
/* no space */
sctp_m_freem(m_notify);
return;
}
/* append to socket */
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
0, 0, stcb->asoc.context, 0, 0, 0,
Add the add-stream capability. Still needs more testing.. MFC after: 1 month
2009-02-20 15:03:54 +00:00
m_notify);
if (control == NULL) {
/* no memory */
sctp_m_freem(m_notify);
return;
}
control->spec_flags = M_NOTIFICATION;
control->length = SCTP_BUF_LEN(m_notify);
/* not that we need this */
control->tail_mbuf = m_notify;
sctp_add_to_readq(stcb->sctp_ep, stcb,
control,
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
&stcb->sctp_socket->so_rcv, 1, SCTP_READ_LOCK_NOT_HELD, SCTP_SO_NOT_LOCKED);
Add the add-stream capability. Still needs more testing.. MFC after: 1 month
2009-02-20 15:03:54 +00:00
}
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
static void
sctp_notify_stream_reset(struct sctp_tcb *stcb,
int number_entries, uint16_t * list, int flag)
{
struct mbuf *m_notify;
struct sctp_queued_to_read *control;
struct sctp_stream_reset_event *strreset;
int len;
Address clang warnings. MFC after: 3 days
2012-05-06 14:50:54 +00:00
if ((stcb == NULL) ||
(sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_STREAM_RESETEVNT))) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* event not enabled */
return;
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
}
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(MCLBYTES, 0, M_NOWAIT, 1, MT_DATA);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (m_notify == NULL)
/* no space left */
return;
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_LEN(m_notify) = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
len = sizeof(struct sctp_stream_reset_event) + (number_entries * sizeof(uint16_t));
if (len > M_TRAILINGSPACE(m_notify)) {
/* never enough room */
sctp_m_freem(m_notify);
return;
}
strreset = mtod(m_notify, struct sctp_stream_reset_event *);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
memset(strreset, 0, len);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
strreset->strreset_type = SCTP_STREAM_RESET_EVENT;
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
strreset->strreset_flags = flag;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
strreset->strreset_length = len;
strreset->strreset_assoc_id = sctp_get_associd(stcb);
if (number_entries) {
int i;
for (i = 0; i < number_entries; i++) {
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
strreset->strreset_stream_list[i] = ntohs(list[i]);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_LEN(m_notify) = len;
SCTP_BUF_NEXT(m_notify) = NULL;
if (sctp_sbspace(&stcb->asoc, &stcb->sctp_socket->so_rcv) < SCTP_BUF_LEN(m_notify)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* no space */
sctp_m_freem(m_notify);
return;
}
/* append to socket */
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
0, 0, stcb->asoc.context, 0, 0, 0,
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
m_notify);
if (control == NULL) {
/* no memory */
sctp_m_freem(m_notify);
return;
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
control->spec_flags = M_NOTIFICATION;
control->length = SCTP_BUF_LEN(m_notify);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* not that we need this */
control->tail_mbuf = m_notify;
sctp_add_to_readq(stcb->sctp_ep, stcb,
control,
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
&stcb->sctp_socket->so_rcv, 1, SCTP_READ_LOCK_NOT_HELD, SCTP_SO_NOT_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Support SCTP_REMOTE_ERROR notification. MFC after: 3 days
2012-05-13 22:27:54 +00:00
static void
sctp_notify_remote_error(struct sctp_tcb *stcb, uint16_t error, struct sctp_error_chunk *chunk)
{
struct mbuf *m_notify;
struct sctp_remote_error *sre;
struct sctp_queued_to_read *control;
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
unsigned int notif_len;
uint16_t chunk_len;
Support SCTP_REMOTE_ERROR notification. MFC after: 3 days
2012-05-13 22:27:54 +00:00
if ((stcb == NULL) ||
sctp_stcb_is_feature_off(stcb->sctp_ep, stcb, SCTP_PCB_FLAGS_RECVPEERERR)) {
return;
}
if (chunk != NULL) {
Use htons()/ntohs() appropriately. These issues were reported by Andrew Galante. MFC after: 3 days
2013-11-07 16:37:12 +00:00
chunk_len = ntohs(chunk->ch.chunk_length);
Support SCTP_REMOTE_ERROR notification. MFC after: 3 days
2012-05-13 22:27:54 +00:00
} else {
chunk_len = 0;
}
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
notif_len = (unsigned int)(sizeof(struct sctp_remote_error) + chunk_len);
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(notif_len, 0, M_NOWAIT, 1, MT_DATA);
Support SCTP_REMOTE_ERROR notification. MFC after: 3 days
2012-05-13 22:27:54 +00:00
if (m_notify == NULL) {
/* Retry with smaller value. */
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
notif_len = (unsigned int)sizeof(struct sctp_remote_error);
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
m_notify = sctp_get_mbuf_for_msg(notif_len, 0, M_NOWAIT, 1, MT_DATA);
Support SCTP_REMOTE_ERROR notification. MFC after: 3 days
2012-05-13 22:27:54 +00:00
if (m_notify == NULL) {
return;
}
}
SCTP_BUF_NEXT(m_notify) = NULL;
sre = mtod(m_notify, struct sctp_remote_error *);
Initialize notification strucuture. This was missed in an earlier commit MFC after: 3 days
2014-07-24 18:06:18 +00:00
memset(sre, 0, notif_len);
Support SCTP_REMOTE_ERROR notification. MFC after: 3 days
2012-05-13 22:27:54 +00:00
sre->sre_type = SCTP_REMOTE_ERROR;
sre->sre_flags = 0;
sre->sre_length = sizeof(struct sctp_remote_error);
sre->sre_error = error;
sre->sre_assoc_id = sctp_get_associd(stcb);
if (notif_len > sizeof(struct sctp_remote_error)) {
memcpy(sre->sre_data, chunk, chunk_len);
sre->sre_length += chunk_len;
}
SCTP_BUF_LEN(m_notify) = sre->sre_length;
control = sctp_build_readq_entry(stcb, stcb->asoc.primary_destination,
0, 0, stcb->asoc.context, 0, 0, 0,
m_notify);
if (control != NULL) {
control->length = SCTP_BUF_LEN(m_notify);
/* not that we need this */
control->tail_mbuf = m_notify;
control->spec_flags = M_NOTIFICATION;
sctp_add_to_readq(stcb->sctp_ep, stcb,
control,
&stcb->sctp_socket->so_rcv, 1,
SCTP_READ_LOCK_NOT_HELD, SCTP_SO_NOT_LOCKED);
} else {
sctp_m_freem(m_notify);
}
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
void
sctp_ulp_notify(uint32_t notification, struct sctp_tcb *stcb,
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
uint32_t error, void *data, int so_locked
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
if ((stcb == NULL) ||
(stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE) ||
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
(stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_SOCKET_ALLGONE) ||
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
(stcb->asoc.state & SCTP_STATE_CLOSED_SOCKET)) {
/* If the socket is gone we are out of here */
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
}
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
if (stcb->sctp_socket->so_rcv.sb_state & SBS_CANTRCVMORE) {
return;
}
Remove redundant checks. MFC after: 2 weeks
2013-05-28 09:25:58 +00:00
if ((stcb->asoc.state & SCTP_STATE_COOKIE_WAIT) ||
(stcb->asoc.state & SCTP_STATE_COOKIE_ECHOED)) {
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
if ((notification == SCTP_NOTIFY_INTERFACE_DOWN) ||
(notification == SCTP_NOTIFY_INTERFACE_UP) ||
(notification == SCTP_NOTIFY_INTERFACE_CONFIRMED)) {
/* Don't report these in front states */
return;
}
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
switch (notification) {
case SCTP_NOTIFY_ASSOC_UP:
if (stcb->asoc.assoc_up_sent == 0) {
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
sctp_notify_assoc_change(SCTP_COMM_UP, stcb, error, NULL, 0, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
stcb->asoc.assoc_up_sent = 1;
}
- During shutdown pending, when the last sack came in and the last message on the send stream was "null" but still there, a state we allow, we could get hung and not clean it up and wait for the shutdown guard timer to clear the association without a graceful close. Fix this so that that we properly clean up. - Added support for Multiple ASCONF per new RFC. We only (so far) accept input of these and cannot yet generate a multi-asconf. - Sysctl'd support for experimental Fast Handover feature. Always disabled unless sysctl or socket option changes to enable. - Error case in add-ip where the peer supports AUTH and ADD-IP but does NOT require AUTH of ASCONF/ASCONF-ACK. We need to ABORT in this case. - According to the Kyoto summit of socket api developers (Solaris, Linux, BSD). We need to have: o non-eeor mode messages be atomic - Fixed o Allow implicit setup of an assoc in 1-2-1 model if using the sctp_**() send calls - Fixed o Get rid of HAVE_XXX declarations - Done o add a sctp_pr_policy in hole in sndrcvinfo structure - Done o add a PR_SCTP_POLICY_VALID type flag - yet to-do in a future patch! - Optimize sctp6 calls to reuse code in sctp_usrreq. Also optimize when we close sending out the data and disabling Nagle. - Change key concatenation order to match the auth RFC - When sending OOTB shutdown_complete always do csum. - Don't send PKT-DROP to a PKT-DROP - For abort chunks just always checksums same for shutdown-complete. - inpcb_free front state had a bug where in queue data could wedge an assoc. We need to just abandon ones in front states (free_assoc). - If a peer sends us a 64k abort, we would try to assemble a response packet which may be larger than 64k. This then would be dropped by IP. Instead make a "minimum" size for us 64k-2k (we want at least 2k for our initack). If we receive such an init discard it early without all the processing. - When we peel off we must increment the tcb ref count to keep it from being freed from underneath us. - handling fwd-tsn had bugs that caused memory overwrites when given faulty data, fixed so can't happen and we also stop at the first bad stream no. - Fixed so comm-up generates the adaption indication. - peeloff did not get the hmac params copied. - fix it so we lock the addr list when doing src-addr selection (in future we need to use a multi-reader/one writer lock here) - During lowlevel output, we could end up with a _l_addr set to null if the iterator is calling the output routine. This means we would possibly crash when we gather the MTU info. Fix so we only do the gather where we have a src address cached. - we need to be sure to set abort flag on conn state when we receive an abort. - peeloff could leak a socket. Moved code so the close will find the socket if the peeloff fails (uipc_syscalls.c) Approved by: re@freebsd.org(Ken Smith)
2007-08-27 05:19:48 +00:00
if (stcb->asoc.adaptation_needed && (stcb->asoc.adaptation_sent == 0)) {
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
sctp_notify_adaptation_layer(stcb);
- During shutdown pending, when the last sack came in and the last message on the send stream was "null" but still there, a state we allow, we could get hung and not clean it up and wait for the shutdown guard timer to clear the association without a graceful close. Fix this so that that we properly clean up. - Added support for Multiple ASCONF per new RFC. We only (so far) accept input of these and cannot yet generate a multi-asconf. - Sysctl'd support for experimental Fast Handover feature. Always disabled unless sysctl or socket option changes to enable. - Error case in add-ip where the peer supports AUTH and ADD-IP but does NOT require AUTH of ASCONF/ASCONF-ACK. We need to ABORT in this case. - According to the Kyoto summit of socket api developers (Solaris, Linux, BSD). We need to have: o non-eeor mode messages be atomic - Fixed o Allow implicit setup of an assoc in 1-2-1 model if using the sctp_**() send calls - Fixed o Get rid of HAVE_XXX declarations - Done o add a sctp_pr_policy in hole in sndrcvinfo structure - Done o add a PR_SCTP_POLICY_VALID type flag - yet to-do in a future patch! - Optimize sctp6 calls to reuse code in sctp_usrreq. Also optimize when we close sending out the data and disabling Nagle. - Change key concatenation order to match the auth RFC - When sending OOTB shutdown_complete always do csum. - Don't send PKT-DROP to a PKT-DROP - For abort chunks just always checksums same for shutdown-complete. - inpcb_free front state had a bug where in queue data could wedge an assoc. We need to just abandon ones in front states (free_assoc). - If a peer sends us a 64k abort, we would try to assemble a response packet which may be larger than 64k. This then would be dropped by IP. Instead make a "minimum" size for us 64k-2k (we want at least 2k for our initack). If we receive such an init discard it early without all the processing. - When we peel off we must increment the tcb ref count to keep it from being freed from underneath us. - handling fwd-tsn had bugs that caused memory overwrites when given faulty data, fixed so can't happen and we also stop at the first bad stream no. - Fixed so comm-up generates the adaption indication. - peeloff did not get the hmac params copied. - fix it so we lock the addr list when doing src-addr selection (in future we need to use a multi-reader/one writer lock here) - During lowlevel output, we could end up with a _l_addr set to null if the iterator is calling the output routine. This means we would possibly crash when we gather the MTU info. Fix so we only do the gather where we have a src address cached. - we need to be sure to set abort flag on conn state when we receive an abort. - peeloff could leak a socket. Moved code so the close will find the socket if the peeloff fails (uipc_syscalls.c) Approved by: re@freebsd.org(Ken Smith)
2007-08-27 05:19:48 +00:00
}
Add support for the SCTP_AUTH_SUPPORTED and SCTP_ASCONF_SUPPORTED socket options. Add also a sysctl to control the support of ASCONF. MFC after: 1 week
2014-08-12 11:30:16 +00:00
if (stcb->asoc.auth_supported == 0) {
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
sctp_ulp_notify(SCTP_NOTIFY_NO_PEER_AUTH, stcb, 0,
NULL, so_locked);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_NOTIFY_ASSOC_DOWN:
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
sctp_notify_assoc_change(SCTP_SHUTDOWN_COMP, stcb, error, NULL, 0, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_NOTIFY_INTERFACE_DOWN:
{
struct sctp_nets *net;
net = (struct sctp_nets *)data;
sctp_notify_peer_addr_change(stcb, SCTP_ADDR_UNREACHABLE,
Keep track on the socket lock state. This fixes a bug showing up on Mac OS X. MFC after: 3 days
2015-03-10 22:38:10 +00:00
(struct sockaddr *)&net->ro._l_addr, error, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
}
case SCTP_NOTIFY_INTERFACE_UP:
{
struct sctp_nets *net;
net = (struct sctp_nets *)data;
sctp_notify_peer_addr_change(stcb, SCTP_ADDR_AVAILABLE,
Keep track on the socket lock state. This fixes a bug showing up on Mac OS X. MFC after: 3 days
2015-03-10 22:38:10 +00:00
(struct sockaddr *)&net->ro._l_addr, error, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
}
case SCTP_NOTIFY_INTERFACE_CONFIRMED:
{
struct sctp_nets *net;
net = (struct sctp_nets *)data;
sctp_notify_peer_addr_change(stcb, SCTP_ADDR_CONFIRMED,
Keep track on the socket lock state. This fixes a bug showing up on Mac OS X. MFC after: 3 days
2015-03-10 22:38:10 +00:00
(struct sockaddr *)&net->ro._l_addr, error, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
}
case SCTP_NOTIFY_SPECIAL_SP_FAIL:
sctp_notify_send_failed2(stcb, error,
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
(struct sctp_stream_queue_pending *)data, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
case SCTP_NOTIFY_SENT_DG_FAIL:
sctp_notify_send_failed(stcb, 1, error,
(struct sctp_tmit_chunk *)data, so_locked);
break;
case SCTP_NOTIFY_UNSENT_DG_FAIL:
sctp_notify_send_failed(stcb, 0, error,
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
(struct sctp_tmit_chunk *)data, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_NOTIFY_PARTIAL_DELVIERY_INDICATION:
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
{
uint32_t val;
val = *((uint32_t *) data);
* Fix a bug where PR-SCTP settings are ignore when using implicit association setup. * Fix a bug where message with illegal stream ids are not deleted. * Fix a crash when reporting back unsent messages from the send_queue. * Fix a bug related to INIT retransmission when the socket is already closed. * Fix a bug where associations were stalled when partial delivery API was enabled. * Fix a bug where the receive buffer size was smaller than the partial_delivery_point. Approved by: re, rrs (mentor) MFC after: One day.
2009-08-15 21:10:52 +00:00
sctp_notify_partial_delivery_indication(stcb, error, val, so_locked);
break;
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
}
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
case SCTP_NOTIFY_ASSOC_LOC_ABORTED:
Remove redundant checks. MFC after: 2 weeks
2013-05-28 09:25:58 +00:00
if (((stcb->asoc.state & SCTP_STATE_MASK) == SCTP_STATE_COOKIE_WAIT) ||
((stcb->asoc.state & SCTP_STATE_MASK) == SCTP_STATE_COOKIE_ECHOED)) {
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
sctp_notify_assoc_change(SCTP_CANT_STR_ASSOC, stcb, error, data, 0, so_locked);
- fix source address selection when picking an acceptable address - name change of prefered -> preferred - CMT fast recover code added. - Comment fixes in CMT. - We were not giving a reason of cant_start_asoc per socket api if we failed to get init/or/cookie to bring up an assoc. Change so we don't just give a generic "comm lost" but look at actual states of dying assoc. - change "crc32" arguments to "crc32c" to silence strict/noisy compiler warnings when crc32() is also declared - A few minor tweaks to get the portable stuff truely portable for sctp6_usrreq.c :-D - one-2-one style vrf match problem. - window recovery would leave chks marked for retran during window probes on the sent queue. This would then cause an out-of-order problem and assure that the flight size "problem" would occur. - Solves a flight size logging issue that caused rwnd overruns, flight size off as well as false retransmissions.g - Macroize the up and down of flight size. - Fix a ECNE bug in its counting. - The strict_sacks options was causing aborts when window probing was active, fix to make strict sacks a bit smarter about what the next unsent TSN is. - Fixes a one-2-one wakeup bug found by Martin Kulas. - If-defed out form, Andre's copy routines pending his commit of at least m_last().. need to adjust for 6.2 as well.. since m_last won't exist. Reviewed by: gnn
2007-04-14 09:44:09 +00:00
} else {
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
sctp_notify_assoc_change(SCTP_COMM_LOST, stcb, error, data, 0, so_locked);
}
break;
case SCTP_NOTIFY_ASSOC_REM_ABORTED:
Remove redundant checks. MFC after: 2 weeks
2013-05-28 09:25:58 +00:00
if (((stcb->asoc.state & SCTP_STATE_MASK) == SCTP_STATE_COOKIE_WAIT) ||
((stcb->asoc.state & SCTP_STATE_MASK) == SCTP_STATE_COOKIE_ECHOED)) {
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
sctp_notify_assoc_change(SCTP_CANT_STR_ASSOC, stcb, error, data, 1, so_locked);
} else {
sctp_notify_assoc_change(SCTP_COMM_LOST, stcb, error, data, 1, so_locked);
- fix source address selection when picking an acceptable address - name change of prefered -> preferred - CMT fast recover code added. - Comment fixes in CMT. - We were not giving a reason of cant_start_asoc per socket api if we failed to get init/or/cookie to bring up an assoc. Change so we don't just give a generic "comm lost" but look at actual states of dying assoc. - change "crc32" arguments to "crc32c" to silence strict/noisy compiler warnings when crc32() is also declared - A few minor tweaks to get the portable stuff truely portable for sctp6_usrreq.c :-D - one-2-one style vrf match problem. - window recovery would leave chks marked for retran during window probes on the sent queue. This would then cause an out-of-order problem and assure that the flight size "problem" would occur. - Solves a flight size logging issue that caused rwnd overruns, flight size off as well as false retransmissions.g - Macroize the up and down of flight size. - Fix a ECNE bug in its counting. - The strict_sacks options was causing aborts when window probing was active, fix to make strict sacks a bit smarter about what the next unsent TSN is. - Fixes a one-2-one wakeup bug found by Martin Kulas. - If-defed out form, Andre's copy routines pending his commit of at least m_last().. need to adjust for 6.2 as well.. since m_last won't exist. Reviewed by: gnn
2007-04-14 09:44:09 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_NOTIFY_ASSOC_RESTART:
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
sctp_notify_assoc_change(SCTP_RESTART, stcb, error, NULL, 0, so_locked);
Add support for the SCTP_AUTH_SUPPORTED and SCTP_ASCONF_SUPPORTED socket options. Add also a sysctl to control the support of ASCONF. MFC after: 1 week
2014-08-12 11:30:16 +00:00
if (stcb->asoc.auth_supported == 0) {
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
sctp_ulp_notify(SCTP_NOTIFY_NO_PEER_AUTH, stcb, 0,
NULL, so_locked);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_NOTIFY_STR_RESET_SEND:
Use the flags defined in RFC 6525 in the stream reset event.
2012-04-23 20:45:31 +00:00
sctp_notify_stream_reset(stcb, error, ((uint16_t *) data), SCTP_STREAM_RESET_OUTGOING_SSN);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_NOTIFY_STR_RESET_RECV:
Use the flags defined in RFC 6525 in the stream reset event.
2012-04-23 20:45:31 +00:00
sctp_notify_stream_reset(stcb, error, ((uint16_t *) data), SCTP_STREAM_RESET_INCOMING);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_NOTIFY_STR_RESET_FAILED_OUT:
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
sctp_notify_stream_reset(stcb, error, ((uint16_t *) data),
Use the flags defined in RFC 6525 in the stream reset event.
2012-04-23 20:45:31 +00:00
(SCTP_STREAM_RESET_OUTGOING_SSN | SCTP_STREAM_RESET_FAILED));
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
Provide the flags in the SCTP stream reconfig related notification as specified in RFC 6525. MFC after: 3 days
2012-05-05 21:41:16 +00:00
case SCTP_NOTIFY_STR_RESET_DENIED_OUT:
sctp_notify_stream_reset(stcb, error, ((uint16_t *) data),
(SCTP_STREAM_RESET_OUTGOING_SSN | SCTP_STREAM_RESET_DENIED));
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
case SCTP_NOTIFY_STR_RESET_FAILED_IN:
Make stream our stream reset implementation compliant to RFC6525. MFC after: 1 month
2012-03-29 13:36:53 +00:00
sctp_notify_stream_reset(stcb, error, ((uint16_t *) data),
Use the flags defined in RFC 6525 in the stream reset event.
2012-04-23 20:45:31 +00:00
(SCTP_STREAM_RESET_INCOMING | SCTP_STREAM_RESET_FAILED));
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
Provide the flags in the SCTP stream reconfig related notification as specified in RFC 6525. MFC after: 3 days
2012-05-05 21:41:16 +00:00
case SCTP_NOTIFY_STR_RESET_DENIED_IN:
sctp_notify_stream_reset(stcb, error, ((uint16_t *) data),
(SCTP_STREAM_RESET_INCOMING | SCTP_STREAM_RESET_DENIED));
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
case SCTP_NOTIFY_ASCONF_ADD_IP:
sctp_notify_peer_addr_change(stcb, SCTP_ADDR_ADDED, data,
Keep track on the socket lock state. This fixes a bug showing up on Mac OS X. MFC after: 3 days
2015-03-10 22:38:10 +00:00
error, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_NOTIFY_ASCONF_DELETE_IP:
sctp_notify_peer_addr_change(stcb, SCTP_ADDR_REMOVED, data,
Keep track on the socket lock state. This fixes a bug showing up on Mac OS X. MFC after: 3 days
2015-03-10 22:38:10 +00:00
error, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_NOTIFY_ASCONF_SET_PRIMARY:
sctp_notify_peer_addr_change(stcb, SCTP_ADDR_MADE_PRIM, data,
Keep track on the socket lock state. This fixes a bug showing up on Mac OS X. MFC after: 3 days
2015-03-10 22:38:10 +00:00
error, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
case SCTP_NOTIFY_PEER_SHUTDOWN:
sctp_notify_shutdown_event(stcb);
break;
case SCTP_NOTIFY_AUTH_NEW_KEY:
Fix a type in an SCTP AUTH related notification. Keep the old name for backwards compatibility. Spotted by Irene Ruengeler. MFC after: 3 days
2012-04-26 11:07:15 +00:00
sctp_notify_authentication(stcb, SCTP_AUTH_NEW_KEY, error,
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
(uint16_t) (uintptr_t) data,
so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
case SCTP_NOTIFY_AUTH_FREE_KEY:
sctp_notify_authentication(stcb, SCTP_AUTH_FREE_KEY, error,
(uint16_t) (uintptr_t) data,
so_locked);
break;
case SCTP_NOTIFY_NO_PEER_AUTH:
sctp_notify_authentication(stcb, SCTP_AUTH_NO_AUTH, error,
(uint16_t) (uintptr_t) data,
so_locked);
break;
case SCTP_NOTIFY_SENDER_DRY:
sctp_notify_sender_dry_event(stcb, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
Support SCTP_REMOTE_ERROR notification. MFC after: 3 days
2012-05-13 22:27:54 +00:00
case SCTP_NOTIFY_REMOTE_ERROR:
sctp_notify_remote_error(stcb, error, data);
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
default:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTPDBG(SCTP_DEBUG_UTIL1, "%s: unknown notification %xh (%u)\n",
Use __func__ instead of __FUNCTION__. This allows to compile the userland stack without errors using gcc5. Thanks to saghul for makeing me aware and providing the patch. MFC after: 1 week
2015-10-19 11:17:54 +00:00
__func__, notification, notification);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
} /* end switch */
}
void
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
sctp_report_all_outbound(struct sctp_tcb *stcb, uint16_t error, int holds_lock, int so_locked
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
struct sctp_association *asoc;
struct sctp_stream_out *outs;
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
struct sctp_tmit_chunk *chk, *nchk;
struct sctp_stream_queue_pending *sp, *nsp;
In a true restart case, the send_lock was not being aquired. This meant that when we cleanup the outbound we may have one in transit to be added with the old sequence number. This is bad since then we loose a message :( Also the report_outbound needed to have the right lock when its called which it did not.. I added the lock with of course a flag since we want to have the lock before we call it in the restart case. This also fixed the FIX ME case where, in the cookie collision case, we mark for retransmit any that were bundled with the cookie that was dropped. This also means changes to the output routine so we can assure getting the COOKIE-ACK sent BEFORE we retransmit the Data. Approved by: gnn
2006-11-11 22:44:12 +00:00
int i;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (stcb == NULL) {
return;
}
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
asoc = &stcb->asoc;
if (asoc->state & SCTP_STATE_ABOUT_TO_BE_FREED) {
This fixes a crash in SCTP. It was possible to have a large number of packets queued to a crashing process. In a specific case you may get 2 ABORT's back (from say two packets in flight). If the aborts happened to be processed at the same time its possible to have one free the association while the other is trying to report all the outbound packets. When this occured it could lead to a crash. MFC after: 3 days
2010-07-03 14:03:31 +00:00
/* already being freed */
return;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if ((stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE) ||
(stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_SOCKET_ALLGONE) ||
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
(asoc->state & SCTP_STATE_CLOSED_SOCKET)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
}
/* now through all the gunk freeing chunks */
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (holds_lock == 0) {
In a true restart case, the send_lock was not being aquired. This meant that when we cleanup the outbound we may have one in transit to be added with the old sequence number. This is bad since then we loose a message :( Also the report_outbound needed to have the right lock when its called which it did not.. I added the lock with of course a flag since we want to have the lock before we call it in the restart case. This also fixed the FIX ME case where, in the cookie collision case, we mark for retransmit any that were bundled with the cookie that was dropped. This also means changes to the output routine so we can assure getting the COOKIE-ACK sent BEFORE we retransmit the Data. Approved by: gnn
2006-11-11 22:44:12 +00:00
SCTP_TCB_SEND_LOCK(stcb);
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
}
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
/* sent queue SHOULD be empty */
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
TAILQ_FOREACH_SAFE(chk, &asoc->sent_queue, sctp_next, nchk) {
TAILQ_REMOVE(&asoc->sent_queue, chk, sctp_next);
asoc->sent_queue_cnt--;
Get the accounting working. We now have counters how many chunks for each SCTP outgoing stream are in the send and sent queue. While there, improve the naming of NR-SACK related constants recently introduced. MFC after: 1 week
2012-11-16 19:39:10 +00:00
if (chk->sent != SCTP_DATAGRAM_NR_ACKED) {
Add per outgoing stream accounting for chunks in the send and sent queue. This provides no functional change, but is a preparation for an upcoming stream reset improvement. Done with rrs@. MFC after: 1 week
2012-11-07 22:11:38 +00:00
if (asoc->strmout[chk->rec.data.stream_number].chunks_on_queues > 0) {
asoc->strmout[chk->rec.data.stream_number].chunks_on_queues--;
#ifdef INVARIANTS
} else {
panic("No chunks on the queues for sid %u.", chk->rec.data.stream_number);
#endif
}
}
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
if (chk->data != NULL) {
sctp_free_bufspace(stcb, asoc, chk, 1);
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
sctp_ulp_notify(SCTP_NOTIFY_SENT_DG_FAIL, stcb,
error, chk, so_locked);
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
if (chk->data) {
sctp_m_freem(chk->data);
chk->data = NULL;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
Fix a locking issue showing up on Mac OS X when subscribing to authentication events. DTLS/SCTP renegotiations trigger the bug. MFC after: 2 weeks.
2011-05-08 09:11:59 +00:00
sctp_free_a_chunk(stcb, chk, so_locked);
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
/* sa_ignore FREED_MEMORY */
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
/* pending send queue SHOULD be empty */
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
TAILQ_FOREACH_SAFE(chk, &asoc->send_queue, sctp_next, nchk) {
TAILQ_REMOVE(&asoc->send_queue, chk, sctp_next);
asoc->send_queue_cnt--;
Add per outgoing stream accounting for chunks in the send and sent queue. This provides no functional change, but is a preparation for an upcoming stream reset improvement. Done with rrs@. MFC after: 1 week
2012-11-07 22:11:38 +00:00
if (asoc->strmout[chk->rec.data.stream_number].chunks_on_queues > 0) {
asoc->strmout[chk->rec.data.stream_number].chunks_on_queues--;
#ifdef INVARIANTS
} else {
panic("No chunks on the queues for sid %u.", chk->rec.data.stream_number);
#endif
}
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
if (chk->data != NULL) {
sctp_free_bufspace(stcb, asoc, chk, 1);
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
sctp_ulp_notify(SCTP_NOTIFY_UNSENT_DG_FAIL, stcb,
error, chk, so_locked);
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
if (chk->data) {
sctp_m_freem(chk->data);
chk->data = NULL;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
}
Fix a locking issue showing up on Mac OS X when subscribing to authentication events. DTLS/SCTP renegotiations trigger the bug. MFC after: 2 weeks.
2011-05-08 09:11:59 +00:00
sctp_free_a_chunk(stcb, chk, so_locked);
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
/* sa_ignore FREED_MEMORY */
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
}
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
for (i = 0; i < asoc->streamoutcnt; i++) {
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
/* For each stream */
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
outs = &asoc->strmout[i];
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
/* clean up any sends there */
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
asoc->locked_on_sending = NULL;
TAILQ_FOREACH_SAFE(sp, &outs->outqueue, next, nsp) {
asoc->stream_queue_cnt--;
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
TAILQ_REMOVE(&outs->outqueue, sp, next);
sctp_free_spbufspace(stcb, asoc, sp);
if (sp->data) {
This fixes a crash in SCTP. It was possible to have a large number of packets queued to a crashing process. In a specific case you may get 2 ABORT's back (from say two packets in flight). If the aborts happened to be processed at the same time its possible to have one free the association while the other is trying to report all the outbound packets. When this occured it could lead to a crash. MFC after: 3 days
2010-07-03 14:03:31 +00:00
sctp_ulp_notify(SCTP_NOTIFY_SPECIAL_SP_FAIL, stcb,
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
error, (void *)sp, so_locked);
This fixes a crash in SCTP. It was possible to have a large number of packets queued to a crashing process. In a specific case you may get 2 ABORT's back (from say two packets in flight). If the aborts happened to be processed at the same time its possible to have one free the association while the other is trying to report all the outbound packets. When this occured it could lead to a crash. MFC after: 3 days
2010-07-03 14:03:31 +00:00
if (sp->data) {
sctp_m_freem(sp->data);
sp->data = NULL;
Fix a refcount bug when freeing an association. While there: Change code to be consistent. Discussed with rrs@. MFC after: 3 days
2012-07-17 13:03:47 +00:00
sp->tail_mbuf = NULL;
sp->length = 0;
This fixes a crash in SCTP. It was possible to have a large number of packets queued to a crashing process. In a specific case you may get 2 ABORT's back (from say two packets in flight). If the aborts happened to be processed at the same time its possible to have one free the association while the other is trying to report all the outbound packets. When this occured it could lead to a crash. MFC after: 3 days
2010-07-03 14:03:31 +00:00
}
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
}
Delay the assignment of a path for DATA chunk until they hit the sent_queue. Honor a given path when the SCTP_ADDR_OVER flag is set. MFC after: 2 weeks.
2010-09-15 23:10:45 +00:00
if (sp->net) {
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
sctp_free_remote_addr(sp->net);
Delay the assignment of a path for DATA chunk until they hit the sent_queue. Honor a given path when the SCTP_ADDR_OVER flag is set. MFC after: 2 weeks.
2010-09-15 23:10:45 +00:00
sp->net = NULL;
}
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
/* Free the chunk */
Fix a locking issue showing up on Mac OS X when subscribing to authentication events. DTLS/SCTP renegotiations trigger the bug. MFC after: 2 weeks.
2011-05-08 09:11:59 +00:00
sctp_free_a_strmoq(stcb, sp, so_locked);
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
/* sa_ignore FREED_MEMORY */
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
- fix send_failed notification contents - Reorder send failed to be in correct order. - Fixed calulation of init-ack to be right off mbuf lengths instead of the precalculated value. This will fix one 64 bit platform issue.
2007-06-09 13:46:57 +00:00
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (holds_lock == 0) {
In a true restart case, the send_lock was not being aquired. This meant that when we cleanup the outbound we may have one in transit to be added with the old sequence number. This is bad since then we loose a message :( Also the report_outbound needed to have the right lock when its called which it did not.. I added the lock with of course a flag since we want to have the lock before we call it in the restart case. This also fixed the FIX ME case where, in the cookie collision case, we mark for retransmit any that were bundled with the cookie that was dropped. This also means changes to the output routine so we can assure getting the COOKIE-ACK sent BEFORE we retransmit the Data. Approved by: gnn
2006-11-11 22:44:12 +00:00
SCTP_TCB_SEND_UNLOCK(stcb);
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
sctp_abort_notification(struct sctp_tcb *stcb, uint8_t from_peer, uint16_t error,
Provide in the association change notification the received ABORT chunk if case of SCTP_COMM_LOST or SCTP_CANT_STR_ASSOC as required by RFC 6458. MFC after: 3 days
2012-05-12 20:11:35 +00:00
struct sctp_abort_chunk *abort, int so_locked
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (stcb == NULL) {
return;
}
Ensure that 1-to-1 style SCTP sockets can only be connected once. Allow implicit setup also for 1-to-1 style sockets as described in the latest version of the socket API ID. Approved by: re MFC after: 1 month
2011-09-14 19:10:13 +00:00
if ((stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_IN_TCPPOOL) ||
((stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_TCPTYPE) &&
(stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_CONNECTED))) {
stcb->sctp_ep->sctp_flags |= SCTP_PCB_FLAGS_WAS_ABORTED;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if ((stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE) ||
(stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_SOCKET_ALLGONE) ||
(stcb->asoc.state & SCTP_STATE_CLOSED_SOCKET)) {
return;
}
/* Tell them we lost the asoc */
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
sctp_report_all_outbound(stcb, error, 1, so_locked);
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
if (from_peer) {
sctp_ulp_notify(SCTP_NOTIFY_ASSOC_REM_ABORTED, stcb, error, abort, so_locked);
} else {
sctp_ulp_notify(SCTP_NOTIFY_ASSOC_LOC_ABORTED, stcb, error, abort, so_locked);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
sctp_abort_association(struct sctp_inpcb *inp, struct sctp_tcb *stcb,
Pass the src and dst address of a received packet explicitly around. MFC after: 3 days
2012-06-28 16:01:08 +00:00
struct mbuf *m, int iphlen,
struct sockaddr *src, struct sockaddr *dst,
struct sctphdr *sh, struct mbuf *op_err,
This is the SCTP specific companion of https://svnweb.freebsd.org/changeset/base/275358 which was provided by Hans Petter Selasky.
2014-12-04 21:17:50 +00:00
uint8_t mflowtype, uint32_t mflowid,
- Adds support for the multi-asconf (From Kozuka-san) - Adds some prepwork (Not all yet) for vimage in particular support the delete the sctppcbinfo.xx structs. There is still a leak in here if it were to be called plus we stil need the regrouping (From Me and Michael Tuexen) - Adds support for UDP tunneling. For BSD there is no socket yet setup so its disabled, but major argument changes are in here to emcompass the passing of the port number (zero when you don't have a udp tunnel, the default for BSD). Will add some hooks in UDP here shortly (discussed with Robert) that will allow easy tunneling. (Mainly from Peter Lei and Michael Tuexen with some BSD work from me :-D) - Some ease for windows, evidently leave is reserved by their compile move label leave: -> out: MFC after: 1 week
2008-05-20 13:47:46 +00:00
uint32_t vrf_id, uint16_t port)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
uint32_t vtag;
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
struct socket *so;
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
vtag = 0;
if (stcb != NULL) {
/* We have a TCB to abort, send notification too */
vtag = stcb->asoc.peer_vtag;
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
sctp_abort_notification(stcb, 0, 0, NULL, SCTP_SO_NOT_LOCKED);
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
/* get the assoc vrf id and table id */
vrf_id = stcb->asoc.vrf_id;
- change number assignments for SHA225-512 (match artisync for bakeoff.. using the next sequential ones) - In cookie processing 1-2-1, we did not increment the stcb refcnt before releasing the tcb lock. We need to do this to keep the tcb from being freed by a abort or ?? unlikely but worth doing. Also get rid of unneed INP_WLOCK. - extra receive info included the rcvinfo which killed the padding/alignment. We now redefine all the fields properly so they both align properly both to 128 bytes. - A peeled off socket would not close without an error due to its misguided idea that sctp_disconnect() was not supported on it. This fixes it so it goes through the proper path. - When an assoc was being deleted after abort (via a timer) a small race condition exists where we might take a packet for the old assoc (since we are waiting for a cleanup timer). This state especially happens in mac. We now add a state in the asoc so these can properly handle the packet as OOTB. Approved by: re@freebsd.org(Ken Smith)
2007-08-06 15:46:46 +00:00
stcb->asoc.state |= SCTP_STATE_WAS_ABORTED;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Pass the src and dst address of a received packet explicitly around. MFC after: 3 days
2012-06-28 16:01:08 +00:00
sctp_send_abort(m, iphlen, src, dst, sh, vtag, op_err,
Add FIB support for SCTP. This fixes https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=200379 MFC after: 3 days
2015-06-17 15:20:14 +00:00
mflowtype, mflowid, inp->fibnum,
Pass flowid explicitly through the stack instead of taking it from the mbuf chain at different places. While there: Fix several bugs related to VRFs. MFC after: 3 days
2012-06-14 06:54:48 +00:00
vrf_id, port);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb != NULL) {
/* Ok, now lets free it */
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
so = SCTP_INP_SO(inp);
atomic_add_int(&stcb->asoc.refcnt, 1);
SCTP_TCB_UNLOCK(stcb);
SCTP_SOCKET_LOCK(so, 1);
SCTP_TCB_LOCK(stcb);
atomic_subtract_int(&stcb->asoc.refcnt, 1);
#endif
Bugfix: Do correct accounting using the MIB counters when an association is aborted via sctp_abort_association(). MFC after: 3 days.
2010-12-12 20:50:44 +00:00
SCTP_STAT_INCR_COUNTER32(sctps_aborted);
if ((SCTP_GET_STATE(&stcb->asoc) == SCTP_STATE_OPEN) ||
(SCTP_GET_STATE(&stcb->asoc) == SCTP_STATE_SHUTDOWN_RECEIVED)) {
SCTP_STAT_DECR_GAUGE32(sctps_currestab);
}
Some more debug info cleanup. MFC after: 3 days
2015-05-28 16:39:22 +00:00
(void)sctp_free_assoc(inp, stcb, SCTP_NORMAL_PROC,
SCTP_FROM_SCTPUTIL + SCTP_LOC_4);
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
SCTP_SOCKET_UNLOCK(so, 1);
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
#ifdef SCTP_ASOCLOG_OF_TSNS
void
sctp_print_out_track_log(struct sctp_tcb *stcb)
{
- added pre-checks to the bindx call. - use proper tick gathering macro instead of ticks directly. - Placed reasonable boundaries on sets that a user can do that are converted to ticks from ms. - Fix CMT_PF to always check to be sure CMT is on. - Fix ticks use of CMT_PF. - put back code to allow asconfs to be queued while INITs are in flight and before the assoc is established. - During window probes, an ack'd packet might be left with the window probe mark on it causing it to be retransmitted. Change so that the flight decrease macro clears the window_probe mark. - Additional logging flight size/reading and ASOC LOG. This is only enabled if you manually insert things into opt_sctp.h since its a set of debug code only. - Found an interesting SMP race in the way data was appended which could cause a reader to lose a part of a message, had to reorder when we marked the message was complete to after the data was appended. - bug in ADD-IP for the subset bound socket case when the peer has only one address - fix ASCONF implicit success/error handling case - proper support of jails in Freebsd 6> - copy out the timeval for the 64 bit sparc world on cookie-echo alignment error crashes without this). Approved by: re(Ken Smith)
2007-07-17 20:58:26 +00:00
#ifdef NOSIY_PRINTS
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
int i;
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("Last ep reason:%x\n", stcb->sctp_ep->last_abort_code);
SCTP_PRINTF("IN bound TSN log-aaa\n");
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
if ((stcb->asoc.tsn_in_at == 0) && (stcb->asoc.tsn_in_wrapped == 0)) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("None rcvd\n");
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
goto none_in;
}
if (stcb->asoc.tsn_in_wrapped) {
for (i = stcb->asoc.tsn_in_at; i < SCTP_TSN_LOG_SIZE; i++) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("TSN:%x strm:%d seq:%d flags:%x sz:%d\n",
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
stcb->asoc.in_tsnlog[i].tsn,
stcb->asoc.in_tsnlog[i].strm,
stcb->asoc.in_tsnlog[i].seq,
stcb->asoc.in_tsnlog[i].flgs,
stcb->asoc.in_tsnlog[i].sz);
}
}
if (stcb->asoc.tsn_in_at) {
for (i = 0; i < stcb->asoc.tsn_in_at; i++) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("TSN:%x strm:%d seq:%d flags:%x sz:%d\n",
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
stcb->asoc.in_tsnlog[i].tsn,
stcb->asoc.in_tsnlog[i].strm,
stcb->asoc.in_tsnlog[i].seq,
stcb->asoc.in_tsnlog[i].flgs,
stcb->asoc.in_tsnlog[i].sz);
}
}
none_in:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("OUT bound TSN log-aaa\n");
if ((stcb->asoc.tsn_out_at == 0) &&
(stcb->asoc.tsn_out_wrapped == 0)) {
SCTP_PRINTF("None sent\n");
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
}
if (stcb->asoc.tsn_out_wrapped) {
for (i = stcb->asoc.tsn_out_at; i < SCTP_TSN_LOG_SIZE; i++) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("TSN:%x strm:%d seq:%d flags:%x sz:%d\n",
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
stcb->asoc.out_tsnlog[i].tsn,
stcb->asoc.out_tsnlog[i].strm,
stcb->asoc.out_tsnlog[i].seq,
stcb->asoc.out_tsnlog[i].flgs,
stcb->asoc.out_tsnlog[i].sz);
}
}
if (stcb->asoc.tsn_out_at) {
for (i = 0; i < stcb->asoc.tsn_out_at; i++) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("TSN:%x strm:%d seq:%d flags:%x sz:%d\n",
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
stcb->asoc.out_tsnlog[i].tsn,
stcb->asoc.out_tsnlog[i].strm,
stcb->asoc.out_tsnlog[i].seq,
stcb->asoc.out_tsnlog[i].flgs,
stcb->asoc.out_tsnlog[i].sz);
}
}
- added pre-checks to the bindx call. - use proper tick gathering macro instead of ticks directly. - Placed reasonable boundaries on sets that a user can do that are converted to ticks from ms. - Fix CMT_PF to always check to be sure CMT is on. - Fix ticks use of CMT_PF. - put back code to allow asconfs to be queued while INITs are in flight and before the assoc is established. - During window probes, an ack'd packet might be left with the window probe mark on it causing it to be retransmitted. Change so that the flight decrease macro clears the window_probe mark. - Additional logging flight size/reading and ASOC LOG. This is only enabled if you manually insert things into opt_sctp.h since its a set of debug code only. - Found an interesting SMP race in the way data was appended which could cause a reader to lose a part of a message, had to reorder when we marked the message was complete to after the data was appended. - bug in ADD-IP for the subset bound socket case when the peer has only one address - fix ASCONF implicit success/error handling case - proper support of jails in Freebsd 6> - copy out the timeval for the 64 bit sparc world on cookie-echo alignment error crashes without this). Approved by: re(Ken Smith)
2007-07-17 20:58:26 +00:00
#endif
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
}
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
void
sctp_abort_an_association(struct sctp_inpcb *inp, struct sctp_tcb *stcb,
Provide in the association change notification the received ABORT chunk if case of SCTP_COMM_LOST or SCTP_CANT_STR_ASSOC as required by RFC 6458. MFC after: 3 days
2012-05-12 20:11:35 +00:00
struct mbuf *op_err,
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
int so_locked
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
struct socket *so;
#endif
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
so = SCTP_INP_SO(inp);
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (stcb == NULL) {
/* Got to have a TCB */
if (inp->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE) {
Use LIST_EMPTY when appropriate. MFC after: 1 week
2013-06-02 10:35:08 +00:00
if (LIST_EMPTY(&inp->sctp_asoc_list)) {
- Get rid of the sctp_inpcb_free() "magic numbers", now they are sensible defines that tell what you are directing the function to do.
2007-05-08 15:53:03 +00:00
sctp_inpcb_free(inp, SCTP_FREE_SHOULD_USE_ABORT,
SCTP_CALLED_DIRECTLY_NOCMPSET);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
return;
- change number assignments for SHA225-512 (match artisync for bakeoff.. using the next sequential ones) - In cookie processing 1-2-1, we did not increment the stcb refcnt before releasing the tcb lock. We need to do this to keep the tcb from being freed by a abort or ?? unlikely but worth doing. Also get rid of unneed INP_WLOCK. - extra receive info included the rcvinfo which killed the padding/alignment. We now redefine all the fields properly so they both align properly both to 128 bytes. - A peeled off socket would not close without an error due to its misguided idea that sctp_disconnect() was not supported on it. This fixes it so it goes through the proper path. - When an assoc was being deleted after abort (via a timer) a small race condition exists where we might take a packet for the old assoc (since we are waiting for a cleanup timer). This state especially happens in mac. We now add a state in the asoc so these can properly handle the packet as OOTB. Approved by: re@freebsd.org(Ken Smith)
2007-08-06 15:46:46 +00:00
} else {
stcb->asoc.state |= SCTP_STATE_WAS_ABORTED;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
/* notify the ulp */
Provide in the association change notification the received ABORT chunk if case of SCTP_COMM_LOST or SCTP_CANT_STR_ASSOC as required by RFC 6458. MFC after: 3 days
2012-05-12 20:11:35 +00:00
if ((inp->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE) == 0) {
Use ECONNABORTED in cases where the ABORT was sent to the peer. MFC after: 3 days
2012-05-13 16:56:16 +00:00
sctp_abort_notification(stcb, 0, 0, NULL, so_locked);
Provide in the association change notification the received ABORT chunk if case of SCTP_COMM_LOST or SCTP_CANT_STR_ASSOC as required by RFC 6458. MFC after: 3 days
2012-05-12 20:11:35 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* notify the peer */
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
sctp_send_abort_tcb(stcb, op_err, so_locked);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_STAT_INCR_COUNTER32(sctps_aborted);
if ((SCTP_GET_STATE(&stcb->asoc) == SCTP_STATE_OPEN) ||
(SCTP_GET_STATE(&stcb->asoc) == SCTP_STATE_SHUTDOWN_RECEIVED)) {
SCTP_STAT_DECR_GAUGE32(sctps_currestab);
}
/* now free the asoc */
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
#ifdef SCTP_ASOCLOG_OF_TSNS
sctp_print_out_track_log(stcb);
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
#endif
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
if (!so_locked) {
atomic_add_int(&stcb->asoc.refcnt, 1);
SCTP_TCB_UNLOCK(stcb);
SCTP_SOCKET_LOCK(so, 1);
SCTP_TCB_LOCK(stcb);
atomic_subtract_int(&stcb->asoc.refcnt, 1);
}
- More work on making send lock contention. - Removed free-oqueue cache. - Fix counter for sq entries - Increased the amount of information retained on ASOC_TSN logging on the association. - Made it so with the ASOC_TSN logging on sending or recieving an abort we dump the log. - Went through and added invariant's around some panic's that needed them. - decrements went to atomic_subtact_int instead of add -1 - Removed residual count increment that threw off a strm oq count. - Tracks and complaints if we don't have a LAST fragment and clean up the sp structure. - Track a new stat that counts number of abandoned msgs that happen if you close without reading. - Fix lookup of frag point to be aware of a 0 assoc-id. Reviewed by: gnn
2007-04-19 11:28:43 +00:00
#endif
Some more debug info cleanup. MFC after: 3 days
2015-05-28 16:39:22 +00:00
(void)sctp_free_assoc(inp, stcb, SCTP_NORMAL_PROC,
SCTP_FROM_SCTPUTIL + SCTP_LOC_5);
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
if (!so_locked) {
SCTP_SOCKET_UNLOCK(so, 1);
}
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
void
Pass the src and dst address of a received packet explicitly around. MFC after: 3 days
2012-06-28 16:01:08 +00:00
sctp_handle_ootb(struct mbuf *m, int iphlen, int offset,
struct sockaddr *src, struct sockaddr *dst,
struct sctphdr *sh, struct sctp_inpcb *inp,
* Provide information in error causes in ASCII instead of proprietary binary format. * Add support for a diagnostic information error cause. The code is sysctlable and the default is 0, which means it is not sent. This is joint work with rrs@. MFC after: 1 week
2014-03-16 12:32:16 +00:00
struct mbuf *cause,
Add FIB support for SCTP. This fixes https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=200379 MFC after: 3 days
2015-06-17 15:20:14 +00:00
uint8_t mflowtype, uint32_t mflowid, uint16_t fibnum,
Pass flowid explicitly through the stack instead of taking it from the mbuf chain at different places. While there: Fix several bugs related to VRFs. MFC after: 3 days
2012-06-14 06:54:48 +00:00
uint32_t vrf_id, uint16_t port)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
struct sctp_chunkhdr *ch, chunk_buf;
unsigned int chk_length;
Add an SCTP sysctl "blackhole", similar to the one for TCP. If set to 1, no ABORT is sent back in response to an incoming INIT. If set to 2, no ABORT is sent back in response to an out of the blue packet. If set to 0 (the default), ABORTs are sent. Discussed with rrs@. MFC after: 1 month.
2012-01-08 09:56:24 +00:00
int contains_init_chunk;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_STAT_INCR_COUNTER32(sctps_outoftheblue);
/* Generate a TO address for future reference */
if (inp && (inp->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE)) {
Use LIST_EMPTY when appropriate. MFC after: 1 week
2013-06-02 10:35:08 +00:00
if (LIST_EMPTY(&inp->sctp_asoc_list)) {
- Get rid of the sctp_inpcb_free() "magic numbers", now they are sensible defines that tell what you are directing the function to do.
2007-05-08 15:53:03 +00:00
sctp_inpcb_free(inp, SCTP_FREE_SHOULD_USE_ABORT,
SCTP_CALLED_DIRECTLY_NOCMPSET);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
Add an SCTP sysctl "blackhole", similar to the one for TCP. If set to 1, no ABORT is sent back in response to an incoming INIT. If set to 2, no ABORT is sent back in response to an out of the blue packet. If set to 0 (the default), ABORTs are sent. Discussed with rrs@. MFC after: 1 month.
2012-01-08 09:56:24 +00:00
contains_init_chunk = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
ch = (struct sctp_chunkhdr *)sctp_m_getptr(m, offset,
sizeof(*ch), (uint8_t *) & chunk_buf);
while (ch != NULL) {
chk_length = ntohs(ch->chunk_length);
if (chk_length < sizeof(*ch)) {
/* break to abort land */
break;
}
switch (ch->chunk_type) {
Add an SCTP sysctl "blackhole", similar to the one for TCP. If set to 1, no ABORT is sent back in response to an incoming INIT. If set to 2, no ABORT is sent back in response to an out of the blue packet. If set to 0 (the default), ABORTs are sent. Discussed with rrs@. MFC after: 1 month.
2012-01-08 09:56:24 +00:00
case SCTP_INIT:
contains_init_chunk = 1;
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
case SCTP_PACKET_DROPPED:
/* we don't respond to pkt-dropped */
return;
case SCTP_ABORT_ASSOCIATION:
/* we don't respond with an ABORT to an ABORT */
return;
case SCTP_SHUTDOWN_COMPLETE:
/*
* we ignore it since we are not waiting for it and
* peer is gone
*/
return;
case SCTP_SHUTDOWN_ACK:
Pass the src and dst address of a received packet explicitly around. MFC after: 3 days
2012-06-28 16:01:08 +00:00
sctp_send_shutdown_complete2(src, dst, sh,
Add FIB support for SCTP. This fixes https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=200379 MFC after: 3 days
2015-06-17 15:20:14 +00:00
mflowtype, mflowid, fibnum,
Pass flowid explicitly through the stack instead of taking it from the mbuf chain at different places. While there: Fix several bugs related to VRFs. MFC after: 3 days
2012-06-14 06:54:48 +00:00
vrf_id, port);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
default:
break;
}
offset += SCTP_SIZE32(chk_length);
ch = (struct sctp_chunkhdr *)sctp_m_getptr(m, offset,
sizeof(*ch), (uint8_t *) & chunk_buf);
}
Add an SCTP sysctl "blackhole", similar to the one for TCP. If set to 1, no ABORT is sent back in response to an incoming INIT. If set to 2, no ABORT is sent back in response to an out of the blue packet. If set to 0 (the default), ABORTs are sent. Discussed with rrs@. MFC after: 1 month.
2012-01-08 09:56:24 +00:00
if ((SCTP_BASE_SYSCTL(sctp_blackhole) == 0) ||
((SCTP_BASE_SYSCTL(sctp_blackhole) == 1) &&
(contains_init_chunk == 0))) {
* Provide information in error causes in ASCII instead of proprietary binary format. * Add support for a diagnostic information error cause. The code is sysctlable and the default is 0, which means it is not sent. This is joint work with rrs@. MFC after: 1 week
2014-03-16 12:32:16 +00:00
sctp_send_abort(m, iphlen, src, dst, sh, 0, cause,
Add FIB support for SCTP. This fixes https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=200379 MFC after: 3 days
2015-06-17 15:20:14 +00:00
mflowtype, mflowid, fibnum,
Pass flowid explicitly through the stack instead of taking it from the mbuf chain at different places. While there: Fix several bugs related to VRFs. MFC after: 3 days
2012-06-14 06:54:48 +00:00
vrf_id, port);
Add an SCTP sysctl "blackhole", similar to the one for TCP. If set to 1, no ABORT is sent back in response to an incoming INIT. If set to 2, no ABORT is sent back in response to an out of the blue packet. If set to 0 (the default), ABORTs are sent. Discussed with rrs@. MFC after: 1 month.
2012-01-08 09:56:24 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
/*
* check the inbound datagram to make sure there is not an abort inside it,
* if there is return 1, else return 0.
*/
int
sctp_is_there_an_abort_here(struct mbuf *m, int iphlen, uint32_t * vtagfill)
{
struct sctp_chunkhdr *ch;
struct sctp_init_chunk *init_chk, chunk_buf;
int offset;
unsigned int chk_length;
offset = iphlen + sizeof(struct sctphdr);
ch = (struct sctp_chunkhdr *)sctp_m_getptr(m, offset, sizeof(*ch),
(uint8_t *) & chunk_buf);
while (ch != NULL) {
chk_length = ntohs(ch->chunk_length);
if (chk_length < sizeof(*ch)) {
/* packet is probably corrupt */
break;
}
/* we seem to be ok, is it an abort? */
if (ch->chunk_type == SCTP_ABORT_ASSOCIATION) {
/* yep, tell them */
return (1);
}
if (ch->chunk_type == SCTP_INITIATION) {
/* need to update the Vtag */
init_chk = (struct sctp_init_chunk *)sctp_m_getptr(m,
offset, sizeof(*init_chk), (uint8_t *) & chunk_buf);
if (init_chk != NULL) {
*vtagfill = ntohl(init_chk->init.initiate_tag);
}
}
/* Nope, move to the next chunk */
offset += SCTP_SIZE32(chk_length);
ch = (struct sctp_chunkhdr *)sctp_m_getptr(m, offset,
sizeof(*ch), (uint8_t *) & chunk_buf);
}
return (0);
}
/*
* currently (2/02), ifa_addr embeds scope_id's and don't have sin6_scope_id
* set (i.e. it's 0) so, create this function to compare link local scopes
*/
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#ifdef INET6
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
uint32_t
sctp_is_same_scope(struct sockaddr_in6 *addr1, struct sockaddr_in6 *addr2)
{
struct sockaddr_in6 a, b;
/* save copies */
a = *addr1;
b = *addr2;
if (a.sin6_scope_id == 0)
if (sa6_recoverscope(&a)) {
/* can't get scope, so can't match */
return (0);
}
if (b.sin6_scope_id == 0)
if (sa6_recoverscope(&b)) {
/* can't get scope, so can't match */
return (0);
}
if (a.sin6_scope_id != b.sin6_scope_id)
return (0);
return (1);
}
/*
* returns a sockaddr_in6 with embedded scope recovered and removed
*/
struct sockaddr_in6 *
sctp_recover_scope(struct sockaddr_in6 *addr, struct sockaddr_in6 *store)
{
/* check and strip embedded scope junk */
if (addr->sin6_family == AF_INET6) {
if (IN6_IS_SCOPE_LINKLOCAL(&addr->sin6_addr)) {
if (addr->sin6_scope_id == 0) {
*store = *addr;
if (!sa6_recoverscope(store)) {
/* use the recovered scope */
addr = store;
}
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
} else {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* else, return the original "to" addr */
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
in6_clearscope(&addr->sin6_addr);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
}
return (addr);
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* are the two addresses the same? currently a "scopeless" check returns: 1
* if same, 0 if not
*/
- Matthew's changes to get inlines out, plus a few of my own to deal with the VRF inline function -> becomes a macro now. Submitted by: Matthew Jacobs
2007-06-16 00:33:47 +00:00
int
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_cmpaddr(struct sockaddr *sa1, struct sockaddr *sa2)
{
/* must be valid */
if (sa1 == NULL || sa2 == NULL)
return (0);
/* must be the same family */
if (sa1->sa_family != sa2->sa_family)
return (0);
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
switch (sa1->sa_family) {
#ifdef INET6
case AF_INET6:
{
/* IPv6 addresses */
struct sockaddr_in6 *sin6_1, *sin6_2;
sin6_1 = (struct sockaddr_in6 *)sa1;
sin6_2 = (struct sockaddr_in6 *)sa2;
- Adds support for the multi-asconf (From Kozuka-san) - Adds some prepwork (Not all yet) for vimage in particular support the delete the sctppcbinfo.xx structs. There is still a leak in here if it were to be called plus we stil need the regrouping (From Me and Michael Tuexen) - Adds support for UDP tunneling. For BSD there is no socket yet setup so its disabled, but major argument changes are in here to emcompass the passing of the port number (zero when you don't have a udp tunnel, the default for BSD). Will add some hooks in UDP here shortly (discussed with Robert) that will allow easy tunneling. (Mainly from Peter Lei and Michael Tuexen with some BSD work from me :-D) - Some ease for windows, evidently leave is reserved by their compile move label leave: -> out: MFC after: 1 week
2008-05-20 13:47:46 +00:00
return (SCTP6_ARE_ADDR_EQUAL(sin6_1,
sin6_2));
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
}
#endif
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#ifdef INET
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
case AF_INET:
{
/* IPv4 addresses */
struct sockaddr_in *sin_1, *sin_2;
sin_1 = (struct sockaddr_in *)sa1;
sin_2 = (struct sockaddr_in *)sa2;
return (sin_1->sin_addr.s_addr == sin_2->sin_addr.s_addr);
}
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#endif
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
default:
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* we don't do these... */
return (0);
}
}
void
sctp_print_address(struct sockaddr *sa)
{
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#ifdef INET6
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
char ip6buf[INET6_ADDRSTRLEN];
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#endif
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
switch (sa->sa_family) {
#ifdef INET6
case AF_INET6:
{
struct sockaddr_in6 *sin6;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
sin6 = (struct sockaddr_in6 *)sa;
SCTP_PRINTF("IPv6 address: %s:port:%d scope:%u\n",
ip6_sprintf(ip6buf, &sin6->sin6_addr),
ntohs(sin6->sin6_port),
sin6->sin6_scope_id);
break;
}
#endif
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#ifdef INET
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
case AF_INET:
{
struct sockaddr_in *sin;
unsigned char *p;
sin = (struct sockaddr_in *)sa;
p = (unsigned char *)&sin->sin_addr;
SCTP_PRINTF("IPv4 address: %u.%u.%u.%u:%d\n",
p[0], p[1], p[2], p[3], ntohs(sin->sin_port));
break;
}
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#endif
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
default:
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("?\n");
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
void
sctp_pull_off_control_to_new_inp(struct sctp_inpcb *old_inp,
struct sctp_inpcb *new_inp,
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave (0-2). - Codenomicon security test updates - length checks and such. - Bug in stream reset (2 actually). - setpeerprimary could unlock a null pointer, fixed. - Added a flag in the pcb so netstat can see if we are listening easier. Obtained from: (some of the Listen changes from Weongyo Jeong)
2007-05-02 12:50:13 +00:00
struct sctp_tcb *stcb,
int waitflags)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
/*
* go through our old INP and pull off any control structures that
* belong to stcb and move then to the new inp.
*/
struct socket *old_so, *new_so;
struct sctp_queued_to_read *control, *nctl;
struct sctp_readhead tmp_queue;
struct mbuf *m;
- fixed several places where we did not release INP locks. - fixed a refcount bug in the new ifa structures. - use vrf's from default stcb or inp whenever possible. - Address limits raised to account for a full IP fragmented packet (1000 addresses). - flight size correcting updated to include one message only and to handle case where the peer does not cumack the next segment aka lists 1/1 in sack blocks.. - Various bad init/init-ack handling could cause a panic since we tried to unlock the destroyed mutex. Fixes so we properly exit when we need to destroy an assoc. (Found by Cisco DevTest team :D) - name rename in src-addr-selection from pass to sifa. - route structure typedef'd to allow different platforms and updated into sctp_os_bsd file. - Max retransmissions a chunk can be made added. Reviewed by: gnn
2007-04-03 11:15:32 +00:00
int error = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
old_so = old_inp->sctp_socket;
new_so = new_inp->sctp_socket;
TAILQ_INIT(&tmp_queue);
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave (0-2). - Codenomicon security test updates - length checks and such. - Bug in stream reset (2 actually). - setpeerprimary could unlock a null pointer, fixed. - Added a flag in the pcb so netstat can see if we are listening easier. Obtained from: (some of the Listen changes from Weongyo Jeong)
2007-05-02 12:50:13 +00:00
error = sblock(&old_so->so_rcv, waitflags);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (error) {
/*
* Gak, can't get sblock, we have a problem. data will be
* left stranded.. and we don't dare look at it since the
* other thread may be reading something. Oh well, its a
* screwed up app that does a peeloff OR a accept while
* reading from the main socket... actually its only the
* peeloff() case, since I think read will fail on a
* listening socket..
*/
return;
}
/* lock the socket buffers */
SCTP_INP_READ_LOCK(old_inp);
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
TAILQ_FOREACH_SAFE(control, &old_inp->read_queue, next, nctl) {
/* Pull off all for out target stcb */
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (control->stcb == stcb) {
/* remove it we want it */
TAILQ_REMOVE(&old_inp->read_queue, control, next);
TAILQ_INSERT_TAIL(&tmp_queue, control, next);
m = control->data;
while (m) {
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(&old_so->so_rcv, control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBFREE, SCTP_BUF_LEN(m));
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_sbfree(control, stcb, &old_so->so_rcv, m);
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(&old_so->so_rcv, control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBRESULT, 0);
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
m = SCTP_BUF_NEXT(m);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
}
SCTP_INP_READ_UNLOCK(old_inp);
/* Remove the sb-lock on the old socket */
sbunlock(&old_so->so_rcv);
/* Now we move them over to the new socket buffer */
SCTP_INP_READ_LOCK(new_inp);
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
TAILQ_FOREACH_SAFE(control, &tmp_queue, next, nctl) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
TAILQ_INSERT_TAIL(&new_inp->read_queue, control, next);
m = control->data;
while (m) {
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(&new_so->so_rcv, control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBALLOC, SCTP_BUF_LEN(m));
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_sballoc(stcb, &new_so->so_rcv, m);
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(&new_so->so_rcv, control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBRESULT, 0);
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
m = SCTP_BUF_NEXT(m);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
SCTP_INP_READ_UNLOCK(new_inp);
}
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
void
sctp_wakeup_the_read_socket(struct sctp_inpcb *inp)
{
if (inp && inp->sctp_socket) {
if (sctp_is_feature_on(inp, SCTP_PCB_FLAGS_ZERO_COPY_ACTIVE)) {
SCTP_ZERO_COPY_EVENT(inp, inp->sctp_socket);
} else {
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
struct socket *so;
so = SCTP_INP_SO(inp);
if (!so_locked) {
if (stcb) {
atomic_add_int(&stcb->asoc.refcnt, 1);
SCTP_TCB_UNLOCK(stcb);
}
SCTP_SOCKET_LOCK(so, 1);
if (stcb) {
SCTP_TCB_LOCK(stcb);
atomic_subtract_int(&stcb->asoc.refcnt, 1);
}
if (inp->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE) {
SCTP_SOCKET_UNLOCK(so, 1);
return;
}
}
#endif
sctp_sorwakeup(inp, inp->sctp_socket);
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
if (!so_locked) {
SCTP_SOCKET_UNLOCK(so, 1);
}
#endif
}
}
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
void
sctp_add_to_readq(struct sctp_inpcb *inp,
struct sctp_tcb *stcb,
struct sctp_queued_to_read *control,
struct sockbuf *sb,
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
int end,
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
int inp_read_lock_held,
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
int so_locked
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
/*
* Here we must place the control on the end of the socket read
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
* queue AND increment sb_cc so that select will work properly on
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
* read.
*/
struct mbuf *m, *prev = NULL;
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
if (inp == NULL) {
/* Gak, TSNH!! */
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
#ifdef INVARIANTS
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
panic("Gak, inp NULL on add_to_readq");
#endif
return;
}
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
if (inp_read_lock_held == 0)
SCTP_INP_READ_LOCK(inp);
Fix a rece condition in the shutdown handling. The race condition resulted in a panic. MFC after: 3 days
2010-06-18 09:01:44 +00:00
if (inp->sctp_flags & SCTP_PCB_FLAGS_SOCKET_CANT_READ) {
sctp_free_remote_addr(control->whoFrom);
if (control->data) {
sctp_m_freem(control->data);
control->data = NULL;
}
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
sctp_free_a_readq(stcb, control);
Fix a rece condition in the shutdown handling. The race condition resulted in a panic. MFC after: 3 days
2010-06-18 09:01:44 +00:00
if (inp_read_lock_held == 0)
SCTP_INP_READ_UNLOCK(inp);
return;
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
if (!(control->spec_flags & M_NOTIFICATION)) {
atomic_add_int(&inp->total_recvs, 1);
if (!control->do_not_ref_stcb) {
atomic_add_int(&stcb->total_recvs, 1);
}
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
m = control->data;
control->held_length = 0;
control->length = 0;
while (m) {
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (SCTP_BUF_LEN(m) == 0) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Skip mbufs with NO length */
if (prev == NULL) {
/* First one */
control->data = sctp_m_free(m);
m = control->data;
} else {
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_NEXT(prev) = sctp_m_free(m);
m = SCTP_BUF_NEXT(prev);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (m == NULL) {
Remove extraneous semicolons, no functional changes. Submitted by: Marc Balmer <marc@msys.ch> MFC after: 1 week
2010-01-07 21:01:37 +00:00
control->tail_mbuf = prev;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
continue;
}
prev = m;
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(sb, control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBALLOC, SCTP_BUF_LEN(m));
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_sballoc(stcb, sb, m);
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(sb, control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBRESULT, 0);
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
atomic_add_int(&control->length, SCTP_BUF_LEN(m));
m = SCTP_BUF_NEXT(m);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (prev != NULL) {
control->tail_mbuf = prev;
} else {
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
/* Everything got collapsed out?? */
Fix a rece condition in the shutdown handling. The race condition resulted in a panic. MFC after: 3 days
2010-06-18 09:01:44 +00:00
sctp_free_remote_addr(control->whoFrom);
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
sctp_free_a_readq(stcb, control);
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
if (inp_read_lock_held == 0)
SCTP_INP_READ_UNLOCK(inp);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
}
if (end) {
control->end_added = 1;
}
TAILQ_INSERT_TAIL(&inp->read_queue, control, next);
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
control->on_read_q = 1;
Turns out that when a receiver forwards through its TNS's the processing code holds the read lock (when processing a FWD-TSN for pr-sctp). If it finds stranded data that can be given to the application, it calls sctp_add_to_readq(). The readq function also grabs this lock. So if INVAR is on we get a double recurse on a non-recursive lock and panic. This fix will change it so that readq() function gets a flag to tell if the lock is held, if so then it does not get the lock. Approved by: re@freebsd.org (Kostik Belousov) MFC after: 1 week
2009-07-28 14:09:06 +00:00
if (inp_read_lock_held == 0)
SCTP_INP_READ_UNLOCK(inp);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (inp && inp->sctp_socket) {
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
sctp_wakeup_the_read_socket(inp);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
int
sctp_append_to_readq(struct sctp_inpcb *inp,
struct sctp_tcb *stcb,
struct sctp_queued_to_read *control,
struct mbuf *m,
int end,
int ctls_cumack,
struct sockbuf *sb)
{
/*
* A partial delivery API event is underway. OR we are appending on
* the reassembly queue.
*
* If PDAPI this means we need to add m to the end of the data.
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
* Increase the length in the control AND increment the sb_cc.
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
* Otherwise sb is NULL and all we need to do is put it at the end
* of the mbuf chain.
*/
int len = 0;
struct mbuf *mm, *tail = NULL, *prev = NULL;
if (inp) {
SCTP_INP_READ_LOCK(inp);
}
if (control == NULL) {
get_out:
if (inp) {
SCTP_INP_READ_UNLOCK(inp);
}
return (-1);
}
Fix a rece condition in the shutdown handling. The race condition resulted in a panic. MFC after: 3 days
2010-06-18 09:01:44 +00:00
if (inp && (inp->sctp_flags & SCTP_PCB_FLAGS_SOCKET_CANT_READ)) {
SCTP_INP_READ_UNLOCK(inp);
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
return (0);
Fix a rece condition in the shutdown handling. The race condition resulted in a panic. MFC after: 3 days
2010-06-18 09:01:44 +00:00
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (control->end_added) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* huh this one is complete? */
goto get_out;
}
mm = m;
if (mm == NULL) {
goto get_out;
}
while (mm) {
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (SCTP_BUF_LEN(mm) == 0) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Skip mbufs with NO lenght */
if (prev == NULL) {
/* First one */
m = sctp_m_free(mm);
mm = m;
} else {
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_NEXT(prev) = sctp_m_free(mm);
mm = SCTP_BUF_NEXT(prev);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
continue;
}
prev = mm;
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
len += SCTP_BUF_LEN(mm);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (sb) {
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(sb, control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBALLOC, SCTP_BUF_LEN(mm));
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_sballoc(stcb, sb, mm);
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(sb, control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBRESULT, 0);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
mm = SCTP_BUF_NEXT(mm);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
if (prev) {
tail = prev;
} else {
/* Really there should always be a prev */
if (m == NULL) {
/* Huh nothing left? */
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
#ifdef INVARIANTS
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
panic("Nothing left to add?");
#else
goto get_out;
#endif
}
tail = m;
}
if (control->tail_mbuf) {
/* append */
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_NEXT(control->tail_mbuf) = m;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
control->tail_mbuf = tail;
} else {
/* nothing there */
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
#ifdef INVARIANTS
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (control->data != NULL) {
panic("This should NOT happen");
}
#endif
control->data = m;
control->tail_mbuf = tail;
}
- added pre-checks to the bindx call. - use proper tick gathering macro instead of ticks directly. - Placed reasonable boundaries on sets that a user can do that are converted to ticks from ms. - Fix CMT_PF to always check to be sure CMT is on. - Fix ticks use of CMT_PF. - put back code to allow asconfs to be queued while INITs are in flight and before the assoc is established. - During window probes, an ack'd packet might be left with the window probe mark on it causing it to be retransmitted. Change so that the flight decrease macro clears the window_probe mark. - Additional logging flight size/reading and ASOC LOG. This is only enabled if you manually insert things into opt_sctp.h since its a set of debug code only. - Found an interesting SMP race in the way data was appended which could cause a reader to lose a part of a message, had to reorder when we marked the message was complete to after the data was appended. - bug in ADD-IP for the subset bound socket case when the peer has only one address - fix ASCONF implicit success/error handling case - proper support of jails in Freebsd 6> - copy out the timeval for the 64 bit sparc world on cookie-echo alignment error crashes without this). Approved by: re(Ken Smith)
2007-07-17 20:58:26 +00:00
atomic_add_int(&control->length, len);
if (end) {
/* message is complete */
if (stcb && (control == stcb->asoc.control_pdapi)) {
stcb->asoc.control_pdapi = NULL;
}
control->held_length = 0;
control->end_added = 1;
}
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (stcb == NULL) {
control->do_not_ref_stcb = 1;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* When we are appending in partial delivery, the cum-ack is used
* for the actual pd-api highest tsn on this mbuf. The true cum-ack
* is populated in the outbound sinfo structure from the true cumack
* if the association exists...
*/
control->sinfo_tsn = control->sinfo_cumtsn = ctls_cumack;
if (inp) {
SCTP_INP_READ_UNLOCK(inp);
}
if (inp && inp->sctp_socket) {
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
if (sctp_is_feature_on(inp, SCTP_PCB_FLAGS_ZERO_COPY_ACTIVE)) {
SCTP_ZERO_COPY_EVENT(inp, inp->sctp_socket);
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
} else {
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
struct socket *so;
so = SCTP_INP_SO(inp);
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
if (stcb) {
atomic_add_int(&stcb->asoc.refcnt, 1);
SCTP_TCB_UNLOCK(stcb);
}
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
SCTP_SOCKET_LOCK(so, 1);
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
if (stcb) {
SCTP_TCB_LOCK(stcb);
atomic_subtract_int(&stcb->asoc.refcnt, 1);
}
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
if (inp->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE) {
SCTP_SOCKET_UNLOCK(so, 1);
return (0);
}
#endif
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
sctp_sorwakeup(inp, inp->sctp_socket);
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
SCTP_SOCKET_UNLOCK(so, 1);
#endif
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
return (0);
}
/*************HOLD THIS COMMENT FOR PATCH FILE OF
*************ALTERNATE ROUTING CODE
*/
/*************HOLD THIS COMMENT FOR END OF PATCH FILE OF
*************ALTERNATE ROUTING CODE
*/
struct mbuf *
* Provide information in error causes in ASCII instead of proprietary binary format. * Add support for a diagnostic information error cause. The code is sysctlable and the default is 0, which means it is not sent. This is joint work with rrs@. MFC after: 1 week
2014-03-16 12:32:16 +00:00
sctp_generate_cause(uint16_t code, char *info)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
struct mbuf *m;
* Provide information in error causes in ASCII instead of proprietary binary format. * Add support for a diagnostic information error cause. The code is sysctlable and the default is 0, which means it is not sent. This is joint work with rrs@. MFC after: 1 week
2014-03-16 12:32:16 +00:00
struct sctp_gen_error_cause *cause;
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
size_t info_len;
uint16_t len;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
* Provide information in error causes in ASCII instead of proprietary binary format. * Add support for a diagnostic information error cause. The code is sysctlable and the default is 0, which means it is not sent. This is joint work with rrs@. MFC after: 1 week
2014-03-16 12:32:16 +00:00
if ((code == 0) || (info == NULL)) {
return (NULL);
}
info_len = strlen(info);
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
if (info_len > (SCTP_MAX_CAUSE_LENGTH - sizeof(struct sctp_paramhdr))) {
return (NULL);
}
len = (uint16_t) (sizeof(struct sctp_paramhdr) + info_len);
* Provide information in error causes in ASCII instead of proprietary binary format. * Add support for a diagnostic information error cause. The code is sysctlable and the default is 0, which means it is not sent. This is joint work with rrs@. MFC after: 1 week
2014-03-16 12:32:16 +00:00
m = sctp_get_mbuf_for_msg(len, 0, M_NOWAIT, 1, MT_DATA);
if (m != NULL) {
SCTP_BUF_LEN(m) = len;
cause = mtod(m, struct sctp_gen_error_cause *);
cause->code = htons(code);
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
cause->length = htons(len);
* Provide information in error causes in ASCII instead of proprietary binary format. * Add support for a diagnostic information error cause. The code is sysctlable and the default is 0, which means it is not sent. This is joint work with rrs@. MFC after: 1 week
2014-03-16 12:32:16 +00:00
memcpy(cause->info, info, info_len);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
return (m);
}
Send the correct error cause, when a DATA chunk with no user data is received. This bug was reported by Irene Ruengeler. MFC after: 3 days
2014-04-19 19:21:06 +00:00
struct mbuf *
sctp_generate_no_user_data_cause(uint32_t tsn)
{
struct mbuf *m;
struct sctp_error_no_user_data *no_user_data_cause;
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
uint16_t len;
Send the correct error cause, when a DATA chunk with no user data is received. This bug was reported by Irene Ruengeler. MFC after: 3 days
2014-04-19 19:21:06 +00:00
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
len = (uint16_t) sizeof(struct sctp_error_no_user_data);
Send the correct error cause, when a DATA chunk with no user data is received. This bug was reported by Irene Ruengeler. MFC after: 3 days
2014-04-19 19:21:06 +00:00
m = sctp_get_mbuf_for_msg(len, 0, M_NOWAIT, 1, MT_DATA);
if (m != NULL) {
SCTP_BUF_LEN(m) = len;
no_user_data_cause = mtod(m, struct sctp_error_no_user_data *);
no_user_data_cause->cause.code = htons(SCTP_CAUSE_NO_USER_DATA);
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
no_user_data_cause->cause.length = htons(len);
Send the correct error cause, when a DATA chunk with no user data is received. This bug was reported by Irene Ruengeler. MFC after: 3 days
2014-04-19 19:21:06 +00:00
no_user_data_cause->tsn = tsn; /* tsn is passed in as NBO */
}
return (m);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#ifdef SCTP_MBCNT_LOGGING
void
sctp_free_bufspace(struct sctp_tcb *stcb, struct sctp_association *asoc,
struct sctp_tmit_chunk *tp1, int chk_cnt)
{
if (tp1->data == NULL) {
return;
}
asoc->chunks_on_out_queue -= chk_cnt;
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_MBCNT_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_log_mbcnt(SCTP_LOG_MBCNT_DECREASE,
asoc->total_output_queue_size,
tp1->book_size,
0,
tp1->mbcnt);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (asoc->total_output_queue_size >= tp1->book_size) {
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
atomic_add_int(&asoc->total_output_queue_size, -tp1->book_size);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
asoc->total_output_queue_size = 0;
}
if (stcb->sctp_socket && (((stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_IN_TCPPOOL)) ||
((stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_TCPTYPE)))) {
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
if (stcb->sctp_socket->so_snd.sb_cc >= tp1->book_size) {
stcb->sctp_socket->so_snd.sb_cc -= tp1->book_size;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
stcb->sctp_socket->so_snd.sb_cc = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
}
#endif
int
sctp_release_pr_sctp_chunk(struct sctp_tcb *stcb, struct sctp_tmit_chunk *tp1,
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
uint8_t sent, int so_locked
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
#if !defined(__APPLE__) && !defined(SCTP_SO_LOCK_TESTING)
SCTP_UNUSED
#endif
)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
struct sctp_stream_out *strq;
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
struct sctp_tmit_chunk *chk = NULL, *tp2;
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
struct sctp_stream_queue_pending *sp;
uint16_t stream = 0, seq = 0;
uint8_t foundeom = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
int ret_sz = 0;
int notdone;
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
int do_wakeup_routine = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
stream = tp1->rec.data.stream_number;
seq = tp1->rec.data.stream_seq;
Add support for the SCTP_PR_STREAM_STATUS and SCTP_PR_ASSOC_STATUS socket options. This includes managing the correspoing stat counters. Add the SCTP_DETAILED_STR_STATS kernel option to control per policy counters on every stream. The default is off and only an aggregated counter is available. This is sufficient for the RTCWeb usecase. MFC after: 1 week
2014-08-13 15:50:16 +00:00
if (sent || !(tp1->rec.data.rcv_flags & SCTP_DATA_FIRST_FRAG)) {
stcb->asoc.abandoned_sent[0]++;
stcb->asoc.abandoned_sent[PR_SCTP_POLICY(tp1->flags)]++;
stcb->asoc.strmout[stream].abandoned_sent[0]++;
#if defined(SCTP_DETAILED_STR_STATS)
stcb->asoc.strmout[stream].abandoned_sent[PR_SCTP_POLICY(tp1->flags)]++;
#endif
} else {
stcb->asoc.abandoned_unsent[0]++;
stcb->asoc.abandoned_unsent[PR_SCTP_POLICY(tp1->flags)]++;
stcb->asoc.strmout[stream].abandoned_unsent[0]++;
#if defined(SCTP_DETAILED_STR_STATS)
stcb->asoc.strmout[stream].abandoned_unsent[PR_SCTP_POLICY(tp1->flags)]++;
#endif
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
do {
ret_sz += tp1->book_size;
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
if (tp1->data != NULL) {
Many bug fixes (from the IETF hack-fest): - PR-SCTP had major issues when skipping through a multi-part message. o Did not look at socket buffer. o Did not properly handle the reassmebly queue. o The MARKED segments could interfere and un-skip a chunk causing a problem with the proper FWD-TSN. o No FR of FWD-TSN's was being done. - NR-Sack code was basically disabled. It needed fixes that never got into the real code. - CMT code had issues when the two paths were NOT the same b/w. We found a few small bugs, but also the critcal one here was not dividing the rwnd amongst the paths. Obtained from: Michael Tuexen and myself at the IETF hack-fest ;-)
2009-04-04 11:43:32 +00:00
if (tp1->sent < SCTP_DATAGRAM_RESEND) {
sctp_flight_size_decrease(tp1);
sctp_total_flight_decrease(stcb, tp1);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_free_bufspace(stcb, &stcb->asoc, tp1, 1);
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
stcb->asoc.peers_rwnd += tp1->send_size;
stcb->asoc.peers_rwnd += SCTP_BASE_SYSCTL(sctp_peer_chunk_oh);
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
if (sent) {
sctp_ulp_notify(SCTP_NOTIFY_SENT_DG_FAIL, stcb, 0, tp1, so_locked);
} else {
sctp_ulp_notify(SCTP_NOTIFY_UNSENT_DG_FAIL, stcb, 0, tp1, so_locked);
}
Fix a bug in the handling of unreliable messages which results in stalled associations. Approved by: re, rrs (mentor) MFC after: immediately
2009-08-19 12:02:28 +00:00
if (tp1->data) {
sctp_m_freem(tp1->data);
tp1->data = NULL;
}
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
do_wakeup_routine = 1;
if (PR_SCTP_BUF_ENABLED(tp1->flags)) {
stcb->asoc.sent_queue_cnt_removeable--;
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
Many bug fixes (from the IETF hack-fest): - PR-SCTP had major issues when skipping through a multi-part message. o Did not look at socket buffer. o Did not properly handle the reassmebly queue. o The MARKED segments could interfere and un-skip a chunk causing a problem with the proper FWD-TSN. o No FR of FWD-TSN's was being done. - NR-Sack code was basically disabled. It needed fixes that never got into the real code. - CMT code had issues when the two paths were NOT the same b/w. We found a few small bugs, but also the critcal one here was not dividing the rwnd amongst the paths. Obtained from: Michael Tuexen and myself at the IETF hack-fest ;-)
2009-04-04 11:43:32 +00:00
tp1->sent = SCTP_FORWARD_TSN_SKIP;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if ((tp1->rec.data.rcv_flags & SCTP_DATA_NOT_FRAG) ==
SCTP_DATA_NOT_FRAG) {
/* not frag'ed we ae done */
notdone = 0;
foundeom = 1;
} else if (tp1->rec.data.rcv_flags & SCTP_DATA_LAST_FRAG) {
/* end of frag, we are done */
notdone = 0;
foundeom = 1;
} else {
/*
* Its a begin or middle piece, we must mark all of
* it
*/
notdone = 1;
tp1 = TAILQ_NEXT(tp1, sctp_next);
}
} while (tp1 && notdone);
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
if (foundeom == 0) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* The multi-part message was scattered across the send and
* sent queue.
*/
Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH, TAILQ_FOREACH_SAFE where appropriate. No functional change. MFC after: 3 months.
2010-12-30 16:56:20 +00:00
TAILQ_FOREACH_SAFE(tp1, &stcb->asoc.send_queue, sctp_next, tp2) {
if ((tp1->rec.data.stream_number != stream) ||
(tp1->rec.data.stream_seq != seq)) {
break;
}
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
/*
* save to chk in case we have some on stream out
* queue. If so and we have an un-transmitted one we
* don't have to fudge the TSN.
*/
chk = tp1;
ret_sz += tp1->book_size;
sctp_free_bufspace(stcb, &stcb->asoc, tp1, 1);
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications the correct ssf_error or ssfe_error as required by RFC 6458. MFC after: 3 days
2012-05-13 19:32:49 +00:00
if (sent) {
sctp_ulp_notify(SCTP_NOTIFY_SENT_DG_FAIL, stcb, 0, tp1, so_locked);
} else {
sctp_ulp_notify(SCTP_NOTIFY_UNSENT_DG_FAIL, stcb, 0, tp1, so_locked);
}
Fix a bug in the handling of unreliable messages which results in stalled associations. Approved by: re, rrs (mentor) MFC after: immediately
2009-08-19 12:02:28 +00:00
if (tp1->data) {
sctp_m_freem(tp1->data);
tp1->data = NULL;
}
Many bug fixes (from the IETF hack-fest): - PR-SCTP had major issues when skipping through a multi-part message. o Did not look at socket buffer. o Did not properly handle the reassmebly queue. o The MARKED segments could interfere and un-skip a chunk causing a problem with the proper FWD-TSN. o No FR of FWD-TSN's was being done. - NR-Sack code was basically disabled. It needed fixes that never got into the real code. - CMT code had issues when the two paths were NOT the same b/w. We found a few small bugs, but also the critcal one here was not dividing the rwnd amongst the paths. Obtained from: Michael Tuexen and myself at the IETF hack-fest ;-)
2009-04-04 11:43:32 +00:00
/* No flight involved here book the size to 0 */
tp1->book_size = 0;
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
if (tp1->rec.data.rcv_flags & SCTP_DATA_LAST_FRAG) {
foundeom = 1;
}
do_wakeup_routine = 1;
tp1->sent = SCTP_FORWARD_TSN_SKIP;
TAILQ_REMOVE(&stcb->asoc.send_queue, tp1, sctp_next);
/*
* on to the sent queue so we can wait for it to be
* passed by.
*/
TAILQ_INSERT_TAIL(&stcb->asoc.sent_queue, tp1,
sctp_next);
stcb->asoc.send_queue_cnt--;
stcb->asoc.sent_queue_cnt++;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
if (foundeom == 0) {
/*
* Still no eom found. That means there is stuff left on the
* stream out queue.. yuck.
*/
SCTP_TCB_SEND_LOCK(stcb);
Move from early SSN assignment to late SSN assignment. This doesn't change functionality, but makes upcoming change much easier. Developed with rrs@ at the IETF 85. MFC after: 1 week
2012-11-05 20:55:17 +00:00
strq = &stcb->asoc.strmout[stream];
sp = TAILQ_FIRST(&strq->outqueue);
if (sp != NULL) {
sp->discard_rest = 1;
/*
* We may need to put a chunk on the queue that
* holds the TSN that would have been sent with the
* LAST bit.
*/
if (chk == NULL) {
/* Yep, we have to */
sctp_alloc_a_chunk(stcb, chk);
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
if (chk == NULL) {
/*
Move from early SSN assignment to late SSN assignment. This doesn't change functionality, but makes upcoming change much easier. Developed with rrs@ at the IETF 85. MFC after: 1 week
2012-11-05 20:55:17 +00:00
* we are hosed. All we can do is
* nothing.. which will cause an
* abort if the peer is paying
* attention.
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
*/
Move from early SSN assignment to late SSN assignment. This doesn't change functionality, but makes upcoming change much easier. Developed with rrs@ at the IETF 85. MFC after: 1 week
2012-11-05 20:55:17 +00:00
goto oh_well;
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
}
Move from early SSN assignment to late SSN assignment. This doesn't change functionality, but makes upcoming change much easier. Developed with rrs@ at the IETF 85. MFC after: 1 week
2012-11-05 20:55:17 +00:00
memset(chk, 0, sizeof(*chk));
chk->rec.data.rcv_flags = SCTP_DATA_LAST_FRAG;
chk->sent = SCTP_FORWARD_TSN_SKIP;
chk->asoc = &stcb->asoc;
chk->rec.data.stream_seq = strq->next_sequence_send;
chk->rec.data.stream_number = sp->stream;
chk->rec.data.payloadtype = sp->ppid;
chk->rec.data.context = sp->context;
chk->flags = sp->act_flags;
Fix an accounting bug related to the per stream chunk counter. While there, don't refer to a net articifically. MFC after: 3 days
2015-03-24 14:51:46 +00:00
chk->whoTo = NULL;
Move from early SSN assignment to late SSN assignment. This doesn't change functionality, but makes upcoming change much easier. Developed with rrs@ at the IETF 85. MFC after: 1 week
2012-11-05 20:55:17 +00:00
chk->rec.data.TSN_seq = atomic_fetchadd_int(&stcb->asoc.sending_seq, 1);
Fix an accounting bug related to the per stream chunk counter. While there, don't refer to a net articifically. MFC after: 3 days
2015-03-24 14:51:46 +00:00
strq->chunks_on_queues++;
Move from early SSN assignment to late SSN assignment. This doesn't change functionality, but makes upcoming change much easier. Developed with rrs@ at the IETF 85. MFC after: 1 week
2012-11-05 20:55:17 +00:00
TAILQ_INSERT_TAIL(&stcb->asoc.sent_queue, chk, sctp_next);
stcb->asoc.sent_queue_cnt++;
stcb->asoc.pr_sctp_cnt++;
} else {
chk->rec.data.rcv_flags |= SCTP_DATA_LAST_FRAG;
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
}
Move from early SSN assignment to late SSN assignment. This doesn't change functionality, but makes upcoming change much easier. Developed with rrs@ at the IETF 85. MFC after: 1 week
2012-11-05 20:55:17 +00:00
strq->next_sequence_send++;
oh_well:
if (sp->data) {
/*
* Pull any data to free up the SB and allow
* sender to "add more" while we will throw
* away :-)
*/
sctp_free_spbufspace(stcb, &stcb->asoc, sp);
ret_sz += sp->length;
do_wakeup_routine = 1;
sp->some_taken = 1;
sctp_m_freem(sp->data);
sp->data = NULL;
sp->tail_mbuf = NULL;
sp->length = 0;
}
}
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
SCTP_TCB_SEND_UNLOCK(stcb);
}
if (do_wakeup_routine) {
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
Many bug fixes (from the IETF hack-fest): - PR-SCTP had major issues when skipping through a multi-part message. o Did not look at socket buffer. o Did not properly handle the reassmebly queue. o The MARKED segments could interfere and un-skip a chunk causing a problem with the proper FWD-TSN. o No FR of FWD-TSN's was being done. - NR-Sack code was basically disabled. It needed fixes that never got into the real code. - CMT code had issues when the two paths were NOT the same b/w. We found a few small bugs, but also the critcal one here was not dividing the rwnd amongst the paths. Obtained from: Michael Tuexen and myself at the IETF hack-fest ;-)
2009-04-04 11:43:32 +00:00
struct socket *so;
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
so = SCTP_INP_SO(stcb->sctp_ep);
if (!so_locked) {
atomic_add_int(&stcb->asoc.refcnt, 1);
SCTP_TCB_UNLOCK(stcb);
SCTP_SOCKET_LOCK(so, 1);
SCTP_TCB_LOCK(stcb);
atomic_subtract_int(&stcb->asoc.refcnt, 1);
if (stcb->asoc.state & SCTP_STATE_CLOSED_SOCKET) {
/* assoc was freed while we were unlocked */
SCTP_SOCKET_UNLOCK(so, 1);
return (ret_sz);
}
}
#endif
sctp_sowwakeup(stcb->sctp_ep, stcb->sctp_socket);
Whitespace cleanup. MFC after: 3 days
2012-06-25 17:15:09 +00:00
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
Fixes several PR-SCTP releated bugs. - When sending large PR-SCTP messages over a lossy link we would incorrectly calculate the fwd-tsn - When receiving large multipart pr-sctp packets we would incorrectly send back a SACK that would renege improperly on already received packets thus causing unneeded retransmissions.
2009-03-14 13:42:13 +00:00
if (!so_locked) {
SCTP_SOCKET_UNLOCK(so, 1);
}
#endif
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (ret_sz);
}
/*
* checks to see if the given address, sa, is one that is currently known by
* the kernel note: can't distinguish the same address on multiple interfaces
* and doesn't handle multiple addresses with different zone/scope id's note:
* ifa_ifwithaddr() compares the entire sockaddr struct
*/
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
struct sctp_ifa *
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_find_ifa_in_ep(struct sctp_inpcb *inp, struct sockaddr *addr,
int holds_lock)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
struct sctp_laddr *laddr;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (holds_lock == 0) {
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_INP_RLOCK(inp);
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
LIST_FOREACH(laddr, &inp->sctp_addr_list, sctp_nxt_addr) {
if (laddr->ifa == NULL)
continue;
if (addr->sa_family != laddr->ifa->address.sa.sa_family)
continue;
Improve compilation of SCTP code without INET support. Some bugs where fixed while doing this: * ASCONF-ACK messages might use wrong port number when using IPv6. * Checking for additional addresses takes the correct address into account and also does not do more comparisons than necessary. This patch is based on one received from bz@ who was sponsored by The FreeBSD Foundation and iXsystems. MFC after: 1 week
2011-04-30 11:18:16 +00:00
#ifdef INET
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
if (addr->sa_family == AF_INET) {
if (((struct sockaddr_in *)addr)->sin_addr.s_addr ==
laddr->ifa->address.sin.sin_addr.s_addr) {
/* found him. */
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (holds_lock == 0) {
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_INP_RUNLOCK(inp);
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
return (laddr->ifa);
break;
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
}
Improve compilation of SCTP code without INET support. Some bugs where fixed while doing this: * ASCONF-ACK messages might use wrong port number when using IPv6. * Checking for additional addresses takes the correct address into account and also does not do more comparisons than necessary. This patch is based on one received from bz@ who was sponsored by The FreeBSD Foundation and iXsystems. MFC after: 1 week
2011-04-30 11:18:16 +00:00
#endif
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#ifdef INET6
if (addr->sa_family == AF_INET6) {
- Adds support for the multi-asconf (From Kozuka-san) - Adds some prepwork (Not all yet) for vimage in particular support the delete the sctppcbinfo.xx structs. There is still a leak in here if it were to be called plus we stil need the regrouping (From Me and Michael Tuexen) - Adds support for UDP tunneling. For BSD there is no socket yet setup so its disabled, but major argument changes are in here to emcompass the passing of the port number (zero when you don't have a udp tunnel, the default for BSD). Will add some hooks in UDP here shortly (discussed with Robert) that will allow easy tunneling. (Mainly from Peter Lei and Michael Tuexen with some BSD work from me :-D) - Some ease for windows, evidently leave is reserved by their compile move label leave: -> out: MFC after: 1 week
2008-05-20 13:47:46 +00:00
if (SCTP6_ARE_ADDR_EQUAL((struct sockaddr_in6 *)addr,
&laddr->ifa->address.sin6)) {
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
/* found him. */
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (holds_lock == 0) {
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_INP_RUNLOCK(inp);
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
return (laddr->ifa);
break;
}
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#endif
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
}
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
if (holds_lock == 0) {
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_INP_RUNLOCK(inp);
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
return (NULL);
}
Adds a hash table to speed local address lookup on a per VRF basis (BSD has only one VRF currently). Hash table is sized to 16 but may need to be adjusted for machines with large numbers of addresses. Reviewed by: gnn
2007-03-19 11:11:16 +00:00
uint32_t
sctp_get_ifa_hash_val(struct sockaddr *addr)
{
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
switch (addr->sa_family) {
#ifdef INET
case AF_INET:
{
struct sockaddr_in *sin;
Adds a hash table to speed local address lookup on a per VRF basis (BSD has only one VRF currently). Hash table is sized to 16 but may need to be adjusted for machines with large numbers of addresses. Reviewed by: gnn
2007-03-19 11:11:16 +00:00
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
sin = (struct sockaddr_in *)addr;
return (sin->sin_addr.s_addr ^ (sin->sin_addr.s_addr >> 16));
}
#endif
#ifdef INET6
Fix a copy and paste error. MFC after: 3 days
2012-12-27 08:02:58 +00:00
case AF_INET6:
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
{
struct sockaddr_in6 *sin6;
uint32_t hash_of_addr;
Adds a hash table to speed local address lookup on a per VRF basis (BSD has only one VRF currently). Hash table is sized to 16 but may need to be adjusted for machines with large numbers of addresses. Reviewed by: gnn
2007-03-19 11:11:16 +00:00
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
sin6 = (struct sockaddr_in6 *)addr;
hash_of_addr = (sin6->sin6_addr.s6_addr32[0] +
sin6->sin6_addr.s6_addr32[1] +
sin6->sin6_addr.s6_addr32[2] +
sin6->sin6_addr.s6_addr32[3]);
hash_of_addr = (hash_of_addr ^ (hash_of_addr >> 16));
return (hash_of_addr);
}
#endif
default:
break;
Adds a hash table to speed local address lookup on a per VRF basis (BSD has only one VRF currently). Hash table is sized to 16 but may need to be adjusted for machines with large numbers of addresses. Reviewed by: gnn
2007-03-19 11:11:16 +00:00
}
return (0);
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
struct sctp_ifa *
sctp_find_ifa_by_addr(struct sockaddr *addr, uint32_t vrf_id, int holds_lock)
{
struct sctp_ifa *sctp_ifap;
struct sctp_vrf *vrf;
Adds a hash table to speed local address lookup on a per VRF basis (BSD has only one VRF currently). Hash table is sized to 16 but may need to be adjusted for machines with large numbers of addresses. Reviewed by: gnn
2007-03-19 11:11:16 +00:00
struct sctp_ifalist *hash_head;
uint32_t hash_of_addr;
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
- fixed several places where we did not release INP locks. - fixed a refcount bug in the new ifa structures. - use vrf's from default stcb or inp whenever possible. - Address limits raised to account for a full IP fragmented packet (1000 addresses). - flight size correcting updated to include one message only and to handle case where the peer does not cumack the next segment aka lists 1/1 in sack blocks.. - Various bad init/init-ack handling could cause a panic since we tried to unlock the destroyed mutex. Fixes so we properly exit when we need to destroy an assoc. (Found by Cisco DevTest team :D) - name rename in src-addr-selection from pass to sifa. - route structure typedef'd to allow different platforms and updated into sctp_os_bsd file. - Max retransmissions a chunk can be made added. Reviewed by: gnn
2007-04-03 11:15:32 +00:00
if (holds_lock == 0)
- The address lock is changed to a rwlock. This also involves macro changes to have a RLOCK and a WLOCK and placing the correct version within the code. - The INP-INFO lock is changed to a rwlock. - When sctp_shutdown() is called on Mac OS X, the socket lock is held. So call sctp_chunk_output with SCTP_SO_LOCKED and not SCTP_SO_NOT_LOCKED. - Add SCTP_IPI_ADDR_[RW]LOCK and SCTP_IPI_ADDR_[RW]UNLOCK for Mac OS X. - u_int64_t -> uint64_t - add missing addr unlock for error return path Approved by: re@freebsd.org (K Smith)
2007-09-18 15:16:39 +00:00
SCTP_IPI_ADDR_RLOCK();
- fixed several places where we did not release INP locks. - fixed a refcount bug in the new ifa structures. - use vrf's from default stcb or inp whenever possible. - Address limits raised to account for a full IP fragmented packet (1000 addresses). - flight size correcting updated to include one message only and to handle case where the peer does not cumack the next segment aka lists 1/1 in sack blocks.. - Various bad init/init-ack handling could cause a panic since we tried to unlock the destroyed mutex. Fixes so we properly exit when we need to destroy an assoc. (Found by Cisco DevTest team :D) - name rename in src-addr-selection from pass to sifa. - route structure typedef'd to allow different platforms and updated into sctp_os_bsd file. - Max retransmissions a chunk can be made added. Reviewed by: gnn
2007-04-03 11:15:32 +00:00
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
vrf = sctp_find_vrf(vrf_id);
- fixed several places where we did not release INP locks. - fixed a refcount bug in the new ifa structures. - use vrf's from default stcb or inp whenever possible. - Address limits raised to account for a full IP fragmented packet (1000 addresses). - flight size correcting updated to include one message only and to handle case where the peer does not cumack the next segment aka lists 1/1 in sack blocks.. - Various bad init/init-ack handling could cause a panic since we tried to unlock the destroyed mutex. Fixes so we properly exit when we need to destroy an assoc. (Found by Cisco DevTest team :D) - name rename in src-addr-selection from pass to sifa. - route structure typedef'd to allow different platforms and updated into sctp_os_bsd file. - Max retransmissions a chunk can be made added. Reviewed by: gnn
2007-04-03 11:15:32 +00:00
if (vrf == NULL) {
if (holds_lock == 0)
- The address lock is changed to a rwlock. This also involves macro changes to have a RLOCK and a WLOCK and placing the correct version within the code. - The INP-INFO lock is changed to a rwlock. - When sctp_shutdown() is called on Mac OS X, the socket lock is held. So call sctp_chunk_output with SCTP_SO_LOCKED and not SCTP_SO_NOT_LOCKED. - Add SCTP_IPI_ADDR_[RW]LOCK and SCTP_IPI_ADDR_[RW]UNLOCK for Mac OS X. - u_int64_t -> uint64_t - add missing addr unlock for error return path Approved by: re@freebsd.org (K Smith)
2007-09-18 15:16:39 +00:00
SCTP_IPI_ADDR_RUNLOCK();
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
return (NULL);
- fixed several places where we did not release INP locks. - fixed a refcount bug in the new ifa structures. - use vrf's from default stcb or inp whenever possible. - Address limits raised to account for a full IP fragmented packet (1000 addresses). - flight size correcting updated to include one message only and to handle case where the peer does not cumack the next segment aka lists 1/1 in sack blocks.. - Various bad init/init-ack handling could cause a panic since we tried to unlock the destroyed mutex. Fixes so we properly exit when we need to destroy an assoc. (Found by Cisco DevTest team :D) - name rename in src-addr-selection from pass to sifa. - route structure typedef'd to allow different platforms and updated into sctp_os_bsd file. - Max retransmissions a chunk can be made added. Reviewed by: gnn
2007-04-03 11:15:32 +00:00
}
Adds a hash table to speed local address lookup on a per VRF basis (BSD has only one VRF currently). Hash table is sized to 16 but may need to be adjusted for machines with large numbers of addresses. Reviewed by: gnn
2007-03-19 11:11:16 +00:00
hash_of_addr = sctp_get_ifa_hash_val(addr);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
hash_head = &vrf->vrf_addr_hash[(hash_of_addr & vrf->vrf_addr_hashmark)];
- fixed several places where we did not release INP locks. - fixed a refcount bug in the new ifa structures. - use vrf's from default stcb or inp whenever possible. - Address limits raised to account for a full IP fragmented packet (1000 addresses). - flight size correcting updated to include one message only and to handle case where the peer does not cumack the next segment aka lists 1/1 in sack blocks.. - Various bad init/init-ack handling could cause a panic since we tried to unlock the destroyed mutex. Fixes so we properly exit when we need to destroy an assoc. (Found by Cisco DevTest team :D) - name rename in src-addr-selection from pass to sifa. - route structure typedef'd to allow different platforms and updated into sctp_os_bsd file. - Max retransmissions a chunk can be made added. Reviewed by: gnn
2007-04-03 11:15:32 +00:00
if (hash_head == NULL) {
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("hash_of_addr:%x mask:%x table:%x - ",
- The address lock is changed to a rwlock. This also involves macro changes to have a RLOCK and a WLOCK and placing the correct version within the code. - The INP-INFO lock is changed to a rwlock. - When sctp_shutdown() is called on Mac OS X, the socket lock is held. So call sctp_chunk_output with SCTP_SO_LOCKED and not SCTP_SO_NOT_LOCKED. - Add SCTP_IPI_ADDR_[RW]LOCK and SCTP_IPI_ADDR_[RW]UNLOCK for Mac OS X. - u_int64_t -> uint64_t - add missing addr unlock for error return path Approved by: re@freebsd.org (K Smith)
2007-09-18 15:16:39 +00:00
hash_of_addr, (uint32_t) vrf->vrf_addr_hashmark,
(uint32_t) (hash_of_addr & vrf->vrf_addr_hashmark));
- fixed several places where we did not release INP locks. - fixed a refcount bug in the new ifa structures. - use vrf's from default stcb or inp whenever possible. - Address limits raised to account for a full IP fragmented packet (1000 addresses). - flight size correcting updated to include one message only and to handle case where the peer does not cumack the next segment aka lists 1/1 in sack blocks.. - Various bad init/init-ack handling could cause a panic since we tried to unlock the destroyed mutex. Fixes so we properly exit when we need to destroy an assoc. (Found by Cisco DevTest team :D) - name rename in src-addr-selection from pass to sifa. - route structure typedef'd to allow different platforms and updated into sctp_os_bsd file. - Max retransmissions a chunk can be made added. Reviewed by: gnn
2007-04-03 11:15:32 +00:00
sctp_print_address(addr);
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("No such bucket for address\n");
- fixed several places where we did not release INP locks. - fixed a refcount bug in the new ifa structures. - use vrf's from default stcb or inp whenever possible. - Address limits raised to account for a full IP fragmented packet (1000 addresses). - flight size correcting updated to include one message only and to handle case where the peer does not cumack the next segment aka lists 1/1 in sack blocks.. - Various bad init/init-ack handling could cause a panic since we tried to unlock the destroyed mutex. Fixes so we properly exit when we need to destroy an assoc. (Found by Cisco DevTest team :D) - name rename in src-addr-selection from pass to sifa. - route structure typedef'd to allow different platforms and updated into sctp_os_bsd file. - Max retransmissions a chunk can be made added. Reviewed by: gnn
2007-04-03 11:15:32 +00:00
if (holds_lock == 0)
- The address lock is changed to a rwlock. This also involves macro changes to have a RLOCK and a WLOCK and placing the correct version within the code. - The INP-INFO lock is changed to a rwlock. - When sctp_shutdown() is called on Mac OS X, the socket lock is held. So call sctp_chunk_output with SCTP_SO_LOCKED and not SCTP_SO_NOT_LOCKED. - Add SCTP_IPI_ADDR_[RW]LOCK and SCTP_IPI_ADDR_[RW]UNLOCK for Mac OS X. - u_int64_t -> uint64_t - add missing addr unlock for error return path Approved by: re@freebsd.org (K Smith)
2007-09-18 15:16:39 +00:00
SCTP_IPI_ADDR_RUNLOCK();
- fixed several places where we did not release INP locks. - fixed a refcount bug in the new ifa structures. - use vrf's from default stcb or inp whenever possible. - Address limits raised to account for a full IP fragmented packet (1000 addresses). - flight size correcting updated to include one message only and to handle case where the peer does not cumack the next segment aka lists 1/1 in sack blocks.. - Various bad init/init-ack handling could cause a panic since we tried to unlock the destroyed mutex. Fixes so we properly exit when we need to destroy an assoc. (Found by Cisco DevTest team :D) - name rename in src-addr-selection from pass to sifa. - route structure typedef'd to allow different platforms and updated into sctp_os_bsd file. - Max retransmissions a chunk can be made added. Reviewed by: gnn
2007-04-03 11:15:32 +00:00
return (NULL);
}
Adds a hash table to speed local address lookup on a per VRF basis (BSD has only one VRF currently). Hash table is sized to 16 but may need to be adjusted for machines with large numbers of addresses. Reviewed by: gnn
2007-03-19 11:11:16 +00:00
LIST_FOREACH(sctp_ifap, hash_head, next_bucket) {
if (addr->sa_family != sctp_ifap->address.sa.sa_family)
continue;
Improve compilation of SCTP code without INET support. Some bugs where fixed while doing this: * ASCONF-ACK messages might use wrong port number when using IPv6. * Checking for additional addresses takes the correct address into account and also does not do more comparisons than necessary. This patch is based on one received from bz@ who was sponsored by The FreeBSD Foundation and iXsystems. MFC after: 1 week
2011-04-30 11:18:16 +00:00
#ifdef INET
Adds a hash table to speed local address lookup on a per VRF basis (BSD has only one VRF currently). Hash table is sized to 16 but may need to be adjusted for machines with large numbers of addresses. Reviewed by: gnn
2007-03-19 11:11:16 +00:00
if (addr->sa_family == AF_INET) {
if (((struct sockaddr_in *)addr)->sin_addr.s_addr ==
sctp_ifap->address.sin.sin_addr.s_addr) {
/* found him. */
if (holds_lock == 0)
- The address lock is changed to a rwlock. This also involves macro changes to have a RLOCK and a WLOCK and placing the correct version within the code. - The INP-INFO lock is changed to a rwlock. - When sctp_shutdown() is called on Mac OS X, the socket lock is held. So call sctp_chunk_output with SCTP_SO_LOCKED and not SCTP_SO_NOT_LOCKED. - Add SCTP_IPI_ADDR_[RW]LOCK and SCTP_IPI_ADDR_[RW]UNLOCK for Mac OS X. - u_int64_t -> uint64_t - add missing addr unlock for error return path Approved by: re@freebsd.org (K Smith)
2007-09-18 15:16:39 +00:00
SCTP_IPI_ADDR_RUNLOCK();
Adds a hash table to speed local address lookup on a per VRF basis (BSD has only one VRF currently). Hash table is sized to 16 but may need to be adjusted for machines with large numbers of addresses. Reviewed by: gnn
2007-03-19 11:11:16 +00:00
return (sctp_ifap);
break;
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
}
Improve compilation of SCTP code without INET support. Some bugs where fixed while doing this: * ASCONF-ACK messages might use wrong port number when using IPv6. * Checking for additional addresses takes the correct address into account and also does not do more comparisons than necessary. This patch is based on one received from bz@ who was sponsored by The FreeBSD Foundation and iXsystems. MFC after: 1 week
2011-04-30 11:18:16 +00:00
#endif
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#ifdef INET6
if (addr->sa_family == AF_INET6) {
- Adds support for the multi-asconf (From Kozuka-san) - Adds some prepwork (Not all yet) for vimage in particular support the delete the sctppcbinfo.xx structs. There is still a leak in here if it were to be called plus we stil need the regrouping (From Me and Michael Tuexen) - Adds support for UDP tunneling. For BSD there is no socket yet setup so its disabled, but major argument changes are in here to emcompass the passing of the port number (zero when you don't have a udp tunnel, the default for BSD). Will add some hooks in UDP here shortly (discussed with Robert) that will allow easy tunneling. (Mainly from Peter Lei and Michael Tuexen with some BSD work from me :-D) - Some ease for windows, evidently leave is reserved by their compile move label leave: -> out: MFC after: 1 week
2008-05-20 13:47:46 +00:00
if (SCTP6_ARE_ADDR_EQUAL((struct sockaddr_in6 *)addr,
&sctp_ifap->address.sin6)) {
Adds a hash table to speed local address lookup on a per VRF basis (BSD has only one VRF currently). Hash table is sized to 16 but may need to be adjusted for machines with large numbers of addresses. Reviewed by: gnn
2007-03-19 11:11:16 +00:00
/* found him. */
if (holds_lock == 0)
- The address lock is changed to a rwlock. This also involves macro changes to have a RLOCK and a WLOCK and placing the correct version within the code. - The INP-INFO lock is changed to a rwlock. - When sctp_shutdown() is called on Mac OS X, the socket lock is held. So call sctp_chunk_output with SCTP_SO_LOCKED and not SCTP_SO_NOT_LOCKED. - Add SCTP_IPI_ADDR_[RW]LOCK and SCTP_IPI_ADDR_[RW]UNLOCK for Mac OS X. - u_int64_t -> uint64_t - add missing addr unlock for error return path Approved by: re@freebsd.org (K Smith)
2007-09-18 15:16:39 +00:00
SCTP_IPI_ADDR_RUNLOCK();
Adds a hash table to speed local address lookup on a per VRF basis (BSD has only one VRF currently). Hash table is sized to 16 but may need to be adjusted for machines with large numbers of addresses. Reviewed by: gnn
2007-03-19 11:11:16 +00:00
return (sctp_ifap);
break;
}
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#endif
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
}
if (holds_lock == 0)
- The address lock is changed to a rwlock. This also involves macro changes to have a RLOCK and a WLOCK and placing the correct version within the code. - The INP-INFO lock is changed to a rwlock. - When sctp_shutdown() is called on Mac OS X, the socket lock is held. So call sctp_chunk_output with SCTP_SO_LOCKED and not SCTP_SO_NOT_LOCKED. - Add SCTP_IPI_ADDR_[RW]LOCK and SCTP_IPI_ADDR_[RW]UNLOCK for Mac OS X. - u_int64_t -> uint64_t - add missing addr unlock for error return path Approved by: re@freebsd.org (K Smith)
2007-09-18 15:16:39 +00:00
SCTP_IPI_ADDR_RUNLOCK();
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (NULL);
}
static void
- Fixed (Apple) compiler warnings in sctp_input.c, sctputil.c, sctp_output.c - Fixed a LOR in handling a cookie. Turns out create lock is applied. And if we abort processing, this causes LOR. Changed to force the timer to clean up, that way create lock is released.
2007-05-30 22:34:21 +00:00
sctp_user_rcvd(struct sctp_tcb *stcb, uint32_t * freed_so_far, int hold_rlock,
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
uint32_t rwnd_req)
{
/* User pulled some data, do we need a rwnd update? */
int r_unlocked = 0;
uint32_t dif, rwnd;
struct socket *so = NULL;
if (stcb == NULL)
return;
Tons of fixes to get all the 64bit issues removed. This also moves two 16 bit int's to become 32 bit values so we do not have to use atomic_add_16. Most of the changes are %p, casts and other various nasty's that were in the orignal code base. With this commit my machine will now do a build universe.. however I as yet have not tested on a 64bit machine .. it may not work :-(
2006-11-05 13:25:18 +00:00
atomic_add_int(&stcb->asoc.refcnt, 1);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
- window update sacks sent incorrectly after shutdown which caused extra abort from peer. - RTT time calculation was not being done in express sack handling since it refered to an unused variable (rto_pending). Removed variable. - socket buffer high water access macro-ized.
2007-03-20 10:23:11 +00:00
if (stcb->asoc.state & (SCTP_STATE_ABOUT_TO_BE_FREED |
SCTP_STATE_SHUTDOWN_RECEIVED |
- Fixed (Apple) compiler warnings in sctp_input.c, sctputil.c, sctp_output.c - Fixed a LOR in handling a cookie. Turns out create lock is applied. And if we abort processing, this causes LOR. Changed to force the timer to clean up, that way create lock is released.
2007-05-30 22:34:21 +00:00
SCTP_STATE_SHUTDOWN_ACK_SENT)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Pre-check If we are freeing no update */
goto no_lock;
}
SCTP_INP_INCR_REF(stcb->sctp_ep);
if ((stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE) ||
(stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_SOCKET_ALLGONE)) {
goto out;
}
so = stcb->sctp_socket;
if (so == NULL) {
goto out;
}
atomic_add_int(&stcb->freed_by_sorcv_sincelast, *freed_so_far);
/* Have you have freed enough to look */
*freed_so_far = 0;
/* Yep, its worth a look and the lock overhead */
/* Figure out what the rwnd would be */
rwnd = sctp_calc_rwnd(stcb, &stcb->asoc);
if (rwnd >= stcb->asoc.my_last_reported_rwnd) {
dif = rwnd - stcb->asoc.my_last_reported_rwnd;
} else {
dif = 0;
}
if (dif >= rwnd_req) {
if (hold_rlock) {
SCTP_INP_READ_UNLOCK(stcb->sctp_ep);
r_unlocked = 1;
}
if (stcb->asoc.state & SCTP_STATE_ABOUT_TO_BE_FREED) {
/*
* One last check before we allow the guy possibly
* to get in. There is a race, where the guy has not
* reached the gate. In that case
*/
goto out;
}
SCTP_TCB_LOCK(stcb);
if (stcb->asoc.state & SCTP_STATE_ABOUT_TO_BE_FREED) {
/* No reports here */
SCTP_TCB_UNLOCK(stcb);
goto out;
}
SCTP_STAT_INCR(sctps_wu_sacks_sent);
Fix a locking issue showing up on Mac OS X when subscribing to authentication events. DTLS/SCTP renegotiations trigger the bug. MFC after: 2 weeks.
2011-05-08 09:11:59 +00:00
sctp_send_sack(stcb, SCTP_SO_LOCKED);
Code from the hack-session known as the IETF (and a bit of debugging afterwards): - Fix protection code for notification generation. - Decouple associd from vtag - Allow vtags to have less strigent requirements in non-uniqueness. o don't pre-hash them when you issue one in a cookie. o Allow duplicates and use addresses and ports to discriminate amongst the duplicates during lookup. - Add support for the NAT draft draft-ietf-behave-sctpnat-00, this is still experimental and needs more extensive testing with the Jason Butt ipfw changes. - Support for the SENDER_DRY event to get DTLS in OpenSSL working with a set of patches from Michael Tuexen (hopefully heading to OpenSSL soon). - Update the support of SCTP-AUTH by Peter Lei. - Use macros for refcounting. - Fix MTU for UDP encapsulation. - Fix reporting back of unsent data. - Update assoc send counter handling to be consistent with endpoint sent counter. - Fix a bug in PR-SCTP. - Fix so we only send another FWD-TSN when a SACK arrives IF and only if the adv-peer-ack point progressed. However we still make sure a timer is running if we do have an adv_peer_ack point. - Fix PR-SCTP bug where chunks were retransmitted if they are sent unreliable but not abandoned yet. With the help of: Michael Teuxen and Peter Lei :-) MFC after: 4 weeks
2008-12-06 13:19:54 +00:00
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_chunk_output(stcb->sctp_ep, stcb,
- Locking compatiability changes. This involves adding additional flags to many function calls. The flags only get used in BSD when we compile with lock testing. These flags allow apple to escape the "giant" lock it holds on the socket and have more fine-grained locking in the NKE. It also allows us to test (with witness) the locking used by apple via a compile switch (manually applied). Approved by: re@freebsd.org(B Mah)
2007-09-08 11:35:11 +00:00
SCTP_OUTPUT_FROM_USR_RCVD, SCTP_SO_LOCKED);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* make sure no timer is running */
Some more debug info cleanup. MFC after: 3 days
2015-05-28 16:39:22 +00:00
sctp_timer_stop(SCTP_TIMER_TYPE_RECV, stcb->sctp_ep, stcb, NULL,
SCTP_FROM_SCTPUTIL + SCTP_LOC_6);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
SCTP_TCB_UNLOCK(stcb);
} else {
/* Update how much we have pending */
stcb->freed_by_sorcv_sincelast = dif;
}
out:
if (so && r_unlocked && hold_rlock) {
SCTP_INP_READ_LOCK(stcb->sctp_ep);
}
SCTP_INP_DECR_REF(stcb->sctp_ep);
no_lock:
Tons of fixes to get all the 64bit issues removed. This also moves two 16 bit int's to become 32 bit values so we do not have to use atomic_add_16. Most of the changes are %p, casts and other various nasty's that were in the orignal code base. With this commit my machine will now do a build universe.. however I as yet have not tested on a 64bit machine .. it may not work :-(
2006-11-05 13:25:18 +00:00
atomic_add_int(&stcb->asoc.refcnt, -1);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return;
}
int
sctp_sorecvmsg(struct socket *so,
struct uio *uio,
struct mbuf **mp,
struct sockaddr *from,
int fromlen,
int *msg_flags,
struct sctp_sndrcvinfo *sinfo,
int filling_sinfo)
{
/*
* MSG flags we will look at MSG_DONTWAIT - non-blocking IO.
* MSG_PEEK - Look don't touch :-D (only valid with OUT mbuf copy
* mp=NULL thus uio is the copy method to userland) MSG_WAITALL - ??
* On the way out we may send out any combination of:
* MSG_NOTIFICATION MSG_EOR
*
*/
struct sctp_inpcb *inp = NULL;
int my_len = 0;
int cp_len = 0, error = 0;
struct sctp_queued_to_read *control = NULL, *ctl = NULL, *nxt = NULL;
Remove unused variable/assignment. MFC after: 3 weeks.
2010-09-15 23:40:36 +00:00
struct mbuf *m = NULL;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
struct sctp_tcb *stcb = NULL;
int wakeup_read_socket = 0;
int freecnt_applied = 0;
int out_flags = 0, in_flags = 0;
int block_allowed = 1;
- Fixed (Apple) compiler warnings in sctp_input.c, sctputil.c, sctp_output.c - Fixed a LOR in handling a cookie. Turns out create lock is applied. And if we abort processing, this causes LOR. Changed to force the timer to clean up, that way create lock is released.
2007-05-30 22:34:21 +00:00
uint32_t freed_so_far = 0;
- Fix a bug in sctp_calc_rwnd() which resulted in wrong rwnd predictions. - Fix a signedness problem that shows up in some 64 bit platforms (macos). MFC after: 1 week
2007-11-10 00:47:14 +00:00
uint32_t copied_so_far = 0;
- most all includes (#include <>) migrate to the sctp_os_bsd.h file - Finally all splxx() are removed - Count error fixed in mapping array which might cause a wrong cumack generation. - Invariants around panic for case D + printf when no invariants. - one-to-one model race condition fixed by using a pre-formed connection and then completing the work so accept won't happen on a non-formed association. - Some additional paranoia checks in sctp_output. - Locks that were missing in the accept code. Approved by: gnn
2007-01-18 09:58:43 +00:00
int in_eeor_mode = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
int no_rcv_needed = 0;
uint32_t rwnd_req = 0;
int hold_sblock = 0;
int hold_rlock = 0;
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
ssize_t slen = 0;
- Fixed (Apple) compiler warnings in sctp_input.c, sctputil.c, sctp_output.c - Fixed a LOR in handling a cookie. Turns out create lock is applied. And if we abort processing, this causes LOR. Changed to force the timer to clean up, that way create lock is released.
2007-05-30 22:34:21 +00:00
uint32_t held_length = 0;
sblock() implements a sleep lock by interlocking SB_WANT and SB_LOCK flags on each socket buffer with the socket buffer's mutex. This sleep lock is used to serialize I/O on sockets in order to prevent I/O interlacing. This change replaces the custom sleep lock with an sx(9) lock, which results in marginally better performance, better handling of contention during simultaneous socket I/O across multiple threads, and a cleaner separation between the different layers of locking in socket buffers. Specifically, the socket buffer mutex is now solely responsible for serializing simultaneous operation on the socket buffer data structure, and not for I/O serialization. While here, fix two historic bugs: (1) a bug allowing I/O to be occasionally interlaced during long I/O operations (discovere by Isilon). (2) a bug in which failed non-blocking acquisition of the socket buffer I/O serialization lock might be ignored (discovered by sam). SCTP portion of this patch submitted by rrs.
2007-05-03 14:42:42 +00:00
int sockbuf_lock = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
if (uio == NULL) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
return (EINVAL);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (msg_flags) {
in_flags = *msg_flags;
- fix source address selection when picking an acceptable address - name change of prefered -> preferred - CMT fast recover code added. - Comment fixes in CMT. - We were not giving a reason of cant_start_asoc per socket api if we failed to get init/or/cookie to bring up an assoc. Change so we don't just give a generic "comm lost" but look at actual states of dying assoc. - change "crc32" arguments to "crc32c" to silence strict/noisy compiler warnings when crc32() is also declared - A few minor tweaks to get the portable stuff truely portable for sctp6_usrreq.c :-D - one-2-one style vrf match problem. - window recovery would leave chks marked for retran during window probes on the sent queue. This would then cause an out-of-order problem and assure that the flight size "problem" would occur. - Solves a flight size logging issue that caused rwnd overruns, flight size off as well as false retransmissions.g - Macroize the up and down of flight size. - Fix a ECNE bug in its counting. - The strict_sacks options was causing aborts when window probing was active, fix to make strict sacks a bit smarter about what the next unsent TSN is. - Fixes a one-2-one wakeup bug found by Martin Kulas. - If-defed out form, Andre's copy routines pending his commit of at least m_last().. need to adjust for 6.2 as well.. since m_last won't exist. Reviewed by: gnn
2007-04-14 09:44:09 +00:00
if (in_flags & MSG_PEEK)
SCTP_STAT_INCR(sctps_read_peeks);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
in_flags = 0;
}
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
slen = uio->uio_resid;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Pull in and set up our int flags */
if (in_flags & MSG_OOB) {
/* Out of band's NOT supported */
return (EOPNOTSUPP);
}
if ((in_flags & MSG_PEEK) && (mp != NULL)) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (EINVAL);
}
if ((in_flags & (MSG_DONTWAIT
| MSG_NBIO
)) ||
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
SCTP_SO_IS_NBIO(so)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
block_allowed = 0;
}
/* setup the endpoint */
inp = (struct sctp_inpcb *)so->so_pcb;
if (inp == NULL) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(NULL, NULL, NULL, SCTP_FROM_SCTPUTIL, EFAULT);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (EFAULT);
}
- window update sacks sent incorrectly after shutdown which caused extra abort from peer. - RTT time calculation was not being done in express sack handling since it refered to an unused variable (rto_pending). Removed variable. - socket buffer high water access macro-ized.
2007-03-20 10:23:11 +00:00
rwnd_req = (SCTP_SB_LIMIT_RCV(so) >> SCTP_RWND_HIWAT_SHIFT);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Must be at least a MTU's worth */
if (rwnd_req < SCTP_MIN_RWND)
rwnd_req = SCTP_MIN_RWND;
in_eeor_mode = sctp_is_feature_on(inp, SCTP_PCB_FLAGS_EXPLICIT_EOR);
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_RECV_RWND_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_misc_ints(SCTP_SORECV_ENTER,
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
rwnd_req, in_eeor_mode, so->so_rcv.sb_cc, (uint32_t) uio->uio_resid);
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
}
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_RECV_RWND_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_misc_ints(SCTP_SORECV_ENTERPL,
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
rwnd_req, block_allowed, so->so_rcv.sb_cc, (uint32_t) uio->uio_resid);
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
}
Correct two problems relating to sorflush(), which is called to flush read socket buffers in shutdown() and close(): - Call socantrcvmore() before sblock() to dislodge any threads that might be sleeping (potentially indefinitely) while holding sblock(), such as a thread blocked in recv(). - Flag the sblock() call as non-interruptible so that a signal delivered to the thread calling sorflush() doesn't cause sblock() to fail. The sblock() is required to ensure that all other socket consumer threads have, in fact, left, and do not enter, the socket buffer until we're done flushin it. To implement the latter, change the 'flags' argument to sblock() to accept two flags, SBL_WAIT and SBL_NOINTR, rather than one M_WAITOK flag. When SBL_NOINTR is set, it forces a non-interruptible sx acquisition, regardless of the setting of the disposition of SB_NOINTR on the socket buffer; without this change it would be possible for another thread to clear SB_NOINTR between when the socket buffer mutex is released and sblock() is invoked. Reviewed by: bz, kmacy Reported by: Jos Backus <jos at catnook dot com>
2008-01-31 08:22:24 +00:00
error = sblock(&so->so_rcv, (block_allowed ? SBL_WAIT : 0));
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (error) {
goto release_unlocked;
}
Fix possible spurious sbunlock in sctp_sorecvmsg. Reviewed by: tuexen Approved by: trasz (mentor) MFC after: 3 days
2012-11-06 23:04:23 +00:00
sockbuf_lock = 1;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
restart:
sblock() implements a sleep lock by interlocking SB_WANT and SB_LOCK flags on each socket buffer with the socket buffer's mutex. This sleep lock is used to serialize I/O on sockets in order to prevent I/O interlacing. This change replaces the custom sleep lock with an sx(9) lock, which results in marginally better performance, better handling of contention during simultaneous socket I/O across multiple threads, and a cleaner separation between the different layers of locking in socket buffers. Specifically, the socket buffer mutex is now solely responsible for serializing simultaneous operation on the socket buffer data structure, and not for I/O serialization. While here, fix two historic bugs: (1) a bug allowing I/O to be occasionally interlaced during long I/O operations (discovere by Isilon). (2) a bug in which failed non-blocking acquisition of the socket buffer I/O serialization lock might be ignored (discovered by sam). SCTP portion of this patch submitted by rrs.
2007-05-03 14:42:42 +00:00
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
restart_nosblocks:
if (hold_sblock == 0) {
SOCKBUF_LOCK(&so->so_rcv);
hold_sblock = 1;
}
if ((inp->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE) ||
(inp->sctp_flags & SCTP_PCB_FLAGS_SOCKET_ALLGONE)) {
goto out;
}
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
if ((so->so_rcv.sb_state & SBS_CANTRCVMORE) && (so->so_rcv.sb_cc == 0)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (so->so_error) {
error = so->so_error;
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
if ((in_flags & MSG_PEEK) == 0)
so->so_error = 0;
- Found a problem in non-blocking sends. When sending, once the locks are all unlocked to do the copy's in, its possible that other events could then raise the number of bytes outstanding pushing it so not all the message would fit. This would then cause us to send only part of the message. This fix makes it so we keep a "reserved" amount that can be kept in mind when making calculations to send. - rcv msg args with a NULL/NULL for to/tolen will return an error incorrectly for the 1-2-1 model. - We were not doing 0 len return correctly and not setting cantrcv more correctly. Previouly we "fixed" this area by taking out the socantrcv since we then could not get the data out. The correct rix is to still flag the socket but alow a by-pass route to continue to read until all data is consumed. MFC after: 1 week
2007-12-04 14:41:48 +00:00
goto out;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
if (so->so_rcv.sb_cc == 0) {
- Found a problem in non-blocking sends. When sending, once the locks are all unlocked to do the copy's in, its possible that other events could then raise the number of bytes outstanding pushing it so not all the message would fit. This would then cause us to send only part of the message. This fix makes it so we keep a "reserved" amount that can be kept in mind when making calculations to send. - rcv msg args with a NULL/NULL for to/tolen will return an error incorrectly for the 1-2-1 model. - We were not doing 0 len return correctly and not setting cantrcv more correctly. Previouly we "fixed" this area by taking out the socantrcv since we then could not get the data out. The correct rix is to still flag the socket but alow a by-pass route to continue to read until all data is consumed. MFC after: 1 week
2007-12-04 14:41:48 +00:00
/* indicate EOF */
error = 0;
goto out;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
if ((so->so_rcv.sb_cc <= held_length) && block_allowed) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* we need to wait for data */
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
if ((so->so_rcv.sb_cc == 0) &&
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
((inp->sctp_flags & SCTP_PCB_FLAGS_TCPTYPE) ||
(inp->sctp_flags & SCTP_PCB_FLAGS_IN_TCPPOOL))) {
if ((inp->sctp_flags & SCTP_PCB_FLAGS_CONNECTED) == 0) {
/*
* For active open side clear flags for
* re-use passive open is blocked by
* connect.
*/
if (inp->sctp_flags & SCTP_PCB_FLAGS_WAS_ABORTED) {
/*
* You were aborted, passive side
* always hits here
*/
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, ECONNRESET);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
error = ECONNRESET;
}
so->so_state &= ~(SS_ISCONNECTING |
SS_ISDISCONNECTING |
SS_ISCONFIRMING |
SS_ISCONNECTED);
if (error == 0) {
if ((inp->sctp_flags & SCTP_PCB_FLAGS_WAS_CONNECTED) == 0) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, ENOTCONN);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
error = ENOTCONN;
}
}
goto out;
}
}
error = sbwait(&so->so_rcv);
if (error) {
goto out;
}
held_length = 0;
goto restart_nosblocks;
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
} else if (so->so_rcv.sb_cc == 0) {
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
if (so->so_error) {
error = so->so_error;
if ((in_flags & MSG_PEEK) == 0)
so->so_error = 0;
} else {
if ((inp->sctp_flags & SCTP_PCB_FLAGS_TCPTYPE) ||
(inp->sctp_flags & SCTP_PCB_FLAGS_IN_TCPPOOL)) {
if ((inp->sctp_flags & SCTP_PCB_FLAGS_CONNECTED) == 0) {
/*
* For active open side clear flags
* for re-use passive open is
* blocked by connect.
*/
if (inp->sctp_flags & SCTP_PCB_FLAGS_WAS_ABORTED) {
/*
* You were aborted, passive
* side always hits here
*/
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, ECONNRESET);
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
error = ECONNRESET;
}
so->so_state &= ~(SS_ISCONNECTING |
SS_ISDISCONNECTING |
SS_ISCONFIRMING |
SS_ISCONNECTED);
if (error == 0) {
if ((inp->sctp_flags & SCTP_PCB_FLAGS_WAS_CONNECTED) == 0) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, ENOTCONN);
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
error = ENOTCONN;
}
}
goto out;
}
}
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EWOULDBLOCK);
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
error = EWOULDBLOCK;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
goto out;
}
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave (0-2). - Codenomicon security test updates - length checks and such. - Bug in stream reset (2 actually). - setpeerprimary could unlock a null pointer, fixed. - Added a flag in the pcb so netstat can see if we are listening easier. Obtained from: (some of the Listen changes from Weongyo Jeong)
2007-05-02 12:50:13 +00:00
if (hold_sblock == 1) {
SOCKBUF_UNLOCK(&so->so_rcv);
hold_sblock = 0;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* we possibly have data we can read */
- Fixed 1-2-1 model to not worry about associd in sockopts - Fixed RTOinfo for bounding. - Fixed connect() to return ECONNREFUSED when an ABORT is received. - Added comments to direct Static Analysis not to look at some things it does not understand (comments are /* sa_ignore XXXXX */) - Bind when colliding was broken, missing not_found = 1 before checking to see if the port was in use caused endless bind loop. - Cookie life needs to be in milliseconds to conform to socket api. - Cookie life is not supposed to change if its 0, On the assoc level set we changed it to 0 opps. - Two more static analysis issues identified by the cisco tool. Null checks needed. - An issue for sendfile(). Need to validate the correct input argument. - When sending failed due to a no route to host, we leaked the mbuf chain failing to call m_freem(). - Fix #ifdef issue for getting hash block len when HAVE_SHA2 is NOT defined Reviewed by: gnn
2007-05-17 12:16:24 +00:00
/* sa_ignore FREED_MEMORY */
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
control = TAILQ_FIRST(&inp->read_queue);
if (control == NULL) {
/*
* This could be happening since the appender did the
* increment but as not yet did the tailq insert onto the
* read_queue
*/
if (hold_rlock == 0) {
SCTP_INP_READ_LOCK(inp);
}
control = TAILQ_FIRST(&inp->read_queue);
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
if ((control == NULL) && (so->so_rcv.sb_cc != 0)) {
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
#ifdef INVARIANTS
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
panic("Huh, its non zero and nothing on control?");
#endif
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
so->so_rcv.sb_cc = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
SCTP_INP_READ_UNLOCK(inp);
hold_rlock = 0;
goto restart;
}
if ((control->length == 0) &&
(control->do_not_ref_stcb)) {
/*
* Clean up code for freeing assoc that left behind a
* pdapi.. maybe a peer in EEOR that just closed after
* sending and never indicated a EOR.
*/
if (hold_rlock == 0) {
hold_rlock = 1;
SCTP_INP_READ_LOCK(inp);
}
control->held_length = 0;
if (control->data) {
/* Hmm there is data here .. fix */
- Fixed (Apple) compiler warnings in sctp_input.c, sctputil.c, sctp_output.c - Fixed a LOR in handling a cookie. Turns out create lock is applied. And if we abort processing, this causes LOR. Changed to force the timer to clean up, that way create lock is released.
2007-05-30 22:34:21 +00:00
struct mbuf *m_tmp;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
int cnt = 0;
- Fixed (Apple) compiler warnings in sctp_input.c, sctputil.c, sctp_output.c - Fixed a LOR in handling a cookie. Turns out create lock is applied. And if we abort processing, this causes LOR. Changed to force the timer to clean up, that way create lock is released.
2007-05-30 22:34:21 +00:00
m_tmp = control->data;
while (m_tmp) {
cnt += SCTP_BUF_LEN(m_tmp);
if (SCTP_BUF_NEXT(m_tmp) == NULL) {
control->tail_mbuf = m_tmp;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
control->end_added = 1;
}
- Fixed (Apple) compiler warnings in sctp_input.c, sctputil.c, sctp_output.c - Fixed a LOR in handling a cookie. Turns out create lock is applied. And if we abort processing, this causes LOR. Changed to force the timer to clean up, that way create lock is released.
2007-05-30 22:34:21 +00:00
m_tmp = SCTP_BUF_NEXT(m_tmp);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
control->length = cnt;
} else {
/* remove it */
TAILQ_REMOVE(&inp->read_queue, control, next);
/* Add back any hiddend data */
sctp_free_remote_addr(control->whoFrom);
sctp_free_a_readq(stcb, control);
}
if (hold_rlock) {
hold_rlock = 0;
SCTP_INP_READ_UNLOCK(inp);
}
goto restart;
}
* Fix a bug where PR-SCTP settings are ignore when using implicit association setup. * Fix a bug where message with illegal stream ids are not deleted. * Fix a crash when reporting back unsent messages from the send_queue. * Fix a bug related to INIT retransmission when the socket is already closed. * Fix a bug where associations were stalled when partial delivery API was enabled. * Fix a bug where the receive buffer size was smaller than the partial_delivery_point. Approved by: re, rrs (mentor) MFC after: One day.
2009-08-15 21:10:52 +00:00
if ((control->length == 0) &&
(control->end_added == 1)) {
/*
* Do we also need to check for (control->pdapi_aborted ==
* 1)?
*/
if (hold_rlock == 0) {
hold_rlock = 1;
SCTP_INP_READ_LOCK(inp);
}
TAILQ_REMOVE(&inp->read_queue, control, next);
if (control->data) {
#ifdef INVARIANTS
panic("control->data not null but control->length == 0");
#else
SCTP_PRINTF("Strange, data left in the control buffer. Cleaning up.\n");
sctp_m_freem(control->data);
control->data = NULL;
#endif
}
if (control->aux_data) {
sctp_m_free(control->aux_data);
control->aux_data = NULL;
}
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
if (control->on_strm_q) {
panic("About to free ctl:%p so:%p and its in %d",
control, so, control->on_strm_q);
}
* Fix a bug where PR-SCTP settings are ignore when using implicit association setup. * Fix a bug where message with illegal stream ids are not deleted. * Fix a crash when reporting back unsent messages from the send_queue. * Fix a bug related to INIT retransmission when the socket is already closed. * Fix a bug where associations were stalled when partial delivery API was enabled. * Fix a bug where the receive buffer size was smaller than the partial_delivery_point. Approved by: re, rrs (mentor) MFC after: One day.
2009-08-15 21:10:52 +00:00
sctp_free_remote_addr(control->whoFrom);
sctp_free_a_readq(stcb, control);
if (hold_rlock) {
hold_rlock = 0;
SCTP_INP_READ_UNLOCK(inp);
}
goto restart;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (control->length == 0) {
if ((sctp_is_feature_on(inp, SCTP_PCB_FLAGS_FRAG_INTERLEAVE)) &&
(filling_sinfo)) {
/* find a more suitable one then this */
ctl = TAILQ_NEXT(control, next);
while (ctl) {
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
if ((ctl->stcb != control->stcb) && (ctl->length) &&
(ctl->some_taken ||
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
(ctl->spec_flags & M_NOTIFICATION) ||
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
((ctl->do_not_ref_stcb == 0) &&
(ctl->stcb->asoc.strmin[ctl->sinfo_stream].delivery_started == 0)))
) {
/*-
* If we have a different TCB next, and there is data
* present. If we have already taken some (pdapi), OR we can
* ref the tcb and no delivery as started on this stream, we
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
* take it. Note we allow a notification on a different
* assoc to be delivered..
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
*/
control = ctl;
goto found_one;
} else if ((sctp_is_feature_on(inp, SCTP_PCB_FLAGS_INTERLEAVE_STRMS)) &&
(ctl->length) &&
((ctl->some_taken) ||
((ctl->do_not_ref_stcb == 0) &&
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
((ctl->spec_flags & M_NOTIFICATION) == 0) &&
* Fix some race condition in SACK/NR-SACK processing. * Fix handling of mapping arrays when draining mbufs or processing FORWARD-TSN chunks. * Cleanup code (no duplicate code anymore for SACKs and NR-SACKs). Part of this code was developed together with rrs. MFC after: 2 weeks.
2010-04-03 15:40:14 +00:00
(ctl->stcb->asoc.strmin[ctl->sinfo_stream].delivery_started == 0)))) {
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
/*-
* If we have the same tcb, and there is data present, and we
* have the strm interleave feature present. Then if we have
* taken some (pdapi) or we can refer to tht tcb AND we have
* not started a delivery for this stream, we can take it.
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
* Note we do NOT allow a notificaiton on the same assoc to
* be delivered.
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
*/
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
control = ctl;
goto found_one;
}
ctl = TAILQ_NEXT(ctl, next);
}
}
/*
* if we reach here, not suitable replacement is available
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
* <or> fragment interleave is NOT on. So stuff the sb_cc
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
* into the our held count, and its time to sleep again.
*/
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
held_length = so->so_rcv.sb_cc;
control->held_length = so->so_rcv.sb_cc;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
goto restart;
}
/* Clear the held length since there is something to read */
control->held_length = 0;
if (hold_rlock) {
SCTP_INP_READ_UNLOCK(inp);
hold_rlock = 0;
}
found_one:
/*
* If we reach here, control has a some data for us to read off.
* Note that stcb COULD be NULL.
*/
- optimize the initialization of the SB max variables. - Missing lock when sending data and moving it to the outqueue. - If a mbuf alloc fails during moving to outqueue the reassembly of the old mbuf chain was incorrect. - some_taken becomes a counter in sctputil.c instead of a set to 1. - Fix a panic to be only under invarients and have a proper recovery. - msg_flags needed to be set.to the value collected not or'd. MFC after: 1 week
2007-12-06 00:22:55 +00:00
control->some_taken++;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (hold_sblock) {
SOCKBUF_UNLOCK(&so->so_rcv);
hold_sblock = 0;
}
stcb = control->stcb;
if (stcb) {
- Fix a memory overwrite when the mapping array is expanded, size of expansion was not taken int consideration. - Fix so vtag hash is 1 bigger so that it modulo's out correctly, avoids a panic when restart with right modulo happens. - do not dereference stcb when control->do_not_ref_stcb is set - Fix up packet logging to not often use a lock and also to add to options. - Fix some logging option duplication in the sctputil.h
2007-05-30 17:39:45 +00:00
if ((control->do_not_ref_stcb == 0) &&
(stcb->asoc.state & SCTP_STATE_ABOUT_TO_BE_FREED)) {
Tons of fixes to get all the 64bit issues removed. This also moves two 16 bit int's to become 32 bit values so we do not have to use atomic_add_16. Most of the changes are %p, casts and other various nasty's that were in the orignal code base. With this commit my machine will now do a build universe.. however I as yet have not tested on a 64bit machine .. it may not work :-(
2006-11-05 13:25:18 +00:00
if (freecnt_applied == 0)
stcb = NULL;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else if (control->do_not_ref_stcb == 0) {
/* you can't free it on me please */
/*
* The lock on the socket buffer protects us so the
* free code will stop. But since we used the
* socketbuf lock and the sender uses the tcb_lock
* to increment, we need to use the atomic add to
* the refcnt
*/
- Bug fix managing congestion parameter on immediate retransmittion by handover event (fast mobility code) - Fixed problem of mobility code which is caused by remaining parameters in the deleted primary destination. - Add a missing lock. When a peer sends an INIT, and while we are processing it to send an INIT-ACK the socket is closed, we did not hold a lock to keep the socket from going away. Add protection for this case. - Fix so that arwnd is alway uses the minimal rwnd if the user has set the socket buffer smaller. Found this when the test org decided to see what happens when you set in a rwnd of 10 bytes (which is not allowed per RFC .. 4k is minimum). - Fixes so a cookie-echo ootb will NOT cause an abort to be sent. This was happening in a MPI collision case. - Examined all panics and unless there was no recovery, moved any that were not already to INVARANTS. Approved by: re@freebsd.org (gnn)
2007-10-01 03:22:29 +00:00
if (freecnt_applied) {
#ifdef INVARIANTS
- Fixes so we won't try to start a timer when we hold a wq lock for the iterator. Panda uses a silly recursive lock they hold through the timer. - Add poor mans wireshark compile option.. - Allocate and start using SCTP_M_XXX for all SCTP_MALLOC() calls. - sysctl now will get back the refcnt for viewing by onlookers. Reviewed by: gnn
2007-05-29 09:29:03 +00:00
panic("refcnt already incremented");
- Bug fix managing congestion parameter on immediate retransmittion by handover event (fast mobility code) - Fixed problem of mobility code which is caused by remaining parameters in the deleted primary destination. - Add a missing lock. When a peer sends an INIT, and while we are processing it to send an INIT-ACK the socket is closed, we did not hold a lock to keep the socket from going away. Add protection for this case. - Fix so that arwnd is alway uses the minimal rwnd if the user has set the socket buffer smaller. Found this when the test org decided to see what happens when you set in a rwnd of 10 bytes (which is not allowed per RFC .. 4k is minimum). - Fixes so a cookie-echo ootb will NOT cause an abort to be sent. This was happening in a MPI collision case. - Examined all panics and unless there was no recovery, moved any that were not already to INVARANTS. Approved by: re@freebsd.org (gnn)
2007-10-01 03:22:29 +00:00
#else
Use SCTP_PRINTF() instead of printf() in all SCTP sources. MFC after: 3 days
2012-05-04 09:27:00 +00:00
SCTP_PRINTF("refcnt already incremented?\n");
- Bug fix managing congestion parameter on immediate retransmittion by handover event (fast mobility code) - Fixed problem of mobility code which is caused by remaining parameters in the deleted primary destination. - Add a missing lock. When a peer sends an INIT, and while we are processing it to send an INIT-ACK the socket is closed, we did not hold a lock to keep the socket from going away. Add protection for this case. - Fix so that arwnd is alway uses the minimal rwnd if the user has set the socket buffer smaller. Found this when the test org decided to see what happens when you set in a rwnd of 10 bytes (which is not allowed per RFC .. 4k is minimum). - Fixes so a cookie-echo ootb will NOT cause an abort to be sent. This was happening in a MPI collision case. - Examined all panics and unless there was no recovery, moved any that were not already to INVARANTS. Approved by: re@freebsd.org (gnn)
2007-10-01 03:22:29 +00:00
#endif
} else {
atomic_add_int(&stcb->asoc.refcnt, 1);
freecnt_applied = 1;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* Setup to remember how much we have not yet told
* the peer our rwnd has opened up. Note we grab the
* value from the tcb from last time. Note too that
- Fix a memory overwrite when the mapping array is expanded, size of expansion was not taken int consideration. - Fix so vtag hash is 1 bigger so that it modulo's out correctly, avoids a panic when restart with right modulo happens. - do not dereference stcb when control->do_not_ref_stcb is set - Fix up packet logging to not often use a lock and also to add to options. - Fix some logging option duplication in the sctputil.h
2007-05-30 17:39:45 +00:00
* sack sending clears this when a sack is sent,
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
* which is fine. Once we hit the rwnd_req, we then
* will go to the sctp_user_rcvd() that will not
* lock until it KNOWs it MUST send a WUP-SACK.
*/
freed_so_far = stcb->freed_by_sorcv_sincelast;
stcb->freed_by_sorcv_sincelast = 0;
}
}
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
if (stcb &&
((control->spec_flags & M_NOTIFICATION) == 0) &&
Two bugs: - Locks were not being unlocked when an invalid size chunk is sent in. - When a notification comes in, we cannot use it to look up the fragment interleave stream information since its not on a stream.
2007-05-06 00:01:17 +00:00
control->do_not_ref_stcb == 0) {
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave (0-2). - Codenomicon security test updates - length checks and such. - Bug in stream reset (2 actually). - setpeerprimary could unlock a null pointer, fixed. - Added a flag in the pcb so netstat can see if we are listening easier. Obtained from: (some of the Listen changes from Weongyo Jeong)
2007-05-02 12:50:13 +00:00
stcb->asoc.strmin[control->sinfo_stream].delivery_started = 1;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* First lets get off the sinfo and sockaddr info */
if ((sinfo) && filling_sinfo) {
memcpy(sinfo, control, sizeof(struct sctp_nonpad_sndrcvinfo));
nxt = TAILQ_NEXT(control, next);
Add support for the newly added SCTP API. In particular add support for: * SCTP_SNDINFO, SCTP_PRINFO, SCTP_AUTHINFO, SCTP_DSTADDRV4, and SCTP_DSTADDRV6 cmsgs. * SCTP_NXTINFO and SCTP_RCVINFO cmgs. * SCTP_EVENT, SCTP_RECVRCVINFO, SCTP_RECVNXTINFO and SCTP_DEFAULT_SNDINFO socket option. * Special association ids (SCTP_FUTURE_ASSOC, ...) * sctp_recvv() and sctp_sendv() functions. MFC after: 1 month.
2011-06-15 23:50:27 +00:00
if (sctp_is_feature_on(inp, SCTP_PCB_FLAGS_EXT_RCVINFO) ||
sctp_is_feature_on(inp, SCTP_PCB_FLAGS_RECVNXTINFO)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
struct sctp_extrcvinfo *s_extra;
s_extra = (struct sctp_extrcvinfo *)sinfo;
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
if ((nxt) &&
(nxt->length)) {
Fix typos in field names of struct sctp_extrcvinfo. Provide defines to allow applications to compile. Thanks to Jens Hoelscher for making me aware of the typos. MFC after: 1 week
2015-11-06 13:08:16 +00:00
s_extra->serinfo_next_flags = SCTP_NEXT_MSG_AVAIL;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (nxt->sinfo_flags & SCTP_UNORDERED) {
Fix typos in field names of struct sctp_extrcvinfo. Provide defines to allow applications to compile. Thanks to Jens Hoelscher for making me aware of the typos. MFC after: 1 week
2015-11-06 13:08:16 +00:00
s_extra->serinfo_next_flags |= SCTP_NEXT_MSG_IS_UNORDERED;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
if (nxt->spec_flags & M_NOTIFICATION) {
Fix typos in field names of struct sctp_extrcvinfo. Provide defines to allow applications to compile. Thanks to Jens Hoelscher for making me aware of the typos. MFC after: 1 week
2015-11-06 13:08:16 +00:00
s_extra->serinfo_next_flags |= SCTP_NEXT_MSG_IS_NOTIFICATION;
- Copyright updates (aka 2007) - ZONE get now also take a type cast so it does the cast like mtod does. - New macro SCTP_LIST_EMPTY, which in bsd is just LIST_EMPTY - Removal of const in some of the static hmac functions (not needed) - Store length changes to allow for new fields in auth - Auth code updated to current draft (this should be the RFC version we think). - use uint8_t instead of u_char in LOOPBACK address comparison - Some u_int32_t converted to uint32_t (in crc code) - A bug was found in the mib counts for ordered/unordered count, this was fixed (was referencing a freed mbuf). - SCTP_ASOCLOG_OF_TSNS added (code will probably disappear after my testing completes. It allows us to keep a small log on each assoc of the last 40 TSN's in/out and stream assignment. It is NOT in options and so is only good for private builds. - Some CMT changes in prep for Jana fixing his problem with reneging when CMT is enabled (Concurrent Multipath Transfer = CMT). - Some missing mib stats added. - Correction to number of open assoc's count in mib - Correction to os_bsd.h to get right sha2 macros - Add of special AUTH_04 flags so you can compile the code with the old format (in case the peer does not yet support the latest auth code). - Nonce sum was incorrectly being set in when ecn_nonce was NOT on. - LOR in listen with implicit bind found and fixed. - Moved away from using mbuf's for socket options to using just data pointers. The mbufs were used to harmonize NetBSD code since both Net and Open used this method. We have decided to move away from that and more conform to FreeBSD style (which makes more sense). - Very very nasty bug found in some of my "debug" code. The cookie_how collision case tracking had an endless loop in it if you got a second retransmission of a cookie collision case. This would lock up a CPU .. ugly.. - auth function goes to using size_t instead of int which conforms to socketapi better - Found the nasty bug that happens after 9 days of testing.. you get the data chunk, deliver it and due to the reference to a ch-> that every now and then has been deleted (depending on the postion in the mbuf) you have an invalid ch->ch.flags.. and thus you don't advance the stream sequence number.. so you block the stream permanently. The fix is to make local variables of these guys and set them up before you have any chance of trimming the mbuf. - style fix in sctp_util.h, not sure how this got bad maybe in the last patch? (aka it may not be in the real source). - Found interesting bug when using the extended snd/rcv info where we would get an error on receiving with this. Thats because it was NOT padded to the same size as the snd_rcv info. We increase (add the pad) so the two structs are the same size in sctp_uio.h - In sctp_usrreq.c one of the most common things we did for socket options was to cast the pointer and validate the size. This as been macro-ized to help make the code more readable. - in sctputil.c two things, the socketapi class found a missing flag type (the next msg is a notification) and a missing scope recovery was also fixed. Reviewed by: gnn
2007-02-12 23:24:31 +00:00
}
Fix typos in field names of struct sctp_extrcvinfo. Provide defines to allow applications to compile. Thanks to Jens Hoelscher for making me aware of the typos. MFC after: 1 week
2015-11-06 13:08:16 +00:00
s_extra->serinfo_next_aid = nxt->sinfo_assoc_id;
s_extra->serinfo_next_length = nxt->length;
s_extra->serinfo_next_ppid = nxt->sinfo_ppid;
s_extra->serinfo_next_stream = nxt->sinfo_stream;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (nxt->tail_mbuf != NULL) {
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (nxt->end_added) {
Fix typos in field names of struct sctp_extrcvinfo. Provide defines to allow applications to compile. Thanks to Jens Hoelscher for making me aware of the typos. MFC after: 1 week
2015-11-06 13:08:16 +00:00
s_extra->serinfo_next_flags |= SCTP_NEXT_MSG_ISCOMPLETE;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
} else {
/*
* we explicitly 0 this, since the memcpy
* got some other things beyond the older
* sinfo_ that is on the control's structure
* :-D
*/
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
nxt = NULL;
Fix typos in field names of struct sctp_extrcvinfo. Provide defines to allow applications to compile. Thanks to Jens Hoelscher for making me aware of the typos. MFC after: 1 week
2015-11-06 13:08:16 +00:00
s_extra->serinfo_next_flags = SCTP_NO_NEXT_MSG;
s_extra->serinfo_next_aid = 0;
s_extra->serinfo_next_length = 0;
s_extra->serinfo_next_ppid = 0;
s_extra->serinfo_next_stream = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
/*
* update off the real current cum-ack, if we have an stcb.
*/
- Fix a memory overwrite when the mapping array is expanded, size of expansion was not taken int consideration. - Fix so vtag hash is 1 bigger so that it modulo's out correctly, avoids a panic when restart with right modulo happens. - do not dereference stcb when control->do_not_ref_stcb is set - Fix up packet logging to not often use a lock and also to add to options. - Fix some logging option duplication in the sctputil.h
2007-05-30 17:39:45 +00:00
if ((control->do_not_ref_stcb == 0) && stcb)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sinfo->sinfo_cumtsn = stcb->asoc.cumulative_tsn;
/*
* mask off the high bits, we keep the actual chunk bits in
* there.
*/
sinfo->sinfo_flags &= 0x00ff;
- Fixes a case where doing a sysctl would leave locks held when coping out association data. - Fixes a small bug that prevented the SCTP_UNORDERED indication from going up to the app on a recv in the sinfo_flags field.
2007-06-06 00:40:41 +00:00
if ((control->sinfo_flags >> 8) & SCTP_DATA_UNORDERED) {
sinfo->sinfo_flags |= SCTP_UNORDERED;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- added pre-checks to the bindx call. - use proper tick gathering macro instead of ticks directly. - Placed reasonable boundaries on sets that a user can do that are converted to ticks from ms. - Fix CMT_PF to always check to be sure CMT is on. - Fix ticks use of CMT_PF. - put back code to allow asconfs to be queued while INITs are in flight and before the assoc is established. - During window probes, an ack'd packet might be left with the window probe mark on it causing it to be retransmitted. Change so that the flight decrease macro clears the window_probe mark. - Additional logging flight size/reading and ASOC LOG. This is only enabled if you manually insert things into opt_sctp.h since its a set of debug code only. - Found an interesting SMP race in the way data was appended which could cause a reader to lose a part of a message, had to reorder when we marked the message was complete to after the data was appended. - bug in ADD-IP for the subset bound socket case when the peer has only one address - fix ASCONF implicit success/error handling case - proper support of jails in Freebsd 6> - copy out the timeval for the 64 bit sparc world on cookie-echo alignment error crashes without this). Approved by: re(Ken Smith)
2007-07-17 20:58:26 +00:00
#ifdef SCTP_ASOCLOG_OF_TSNS
{
int index, newindex;
struct sctp_pcbtsn_rlog *entry;
do {
index = inp->readlog_index;
newindex = index + 1;
if (newindex >= SCTP_READ_LOG_SIZE) {
newindex = 0;
}
} while (atomic_cmpset_int(&inp->readlog_index, index, newindex) == 0);
entry = &inp->readlog[index];
entry->vtag = control->sinfo_assoc_id;
entry->strm = control->sinfo_stream;
entry->seq = control->sinfo_ssn;
entry->sz = control->length;
entry->flgs = control->sinfo_flags;
}
#endif
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
if ((fromlen > 0) && (from != NULL)) {
union sctp_sockstore store;
size_t len;
Small cleanup of the code. No functional change (in FreeBSD kernel). MFC after: 1 week.
2012-03-22 16:22:16 +00:00
switch (control->whoFrom->ro._l_addr.sa.sa_family) {
#ifdef INET6
case AF_INET6:
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
len = sizeof(struct sockaddr_in6);
store.sin6 = control->whoFrom->ro._l_addr.sin6;
store.sin6.sin6_port = control->port_from;
Small cleanup of the code. No functional change (in FreeBSD kernel). MFC after: 1 week.
2012-03-22 16:22:16 +00:00
break;
#endif
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
#ifdef INET
Small cleanup of the code. No functional change (in FreeBSD kernel). MFC after: 1 week.
2012-03-22 16:22:16 +00:00
case AF_INET:
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#ifdef INET6
if (sctp_is_feature_on(inp, SCTP_PCB_FLAGS_NEEDS_MAPPED_V4)) {
len = sizeof(struct sockaddr_in6);
in6_sin_2_v4mapsin6(&control->whoFrom->ro._l_addr.sin,
&store.sin6);
store.sin6.sin6_port = control->port_from;
} else {
len = sizeof(struct sockaddr_in);
store.sin = control->whoFrom->ro._l_addr.sin;
store.sin.sin_port = control->port_from;
}
#else
len = sizeof(struct sockaddr_in);
store.sin = control->whoFrom->ro._l_addr.sin;
store.sin.sin_port = control->port_from;
#endif
Small cleanup of the code. No functional change (in FreeBSD kernel). MFC after: 1 week.
2012-03-22 16:22:16 +00:00
break;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
#endif
Small cleanup of the code. No functional change (in FreeBSD kernel). MFC after: 1 week.
2012-03-22 16:22:16 +00:00
default:
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
len = 0;
Small cleanup of the code. No functional change (in FreeBSD kernel). MFC after: 1 week.
2012-03-22 16:22:16 +00:00
break;
}
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
memcpy(from, &store, min((size_t)fromlen, len));
#ifdef INET and INET6 consistently. This also fixes a bug, where it was done wrong. MFC after: 3 days
2012-07-15 11:04:49 +00:00
#ifdef INET6
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
Small cleanup of the code. No functional change (in FreeBSD kernel). MFC after: 1 week.
2012-03-22 16:22:16 +00:00
struct sockaddr_in6 lsa6, *from6;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
Small cleanup of the code. No functional change (in FreeBSD kernel). MFC after: 1 week.
2012-03-22 16:22:16 +00:00
from6 = (struct sockaddr_in6 *)from;
sctp_recover_scope_mac(from6, (&lsa6));
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
#endif
}
/* now copy out what data we can */
if (mp == NULL) {
/* copy out each mbuf in the chain up to length */
get_more_data:
m = control->data;
while (m) {
/* Move out all we can */
cp_len = (int)uio->uio_resid;
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
my_len = (int)SCTP_BUF_LEN(m);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (cp_len > my_len) {
/* not enough in this buf */
cp_len = my_len;
}
if (hold_rlock) {
SCTP_INP_READ_UNLOCK(inp);
hold_rlock = 0;
}
if (cp_len > 0)
error = uiomove(mtod(m, char *), cp_len, uio);
/* re-read */
if (inp->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE) {
goto release;
}
- Fix a memory overwrite when the mapping array is expanded, size of expansion was not taken int consideration. - Fix so vtag hash is 1 bigger so that it modulo's out correctly, avoids a panic when restart with right modulo happens. - do not dereference stcb when control->do_not_ref_stcb is set - Fix up packet logging to not often use a lock and also to add to options. - Fix some logging option duplication in the sctputil.h
2007-05-30 17:39:45 +00:00
if ((control->do_not_ref_stcb == 0) && stcb &&
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
stcb->asoc.state & SCTP_STATE_ABOUT_TO_BE_FREED) {
no_rcv_needed = 1;
}
if (error) {
/* error we are out of here */
goto release;
}
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
SCTP_INP_READ_LOCK(inp);
hold_rlock = 1;
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (cp_len == SCTP_BUF_LEN(m)) {
if ((SCTP_BUF_NEXT(m) == NULL) &&
(control->end_added)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
out_flags |= MSG_EOR;
Fix some CLANG warnings. One clang warning is left due to the fact that its bogus.. nam->sa_family will not change from AF_INET6 to AF_INET (but clang thinks it does ;-D)
2010-09-05 13:41:45 +00:00
if ((control->do_not_ref_stcb == 0) &&
(control->stcb != NULL) &&
((control->spec_flags & M_NOTIFICATION) == 0))
Fixes cut and paste bug using wrong pointer reference.
2007-04-23 00:51:49 +00:00
control->stcb->asoc.strmin[control->sinfo_stream].delivery_started = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (control->spec_flags & M_NOTIFICATION) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
out_flags |= MSG_NOTIFICATION;
}
/* we ate up the mbuf */
if (in_flags & MSG_PEEK) {
/* just looking */
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
m = SCTP_BUF_NEXT(m);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
copied_so_far += cp_len;
} else {
/* dispose of the mbuf */
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(&so->so_rcv,
control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBFREE, SCTP_BUF_LEN(m));
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_sbfree(control, stcb, &so->so_rcv, m);
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(&so->so_rcv,
control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBRESULT, 0);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
copied_so_far += cp_len;
freed_so_far += cp_len;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
freed_so_far += MSIZE;
- added pre-checks to the bindx call. - use proper tick gathering macro instead of ticks directly. - Placed reasonable boundaries on sets that a user can do that are converted to ticks from ms. - Fix CMT_PF to always check to be sure CMT is on. - Fix ticks use of CMT_PF. - put back code to allow asconfs to be queued while INITs are in flight and before the assoc is established. - During window probes, an ack'd packet might be left with the window probe mark on it causing it to be retransmitted. Change so that the flight decrease macro clears the window_probe mark. - Additional logging flight size/reading and ASOC LOG. This is only enabled if you manually insert things into opt_sctp.h since its a set of debug code only. - Found an interesting SMP race in the way data was appended which could cause a reader to lose a part of a message, had to reorder when we marked the message was complete to after the data was appended. - bug in ADD-IP for the subset bound socket case when the peer has only one address - fix ASCONF implicit success/error handling case - proper support of jails in Freebsd 6> - copy out the timeval for the 64 bit sparc world on cookie-echo alignment error crashes without this). Approved by: re(Ken Smith)
2007-07-17 20:58:26 +00:00
atomic_subtract_int(&control->length, cp_len);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
control->data = sctp_m_free(m);
m = control->data;
/*
* been through it all, must hold sb
* lock ok to null tail
*/
if (control->data == NULL) {
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
#ifdef INVARIANTS
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if ((control->end_added == 0) ||
(TAILQ_NEXT(control, next) == NULL)) {
/*
* If the end is not
* added, OR the
* next is NOT null
* we MUST have the
* lock.
*/
if (mtx_owned(&inp->inp_rdata_mtx) == 0) {
panic("Hmm we don't own the lock?");
}
}
#endif
control->tail_mbuf = NULL;
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
#ifdef INVARIANTS
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if ((control->end_added) && ((out_flags & MSG_EOR) == 0)) {
panic("end_added, nothing left and no MSG_EOR");
}
#endif
}
}
} else {
/* Do we need to trim the mbuf? */
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
if (control->spec_flags & M_NOTIFICATION) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
out_flags |= MSG_NOTIFICATION;
}
if ((in_flags & MSG_PEEK) == 0) {
a) macro-ization of all mbuf and random number access plus timers. This makes the code more portable and able to change out the mbuf or timer system used more easily ;-) b) removal of all use of pkt-hdr's until only the places we need them (before ip_output routines). c) remove a bunch of code not needed due to <b> aka worrying about pkthdr's :-) d) There was one last reorder problem it looks where if a restart occur's and we release and relock (at the point where we setup our alias vtag) we would end up possibly getting the wrong TSN in place. The code that fixed the TSN's just needed to be shifted around BEFORE the release of the lock.. also code that set the state (since this also could contribute). Approved by: gnn
2006-12-29 20:21:42 +00:00
SCTP_BUF_RESV_UF(m, cp_len);
SCTP_BUF_LEN(m) -= cp_len;
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(&so->so_rcv, control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBFREE, cp_len);
}
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
atomic_subtract_int(&so->so_rcv.sb_cc, cp_len);
- Fix a memory overwrite when the mapping array is expanded, size of expansion was not taken int consideration. - Fix so vtag hash is 1 bigger so that it modulo's out correctly, avoids a panic when restart with right modulo happens. - do not dereference stcb when control->do_not_ref_stcb is set - Fix up packet logging to not often use a lock and also to add to options. - Fix some logging option duplication in the sctputil.h
2007-05-30 17:39:45 +00:00
if ((control->do_not_ref_stcb == 0) &&
stcb) {
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
atomic_subtract_int(&stcb->asoc.sb_cc, cp_len);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
copied_so_far += cp_len;
freed_so_far += cp_len;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
freed_so_far += MSIZE;
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(&so->so_rcv, control->do_not_ref_stcb ? NULL : stcb,
SCTP_LOG_SBRESULT, 0);
}
- added pre-checks to the bindx call. - use proper tick gathering macro instead of ticks directly. - Placed reasonable boundaries on sets that a user can do that are converted to ticks from ms. - Fix CMT_PF to always check to be sure CMT is on. - Fix ticks use of CMT_PF. - put back code to allow asconfs to be queued while INITs are in flight and before the assoc is established. - During window probes, an ack'd packet might be left with the window probe mark on it causing it to be retransmitted. Change so that the flight decrease macro clears the window_probe mark. - Additional logging flight size/reading and ASOC LOG. This is only enabled if you manually insert things into opt_sctp.h since its a set of debug code only. - Found an interesting SMP race in the way data was appended which could cause a reader to lose a part of a message, had to reorder when we marked the message was complete to after the data was appended. - bug in ADD-IP for the subset bound socket case when the peer has only one address - fix ASCONF implicit success/error handling case - proper support of jails in Freebsd 6> - copy out the timeval for the 64 bit sparc world on cookie-echo alignment error crashes without this). Approved by: re(Ken Smith)
2007-07-17 20:58:26 +00:00
atomic_subtract_int(&control->length, cp_len);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
} else {
copied_so_far += cp_len;
}
}
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
if ((out_flags & MSG_EOR) || (uio->uio_resid == 0)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
break;
}
if (((stcb) && (in_flags & MSG_PEEK) == 0) &&
(control->do_not_ref_stcb == 0) &&
(freed_so_far >= rwnd_req)) {
sctp_user_rcvd(stcb, &freed_so_far, hold_rlock, rwnd_req);
}
} /* end while(m) */
/*
* At this point we have looked at it all and we either have
* a MSG_EOR/or read all the user wants... <OR>
* control->length == 0.
*/
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
if ((out_flags & MSG_EOR) && ((in_flags & MSG_PEEK) == 0)) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* we are done with this control */
if (control->length == 0) {
if (control->data) {
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
#ifdef INVARIANTS
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
panic("control->data not null at read eor?");
#else
Two major items here: - All printf that was surrounded by #ifdef SCTP_DEBUG moves to a macro that does all of this. This removes all printfs from the code and makes the code more portable and easier to read. - Static Analysis (cisco) - found a few bugs, but mostly we add checks for NULL pointers and such to make the tool happy. We now pass the Cisco SA tools checks except for where it does not understand tailq/lists. We still need to look at the coverity tools output too (this is like the cisco SA tool) and see if it wants us to fix any other items. Hopefully this will be the last major churn in the code other than bug fixes.
2007-05-09 13:30:06 +00:00
SCTP_PRINTF("Strange, data left in the control buffer .. invarients would panic?\n");
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_m_freem(control->data);
control->data = NULL;
#endif
}
done_with_control:
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
if (hold_rlock == 0) {
SCTP_INP_READ_LOCK(inp);
hold_rlock = 1;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
TAILQ_REMOVE(&inp->read_queue, control, next);
/* Add back any hiddend data */
if (control->held_length) {
held_length = 0;
control->held_length = 0;
wakeup_read_socket = 1;
}
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
if (control->aux_data) {
sctp_m_free(control->aux_data);
control->aux_data = NULL;
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
no_rcv_needed = control->do_not_ref_stcb;
sctp_free_remote_addr(control->whoFrom);
control->data = NULL;
This is work done by Michael Tuexen and myself at the IETF. This adds the new I-Data (Interleaved Data) message. This allows a user to be able to have complete freedom from Head Of Line blocking that was previously there due to the in-ability to send multiple large messages without the TSN's being in sequence. The code as been tested with Michaels various packet drill scripts as well as inter-networking between the IETF's location in Argentina and Germany.
2016-04-07 09:10:34 +00:00
if (control->on_strm_q) {
panic("About to free ctl:%p so:%p and its in %d",
control, so, control->on_strm_q);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_free_a_readq(stcb, control);
control = NULL;
- Fix a memory overwrite when the mapping array is expanded, size of expansion was not taken int consideration. - Fix so vtag hash is 1 bigger so that it modulo's out correctly, avoids a panic when restart with right modulo happens. - do not dereference stcb when control->do_not_ref_stcb is set - Fix up packet logging to not often use a lock and also to add to options. - Fix some logging option duplication in the sctputil.h
2007-05-30 17:39:45 +00:00
if ((freed_so_far >= rwnd_req) &&
(no_rcv_needed == 0))
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
sctp_user_rcvd(stcb, &freed_so_far, hold_rlock, rwnd_req);
} else {
/*
* The user did not read all of this
* message, turn off the returned MSG_EOR
* since we are leaving more behind on the
* control to read.
*/
1) Fixes on a number of different collision case LOR's. 2) Fix all "magic numbers" to be constants. 3) A collision case that would generate two associations to the same peer due to a missing lock is fixed. 4) Added tracking of where timers are stopped. Approved by: gnn
2006-12-14 17:02:55 +00:00
#ifdef INVARIANTS
- Fix a memory overwrite when the mapping array is expanded, size of expansion was not taken int consideration. - Fix so vtag hash is 1 bigger so that it modulo's out correctly, avoids a panic when restart with right modulo happens. - do not dereference stcb when control->do_not_ref_stcb is set - Fix up packet logging to not often use a lock and also to add to options. - Fix some logging option duplication in the sctputil.h
2007-05-30 17:39:45 +00:00
if (control->end_added &&
(control->data == NULL) &&
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
(control->tail_mbuf == NULL)) {
panic("Gak, control->length is corrupt?");
}
#endif
no_rcv_needed = control->do_not_ref_stcb;
out_flags &= ~MSG_EOR;
}
}
if (out_flags & MSG_EOR) {
goto release;
}
if ((uio->uio_resid == 0) ||
Address some warnings which showed up on the userland version. MFC after: 1 week
2013-12-27 13:07:00 +00:00
((in_eeor_mode) &&
(copied_so_far >= (uint32_t) max(so->so_rcv.sb_lowat, 1)))) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
goto release;
}
/*
* If I hit here the receiver wants more and this message is
* NOT done (pd-api). So two questions. Can we block? if not
* we are done. Did the user NOT set MSG_WAITALL?
*/
if (block_allowed == 0) {
goto release;
}
/*
* We need to wait for more data a few things: - We don't
* sbunlock() so we don't get someone else reading. - We
* must be sure to account for the case where what is added
* is NOT to our control when we wakeup.
*/
/*
* Do we need to tell the transport a rwnd update might be
* needed before we go to sleep?
*/
if (((stcb) && (in_flags & MSG_PEEK) == 0) &&
((freed_so_far >= rwnd_req) &&
(control->do_not_ref_stcb == 0) &&
(no_rcv_needed == 0))) {
sctp_user_rcvd(stcb, &freed_so_far, hold_rlock, rwnd_req);
}
wait_some_more:
- Macroizes the V6ONLY flag check. - Added a short time wait (not used yet) constant - Corrected the type of the crc32c table (it was unsigned long and really is a uint32_t - Got rid of the user of MHeaders until they are truely needed by lower layers. - Fixed an initialization problem in the readq structure (ordering was off). - Found yet another collision bug when the random number generator returns two numbers on one side (during a collision) that are the same. Also added some tracking of cookies that will go away when we know that we have the last collision bug gone. - Fixed an init bug for book_size_scale, that was causing Early FR code to run when it should not. - Fixed a flight size tracking bug that was associated with Early FR but due to above bug also effected all FR's - Fixed it so Max Burst also will apply to Fast Retransmit. - Fixed a bug in the temporary logging code that allowed a static log array overflow - hashinit_flags is now used. - Two last mcopym's were converted to the macro sctp_m_copym that has always been used by all other places - macro sctp_m_copym was converted to upper case. - We now validate sinfo_flags on input (we did not before). - Fixed a bug that prevented a user from sending data and immediately shuting down with one send operation. - Moved to use hashdestroy instead of free() in our macros. - Fixed an init problem in our timed_wait vtag where we did not fully initialize our time-wait blocks. - Timer stops were re-positioned. - A pcb cleanup method was added, however this probably will not be used in BSD.. unless we make module loadable protocols - I think this fixes the mysterious timer bug.. it was a ordering of locks problem in the way we did timers. It now conforms to the timeout(9) manual (except for the _drain part, we had to do this a different way due to locks). - Fixed error return code so we get either CONNREUSED or CONNRESET depending on where one is in progression - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. - Purged an unused clone macro. - Fixed a read erro code issue where we were NOT getting the proper error when the connection was reset. Approved by: gnn
2007-01-15 15:12:10 +00:00
if (so->so_rcv.sb_state & SBS_CANTRCVMORE) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
goto release;
}
if (inp->sctp_flags & SCTP_PCB_FLAGS_SOCKET_GONE)
goto release;
if (hold_rlock == 1) {
SCTP_INP_READ_UNLOCK(inp);
hold_rlock = 0;
}
if (hold_sblock == 0) {
SOCKBUF_LOCK(&so->so_rcv);
hold_sblock = 1;
}
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
if ((copied_so_far) && (control->length == 0) &&
* Fix some race condition in SACK/NR-SACK processing. * Fix handling of mapping arrays when draining mbufs or processing FORWARD-TSN chunks. * Cleanup code (no duplicate code anymore for SACKs and NR-SACKs). Part of this code was developed together with rrs. MFC after: 2 weeks.
2010-04-03 15:40:14 +00:00
(sctp_is_feature_on(inp, SCTP_PCB_FLAGS_FRAG_INTERLEAVE))) {
- send call has a reference to uio->uio_resid in the recent send code, but uio may be NULL on sendfile calls. Change to use sndlen variable. - EMSGSIZE is not being returned in non-blocking mode and needs a small tweak to look if the msg would ever fit when returning EWOULDBLOCK. - FWD-TSN has a bug in stream processing which could cause a panic. This is a follow on to the codenomicon fix. - PDAPI level 1 and 2 do not work unless the reader gets his returned buffer full. Fix so we can break out when at level 1 or 2. - Fix fast-handoff features to copy across properly on accepted sockets - Fix sctp_peeloff() system call when no true system call exists to screen arguments for errors. In cases where a real system call exists the system call itself does this. - Fix raddr leak in recent add-ip code change for bundled asconfs (even when non-bundled asconfs are received) - Make sure ipi_addr lock is held when walking global addr list. Need to change this lock type to a rwlock(). - Add don't wake flag on both input and output when the socket is closing. - When deleting an address verify the interface is correct before allowing the delete to process. This protects panda and unnumbered. - Clean up old sysctl stuff and get rid of the old Open/Net BSD structures. - Add a function to watch the ranges in the sysctl sets. - When appending in the reassembly queue, validate that the assoc has not gone to about to be freed. If so (in the middle) abort out. Note this especially effects MAC I think due to the lock/unlock they do (or with LOCK testing in place). - Netstat patch to get rid of warnings. - Make sure that no data gets queued to inactive/unconfirmed destinations. This especially effect CMT but also makes a impact on regular SCTP as well. - During init collision when we detect seq number out of sync we need to treat it like Case C and discard the cookie (no invarient needed here). - Atomic access to the random store. - When we declare a vtag good, we need to shove it into the time wait hash to prevent further use. When the tag is put into the assoc hash, we need to remove it from the twait hash (where it will surely be). This prevents duplicate tag assignments. - Move decr-ref count to better protect sysctl out of data. - ltrace error corrections in sctp6_usrreq.c - Add hook for interface up/down to be sent to us. - Make sysctl() exported structures independent of processor architecture. - Fix route and src addr cache clearing for delete address case. - Make sure address marked SCTP_DEL_IP_ADDRESS is never selected as src addr. - in icmp handling fixed so we actually look at the icmp codes to figure out what to do. - Modified mobility code. Reception of DELETE IP ADDRESS for a primary destination and SET PRIMARY for a new primary destination is used for retransmission trigger to the new primary destination. Also, in this case, destination of chunks in send_queue are changed to the new primary destination. - Fix so that we disallow sending by mbuf to ever have EEOR mode set upon it. Approved by: re@freebsd.org (B Mah)
2007-09-08 17:48:46 +00:00
goto release;
}
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
if (so->so_rcv.sb_cc <= control->held_length) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
error = sbwait(&so->so_rcv);
if (error) {
goto release;
}
control->held_length = 0;
}
if (hold_sblock) {
SOCKBUF_UNLOCK(&so->so_rcv);
hold_sblock = 0;
}
if (control->length == 0) {
/* still nothing here */
if (control->end_added == 1) {
/* he aborted, or is done i.e.did a shutdown */
out_flags |= MSG_EOR;
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
if (control->pdapi_aborted) {
Two bugs: - Locks were not being unlocked when an invalid size chunk is sent in. - When a notification comes in, we cannot use it to look up the fragment interleave stream information since its not on a stream.
2007-05-06 00:01:17 +00:00
if ((control->do_not_ref_stcb == 0) && ((control->spec_flags & M_NOTIFICATION) == 0))
Fixes cut and paste bug using wrong pointer reference.
2007-04-23 00:51:49 +00:00
control->stcb->asoc.strmin[control->sinfo_stream].delivery_started = 0;
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
-Fixes first of all the getcred on IPv6 and V4. The copy's were incorrect and so was the locking. -A bug was also found that would create a race and panic when an abort arrived on a socket being read from. -Also fix the reader to get MSG_TRUNC when a partial delivery is aborted. -Also addresses a couple of coverity caught error path memory leaks and a couple of other valid complaints Approved by: gnn
2006-11-08 00:21:13 +00:00
out_flags |= MSG_TRUNC;
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
} else {
Two bugs: - Locks were not being unlocked when an invalid size chunk is sent in. - When a notification comes in, we cannot use it to look up the fragment interleave stream information since its not on a stream.
2007-05-06 00:01:17 +00:00
if ((control->do_not_ref_stcb == 0) && ((control->spec_flags & M_NOTIFICATION) == 0))
Fixes cut and paste bug using wrong pointer reference.
2007-04-23 00:51:49 +00:00
control->stcb->asoc.strmin[control->sinfo_stream].delivery_started = 0;
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
goto done_with_control;
}
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
if (so->so_rcv.sb_cc > held_length) {
control->held_length = so->so_rcv.sb_cc;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
held_length = 0;
}
goto wait_some_more;
} else if (control->data == NULL) {
Tons of fixes to get all the 64bit issues removed. This also moves two 16 bit int's to become 32 bit values so we do not have to use atomic_add_16. Most of the changes are %p, casts and other various nasty's that were in the orignal code base. With this commit my machine will now do a build universe.. however I as yet have not tested on a 64bit machine .. it may not work :-(
2006-11-05 13:25:18 +00:00
/*
* we must re-sync since data is probably being
* added
*/
SCTP_INP_READ_LOCK(inp);
if ((control->length > 0) && (control->data == NULL)) {
/*
* big trouble.. we have the lock and its
* corrupt?
*/
- optimize the initialization of the SB max variables. - Missing lock when sending data and moving it to the outqueue. - If a mbuf alloc fails during moving to outqueue the reassembly of the old mbuf chain was incorrect. - some_taken becomes a counter in sctputil.c instead of a set to 1. - Fix a panic to be only under invarients and have a proper recovery. - msg_flags needed to be set.to the value collected not or'd. MFC after: 1 week
2007-12-06 00:22:55 +00:00
#ifdef INVARIANTS
A couple of minor changes that I missed that Michael had done, most noted in these is the change to non-strict ordering for incoming data (this will make pkt-drill test 14 fail but its expected).
2016-04-07 09:34:41 +00:00
panic("Impossible data==NULL length !=0");
- optimize the initialization of the SB max variables. - Missing lock when sending data and moving it to the outqueue. - If a mbuf alloc fails during moving to outqueue the reassembly of the old mbuf chain was incorrect. - some_taken becomes a counter in sctputil.c instead of a set to 1. - Fix a panic to be only under invarients and have a proper recovery. - msg_flags needed to be set.to the value collected not or'd. MFC after: 1 week
2007-12-06 00:22:55 +00:00
#endif
out_flags |= MSG_EOR;
out_flags |= MSG_TRUNC;
control->length = 0;
SCTP_INP_READ_UNLOCK(inp);
goto done_with_control;
Tons of fixes to get all the 64bit issues removed. This also moves two 16 bit int's to become 32 bit values so we do not have to use atomic_add_16. Most of the changes are %p, casts and other various nasty's that were in the orignal code base. With this commit my machine will now do a build universe.. however I as yet have not tested on a 64bit machine .. it may not work :-(
2006-11-05 13:25:18 +00:00
}
SCTP_INP_READ_UNLOCK(inp);
/* We will fall around to get more data */
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
goto get_more_data;
} else {
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
/*-
* Give caller back the mbuf chain,
* store in uio_resid the length
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
*/
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
wakeup_read_socket = 0;
if ((control->end_added == 0) ||
(TAILQ_NEXT(control, next) == NULL)) {
/* Need to get rlock */
if (hold_rlock == 0) {
SCTP_INP_READ_LOCK(inp);
hold_rlock = 1;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
}
if (control->end_added) {
out_flags |= MSG_EOR;
Address issues found by clang. While there, fix also some style issues. MFC after: 3 months.
2011-12-27 10:16:24 +00:00
if ((control->do_not_ref_stcb == 0) &&
(control->stcb != NULL) &&
((control->spec_flags & M_NOTIFICATION) == 0))
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
control->stcb->asoc.strmin[control->sinfo_stream].delivery_started = 0;
}
if (control->spec_flags & M_NOTIFICATION) {
out_flags |= MSG_NOTIFICATION;
}
uio->uio_resid = control->length;
*mp = control->data;
m = control->data;
while (m) {
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(&so->so_rcv,
control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBFREE, SCTP_BUF_LEN(m));
}
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
sctp_sbfree(control, stcb, &so->so_rcv, m);
freed_so_far += SCTP_BUF_LEN(m);
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
freed_so_far += MSIZE;
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_SB_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
sctp_sblog(&so->so_rcv,
control->do_not_ref_stcb ? NULL : stcb, SCTP_LOG_SBRESULT, 0);
}
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
m = SCTP_BUF_NEXT(m);
}
control->data = control->tail_mbuf = NULL;
control->length = 0;
if (out_flags & MSG_EOR) {
/* Done with this control */
goto done_with_control;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
}
release:
if (hold_rlock == 1) {
SCTP_INP_READ_UNLOCK(inp);
hold_rlock = 0;
}
sblock() implements a sleep lock by interlocking SB_WANT and SB_LOCK flags on each socket buffer with the socket buffer's mutex. This sleep lock is used to serialize I/O on sockets in order to prevent I/O interlacing. This change replaces the custom sleep lock with an sx(9) lock, which results in marginally better performance, better handling of contention during simultaneous socket I/O across multiple threads, and a cleaner separation between the different layers of locking in socket buffers. Specifically, the socket buffer mutex is now solely responsible for serializing simultaneous operation on the socket buffer data structure, and not for I/O serialization. While here, fix two historic bugs: (1) a bug allowing I/O to be occasionally interlaced during long I/O operations (discovere by Isilon). (2) a bug in which failed non-blocking acquisition of the socket buffer I/O serialization lock might be ignored (discovered by sam). SCTP portion of this patch submitted by rrs.
2007-05-03 14:42:42 +00:00
if (hold_sblock == 1) {
SOCKBUF_UNLOCK(&so->so_rcv);
hold_sblock = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
sbunlock(&so->so_rcv);
sblock() implements a sleep lock by interlocking SB_WANT and SB_LOCK flags on each socket buffer with the socket buffer's mutex. This sleep lock is used to serialize I/O on sockets in order to prevent I/O interlacing. This change replaces the custom sleep lock with an sx(9) lock, which results in marginally better performance, better handling of contention during simultaneous socket I/O across multiple threads, and a cleaner separation between the different layers of locking in socket buffers. Specifically, the socket buffer mutex is now solely responsible for serializing simultaneous operation on the socket buffer data structure, and not for I/O serialization. While here, fix two historic bugs: (1) a bug allowing I/O to be occasionally interlaced during long I/O operations (discovere by Isilon). (2) a bug in which failed non-blocking acquisition of the socket buffer I/O serialization lock might be ignored (discovered by sam). SCTP portion of this patch submitted by rrs.
2007-05-03 14:42:42 +00:00
sockbuf_lock = 0;
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
release_unlocked:
if (hold_sblock) {
SOCKBUF_UNLOCK(&so->so_rcv);
hold_sblock = 0;
}
if ((stcb) && (in_flags & MSG_PEEK) == 0) {
if ((freed_so_far >= rwnd_req) &&
(control && (control->do_not_ref_stcb == 0)) &&
(no_rcv_needed == 0))
sctp_user_rcvd(stcb, &freed_so_far, hold_rlock, rwnd_req);
}
out:
The flags value was not always being copied out in the recv routine like it should be. Obtained from: Michael Tuexen
2008-10-18 15:56:52 +00:00
if (msg_flags) {
*msg_flags = out_flags;
}
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
if (((out_flags & MSG_EOR) == 0) &&
((in_flags & MSG_PEEK) == 0) &&
(sinfo) &&
Add support for the newly added SCTP API. In particular add support for: * SCTP_SNDINFO, SCTP_PRINFO, SCTP_AUTHINFO, SCTP_DSTADDRV4, and SCTP_DSTADDRV6 cmsgs. * SCTP_NXTINFO and SCTP_RCVINFO cmgs. * SCTP_EVENT, SCTP_RECVRCVINFO, SCTP_RECVNXTINFO and SCTP_DEFAULT_SNDINFO socket option. * Special association ids (SCTP_FUTURE_ASSOC, ...) * sctp_recvv() and sctp_sendv() functions. MFC after: 1 month.
2011-06-15 23:50:27 +00:00
(sctp_is_feature_on(inp, SCTP_PCB_FLAGS_EXT_RCVINFO) ||
sctp_is_feature_on(inp, SCTP_PCB_FLAGS_RECVNXTINFO))) {
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
struct sctp_extrcvinfo *s_extra;
s_extra = (struct sctp_extrcvinfo *)sinfo;
Fix typos in field names of struct sctp_extrcvinfo. Provide defines to allow applications to compile. Thanks to Jens Hoelscher for making me aware of the typos. MFC after: 1 week
2015-11-06 13:08:16 +00:00
s_extra->serinfo_next_flags = SCTP_NO_NEXT_MSG;
- Somehow the disable fragment option got lost. We could set/clear it but would not do it. Now we will. - Moved to latest socket api for extended sndrcv info struct. - Moved to support all new levels of fragment interleave.
2007-04-22 11:06:27 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (hold_rlock == 1) {
SCTP_INP_READ_UNLOCK(inp);
}
if (hold_sblock) {
SOCKBUF_UNLOCK(&so->so_rcv);
}
sblock() implements a sleep lock by interlocking SB_WANT and SB_LOCK flags on each socket buffer with the socket buffer's mutex. This sleep lock is used to serialize I/O on sockets in order to prevent I/O interlacing. This change replaces the custom sleep lock with an sx(9) lock, which results in marginally better performance, better handling of contention during simultaneous socket I/O across multiple threads, and a cleaner separation between the different layers of locking in socket buffers. Specifically, the socket buffer mutex is now solely responsible for serializing simultaneous operation on the socket buffer data structure, and not for I/O serialization. While here, fix two historic bugs: (1) a bug allowing I/O to be occasionally interlaced during long I/O operations (discovere by Isilon). (2) a bug in which failed non-blocking acquisition of the socket buffer I/O serialization lock might be ignored (discovered by sam). SCTP portion of this patch submitted by rrs.
2007-05-03 14:42:42 +00:00
if (sockbuf_lock) {
sbunlock(&so->so_rcv);
}
Tons of fixes to get all the 64bit issues removed. This also moves two 16 bit int's to become 32 bit values so we do not have to use atomic_add_16. Most of the changes are %p, casts and other various nasty's that were in the orignal code base. With this commit my machine will now do a build universe.. however I as yet have not tested on a 64bit machine .. it may not work :-(
2006-11-05 13:25:18 +00:00
if (freecnt_applied) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/*
* The lock on the socket buffer protects us so the free
* code will stop. But since we used the socketbuf lock and
* the sender uses the tcb_lock to increment, we need to use
* the atomic add to the refcnt.
*/
Tons of fixes to get all the 64bit issues removed. This also moves two 16 bit int's to become 32 bit values so we do not have to use atomic_add_16. Most of the changes are %p, casts and other various nasty's that were in the orignal code base. With this commit my machine will now do a build universe.. however I as yet have not tested on a 64bit machine .. it may not work :-(
2006-11-05 13:25:18 +00:00
if (stcb == NULL) {
- Make strict-sacks be the default. - Change it so that without INVARIANTs there are no panics in SCTP. - sctp_timer changes so that we have a recovery mechanism when the sent list is out of order.
2008-08-28 09:44:07 +00:00
#ifdef INVARIANTS
Tons of fixes to get all the 64bit issues removed. This also moves two 16 bit int's to become 32 bit values so we do not have to use atomic_add_16. Most of the changes are %p, casts and other various nasty's that were in the orignal code base. With this commit my machine will now do a build universe.. however I as yet have not tested on a 64bit machine .. it may not work :-(
2006-11-05 13:25:18 +00:00
panic("stcb for refcnt has gone NULL?");
- Make strict-sacks be the default. - Change it so that without INVARIANTs there are no panics in SCTP. - sctp_timer changes so that we have a recovery mechanism when the sent list is out of order.
2008-08-28 09:44:07 +00:00
goto stage_left;
#else
goto stage_left;
#endif
Tons of fixes to get all the 64bit issues removed. This also moves two 16 bit int's to become 32 bit values so we do not have to use atomic_add_16. Most of the changes are %p, casts and other various nasty's that were in the orignal code base. With this commit my machine will now do a build universe.. however I as yet have not tested on a 64bit machine .. it may not work :-(
2006-11-05 13:25:18 +00:00
}
atomic_add_int(&stcb->asoc.refcnt, -1);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* Save the value back for next time */
stcb->freed_by_sorcv_sincelast = freed_so_far;
}
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_RECV_RWND_LOGGING_ENABLE) {
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
if (stcb) {
sctp_misc_ints(SCTP_SORECV_DONE,
freed_so_far,
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
(uint32_t) ((uio) ? (slen - uio->uio_resid) : slen),
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
stcb->asoc.my_rwnd,
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
so->so_rcv.sb_cc);
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
} else {
sctp_misc_ints(SCTP_SORECV_DONE,
freed_so_far,
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
(uint32_t) ((uio) ? (slen - uio->uio_resid) : slen),
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
0,
Do the renaming of sb_cc to sb_ccc in a way with less code changes by using a macro. This is an alternate approach to https://svnweb.freebsd.org/changeset/base/275326 which is easier to handle upstream. Discussed with: rrs, glebius
2014-12-02 20:29:29 +00:00
so->so_rcv.sb_cc);
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
- Make strict-sacks be the default. - Change it so that without INVARIANTs there are no panics in SCTP. - sctp_timer changes so that we have a recovery mechanism when the sent list is out of order.
2008-08-28 09:44:07 +00:00
stage_left:
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
if (wakeup_read_socket) {
sctp_sorwakeup(inp, so);
}
return (error);
}
#ifdef SCTP_MBUF_LOGGING
struct mbuf *
sctp_m_free(struct mbuf *m)
{
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_MBUF_LOGGING_ENABLE) {
Minimize the usage of SCTP_BUF_IS_EXTENDED. This should help Robert...
2015-01-10 20:49:57 +00:00
sctp_log_mb(m, SCTP_MBUF_IFREE);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
}
return (m_free(m));
}
void
sctp_m_freem(struct mbuf *mb)
{
while (mb != NULL)
mb = sctp_m_free(mb);
}
#endif
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
int
sctp_dynamic_set_primary(struct sockaddr *sa, uint32_t vrf_id)
{
/*
* Given a local address. For all associations that holds the
* address, request a peer-set-primary.
*/
struct sctp_ifa *ifa;
struct sctp_laddr *wi;
ifa = sctp_find_ifa_by_addr(sa, vrf_id, 0);
if (ifa == NULL) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(NULL, NULL, NULL, SCTP_FROM_SCTPUTIL, EADDRNOTAVAIL);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
return (EADDRNOTAVAIL);
}
/*
* Now that we have the ifa we must awaken the iterator with this
* message.
*/
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
wi = SCTP_ZONE_GET(SCTP_BASE_INFO(ipi_zone_laddr), struct sctp_laddr);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
if (wi == NULL) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(NULL, NULL, NULL, SCTP_FROM_SCTPUTIL, ENOMEM);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
return (ENOMEM);
}
/* Now incr the count and int wi structure */
SCTP_INCR_LADDR_COUNT();
bzero(wi, sizeof(*wi));
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
(void)SCTP_GETTIME_TIMEVAL(&wi->start_time);
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
wi->ifa = ifa;
wi->action = SCTP_SET_PRIM_ADDR;
atomic_add_int(&ifa->refcount, 1);
/* Now add it to the work queue */
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
SCTP_WQ_ADDR_LOCK();
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
/*
* Should this really be a tailq? As it is we will process the
* newest first :-0
*/
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
LIST_INSERT_HEAD(&SCTP_BASE_INFO(addr_wq), wi, sctp_nxt_addr);
This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had multiple LOR's and other issues. MFC after: 3 days
2010-05-16 17:03:56 +00:00
SCTP_WQ_ADDR_UNLOCK();
- Sysctl's move to seperate file - moved away from ifn/ifa access to sctp_ifa/sctp_ifn built and managed by the add-ip code. - cleaned up add-ip code to use the iterator - made iterator be a thread, which enables auto-asconf now. - rewrote and cleaned up source address selection (also made it use new structures). - Fixed a couple of memory leaks. - DACK now settable as to how many packets to delay as well as time. - connectx() to latest socket API, new associd arg. - Fixed issue with revoking and loosing potential to send when we inflate the flight size. We now inflate the cwnd too and deflate it later when the revoked chunk is sent or acked. - Got rid of some temp debug code - src addr selection moved to a common file (sctp_output.c) - Support for simple VRF's (we have support for multi-vfr via compile switch that is scrubbed from BSD but we won't need multi-vrf until we first get VRF :-D) - Rest of mib work for address information now done - Limit number of addresses in INIT/INIT-ACK to a #def (30). Reviewed by: gnn
2007-03-15 11:27:14 +00:00
sctp_timer_start(SCTP_TIMER_TYPE_ADDR_WQ,
(struct sctp_inpcb *)NULL,
(struct sctp_tcb *)NULL,
(struct sctp_nets *)NULL);
return (0);
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
int
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
sctp_soreceive(struct socket *so,
struct sockaddr **psa,
struct uio *uio,
struct mbuf **mp0,
struct mbuf **controlp,
int *flagsp)
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
{
int error, fromlen;
uint8_t sockbuf[256];
struct sockaddr *from;
struct sctp_extrcvinfo sinfo;
int filling_sinfo = 1;
struct sctp_inpcb *inp;
inp = (struct sctp_inpcb *)so->so_pcb;
/* pickup the assoc we are reading from */
if (inp == NULL) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
return (EINVAL);
}
Add support for the newly added SCTP API. In particular add support for: * SCTP_SNDINFO, SCTP_PRINFO, SCTP_AUTHINFO, SCTP_DSTADDRV4, and SCTP_DSTADDRV6 cmsgs. * SCTP_NXTINFO and SCTP_RCVINFO cmgs. * SCTP_EVENT, SCTP_RECVRCVINFO, SCTP_RECVNXTINFO and SCTP_DEFAULT_SNDINFO socket option. * Special association ids (SCTP_FUTURE_ASSOC, ...) * sctp_recvv() and sctp_sendv() functions. MFC after: 1 month.
2011-06-15 23:50:27 +00:00
if ((sctp_is_feature_off(inp, SCTP_PCB_FLAGS_RECVDATAIOEVNT) &&
sctp_is_feature_off(inp, SCTP_PCB_FLAGS_RECVRCVINFO) &&
sctp_is_feature_off(inp, SCTP_PCB_FLAGS_RECVNXTINFO)) ||
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
(controlp == NULL)) {
/* user does not want the sndrcv ctl */
filling_sinfo = 0;
}
if (psa) {
from = (struct sockaddr *)sockbuf;
fromlen = sizeof(sockbuf);
from->sa_len = 0;
} else {
from = NULL;
fromlen = 0;
}
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
if (filling_sinfo) {
memset(&sinfo, 0, sizeof(struct sctp_extrcvinfo));
}
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
error = sctp_sorecvmsg(so, uio, mp0, from, fromlen, flagsp,
(struct sctp_sndrcvinfo *)&sinfo, filling_sinfo);
Initialize SCTP cmsg's and notification's buffer before copying out to userland. Submitted by: tuexen Security: CVE-2014-3953 Security: FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
if (controlp != NULL) {
Ok, here it is, we finally add SCTP to current. Note that this work is not just mine, but it is also the works of Peter Lei and Michael Tuexen. They both are my two key other developers working on the project.. and they need ata-boy's too: **** peterlei@cisco.com tuexen@fh-muenster.de **** I did do a make sysent which updated the syscall's and sysproto.. I hope that is correct... without it you don't build since we have new syscalls for SCTP :-0 So go out and look at the NOTES, add option SCTP (make sure inet and inet6 are present too) and play with SCTP. I will see about comitting some test tools I have after I figure out where I should place them. I also have a lib (libsctp.a) that adds some of the missing socketapi functions that I need to put into lib's.. I will talk to George about this :-) There may still be some 64 bit issues in here, none of us have a 64 bit processor to test with yet.. Michael may have a MAC but thats another beast too.. If you have a mac and want to use SCTP contact Michael he maintains a web site with a loadable module with this code :-) Reviewed by: gnn Approved by: gnn
2006-11-03 15:23:16 +00:00
/* copy back the sinfo in a CMSG format */
if (filling_sinfo)
*controlp = sctp_build_ctl_nchunk(inp,
(struct sctp_sndrcvinfo *)&sinfo);
else
*controlp = NULL;
}
if (psa) {
/* copy back the address info */
if (from && from->sa_len) {
*psa = sodupsockaddr(from, M_NOWAIT);
} else {
*psa = NULL;
}
}
return (error);
}
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
int
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
sctp_connectx_helper_add(struct sctp_tcb *stcb, struct sockaddr *addr,
int totaddr, int *error)
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
{
int added = 0;
int i;
struct sctp_inpcb *inp;
struct sockaddr *sa;
size_t incr = 0;
Make sure that SCTP rejects broadcast, multicast and wildcard addresses as remote addresses. Approved by: re MFC after: 1 month.
2011-09-15 08:49:54 +00:00
#ifdef INET
struct sockaddr_in *sin;
#endif
#ifdef INET6
struct sockaddr_in6 *sin6;
#endif
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
sa = addr;
inp = stcb->sctp_ep;
*error = 0;
for (i = 0; i < totaddr; i++) {
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
switch (sa->sa_family) {
#ifdef INET
case AF_INET:
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
incr = sizeof(struct sockaddr_in);
Make sure that SCTP rejects broadcast, multicast and wildcard addresses as remote addresses. Approved by: re MFC after: 1 month.
2011-09-15 08:49:54 +00:00
sin = (struct sockaddr_in *)sa;
if ((sin->sin_addr.s_addr == INADDR_ANY) ||
(sin->sin_addr.s_addr == INADDR_BROADCAST) ||
IN_MULTICAST(ntohl(sin->sin_addr.s_addr))) {
SCTP_LTRACE_ERR_RET(NULL, stcb, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
Some more debug info cleanup. MFC after: 3 days
2015-05-28 16:39:22 +00:00
(void)sctp_free_assoc(inp, stcb, SCTP_NORMAL_PROC,
SCTP_FROM_SCTPUTIL + SCTP_LOC_7);
Make sure that SCTP rejects broadcast, multicast and wildcard addresses as remote addresses. Approved by: re MFC after: 1 month.
2011-09-15 08:49:54 +00:00
*error = EINVAL;
goto out_now;
}
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
if (sctp_add_remote_addr(stcb, sa, NULL, SCTP_DONOT_SETSCOPE, SCTP_ADDR_IS_CONFIRMED)) {
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
/* assoc gone no un-lock */
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(NULL, stcb, NULL, SCTP_FROM_SCTPUTIL, ENOBUFS);
Some more debug info cleanup. MFC after: 3 days
2015-05-28 16:39:22 +00:00
(void)sctp_free_assoc(inp, stcb, SCTP_NORMAL_PROC,
SCTP_FROM_SCTPUTIL + SCTP_LOC_8);
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
*error = ENOBUFS;
goto out_now;
}
added++;
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
break;
#endif
#ifdef INET6
case AF_INET6:
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
incr = sizeof(struct sockaddr_in6);
Make sure that SCTP rejects broadcast, multicast and wildcard addresses as remote addresses. Approved by: re MFC after: 1 month.
2011-09-15 08:49:54 +00:00
sin6 = (struct sockaddr_in6 *)sa;
if (IN6_IS_ADDR_UNSPECIFIED(&sin6->sin6_addr) ||
IN6_IS_ADDR_MULTICAST(&sin6->sin6_addr)) {
SCTP_LTRACE_ERR_RET(NULL, stcb, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
Some more debug info cleanup. MFC after: 3 days
2015-05-28 16:39:22 +00:00
(void)sctp_free_assoc(inp, stcb, SCTP_NORMAL_PROC,
SCTP_FROM_SCTPUTIL + SCTP_LOC_9);
Make sure that SCTP rejects broadcast, multicast and wildcard addresses as remote addresses. Approved by: re MFC after: 1 month.
2011-09-15 08:49:54 +00:00
*error = EINVAL;
goto out_now;
}
The result of a joint work between rrs@ and myself at the IETF: * Decouple the path supervision using a separate HB timer per path. * Add support for potentially failed state. * Bring back RTO.min to 1 second. * Accept packets on IP-addresses already announced via an ASCONF * While there: do some cleanups. Approved by: re@ MFC after: 2 months.
2011-08-03 20:21:00 +00:00
if (sctp_add_remote_addr(stcb, sa, NULL, SCTP_DONOT_SETSCOPE, SCTP_ADDR_IS_CONFIRMED)) {
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
/* assoc gone no un-lock */
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(NULL, stcb, NULL, SCTP_FROM_SCTPUTIL, ENOBUFS);
Some more debug info cleanup. MFC after: 3 days
2015-05-28 16:39:22 +00:00
(void)sctp_free_assoc(inp, stcb, SCTP_NORMAL_PROC,
SCTP_FROM_SCTPUTIL + SCTP_LOC_10);
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
*error = ENOBUFS;
goto out_now;
}
added++;
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
break;
#endif
default:
break;
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
}
sa = (struct sockaddr *)((caddr_t)sa + incr);
}
out_now:
return (added);
}
struct sctp_tcb *
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
sctp_connectx_helper_find(struct sctp_inpcb *inp, struct sockaddr *addr,
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
unsigned int *totaddr,
unsigned int *num_v4, unsigned int *num_v6, int *error,
unsigned int limit, int *bad_addr)
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
{
struct sockaddr *sa;
struct sctp_tcb *stcb = NULL;
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
unsigned int incr, at, i;
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
at = incr = 0;
sa = addr;
*error = *num_v6 = *num_v4 = 0;
/* account and validate addresses */
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
for (i = 0; i < *totaddr; i++) {
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
switch (sa->sa_family) {
#ifdef INET
case AF_INET:
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
if (sa->sa_len != incr) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
*error = EINVAL;
*bad_addr = 1;
return (NULL);
}
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
(*num_v4) += 1;
incr = (unsigned int)sizeof(struct sockaddr_in);
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
break;
#endif
#ifdef INET6
case AF_INET6:
{
struct sockaddr_in6 *sin6;
sin6 = (struct sockaddr_in6 *)sa;
if (IN6_IS_ADDR_V4MAPPED(&sin6->sin6_addr)) {
/* Must be non-mapped for connectx */
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
*error = EINVAL;
*bad_addr = 1;
return (NULL);
}
if (sa->sa_len != incr) {
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
*error = EINVAL;
*bad_addr = 1;
return (NULL);
}
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
(*num_v6) += 1;
incr = (unsigned int)sizeof(struct sockaddr_in6);
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
break;
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
}
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#endif
default:
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
*totaddr = i;
/* we are done */
break;
}
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
if (i == *totaddr) {
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
break;
}
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
SCTP_INP_INCR_REF(inp);
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
stcb = sctp_findassociation_ep_addr(&inp, sa, NULL, NULL, NULL);
if (stcb != NULL) {
/* Already have or am bring up an association */
return (stcb);
- fixed autclose to not allow setting on 1-2-1 model. - bounded cookie-life to 1 second minimum in socket option set. - Delayed_ack_time becomes delayed_ack per new socket api document. - Improve port number selection, we now use low/high bounds and no chance of a endless loop. Only one call to random per bind as well. - fixes so set_peer_primary pre-screens addresses to be valid to this host. - maxseg did not allow setting on an assoc basis. We needed to thus track and use an association value instead of a inp value. - Fixed ep get of HB status to report back properly. - use settings flag to tell if assoc level hb is on off not the timer.. since the timer may still run if unconf address are present. - check for crazy ENABLE/DISABLE conditions. - set and get of pmtud (fixed path mtu) not always taking into account ovh. - Getting PMTU info on stcb only needs to return PMTUD_ENABLED if any net is doing PMTU discovery. - Panic or warning fixed to not do so when a valid ip frag is taking place. - sndrcvinfo appearing in both inp and stcb was full size, instead of the non-pad version. This saves about 92 bytes from each struct by carefully converting to use the smaller version. - one-2-one model get(maxseg) would always get ep value, never the tcb's value. - The delayed ack time could be under a tick, this fixes so it bounds it to at least 1 tick for platforms whos tick is more than a ms. - Fragment interleave level set to wrong default value. - Fragment interleave could not set level 0. - Defered stream reset was broken due to a guard check and ntohl issue. - Found two lock order reversals and fixed. - Tighten up address checking, if the user gives an address the sa_len had better be set properly. - Get asoc by assoc-id would return a locked tcb when it was asked not to if the tcb was in the restart hash. - sysctl to dig down and get more association details Reviewed by: gnn
2007-05-28 11:17:24 +00:00
} else {
SCTP_INP_DECR_REF(inp);
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
}
Improve compilation on windows 64-bit (for the userland stack). MFC after: 1 week
2016-03-27 10:04:25 +00:00
if ((at + incr) > limit) {
- More macros for OS compatabilty - PR-SCTP would ignore FWD-TSN's above a rwnd's worth of TSN's (1 byte msgs).. this left the peer hopelessly out of sync.. or an attacker. So now we abort the assoc. - New IFN hash, also rename hashes to match addr/ifn now that the vrf has multiple. - Do not enable SCTP_PCB_FLAGS_RECVDATAIOEVNT per default as defined in the Socket API ID. - Export MTU information via sysctl. - Vrf's need table id's. This is default for BSD, but may be other things later when BSD fully supports VRFs. - Additional stream reset bug (caught by cisco dev-test). - Additional validations for the address in sending a message (socket api). -------- and ----- - Fix association notifications not to give the active open side false notifications. - Fix so sendfile and SENDALL will work properly (missing flag to say socket sender is done). - Fix Bug that prevented COOKIES from being retransmitted. - Break out connectx into helper sub-models so that iox routines can reuse the helpers. - When an address is added during system init (non-dynamic mode) make sure that the "defer use" flag is not set. ** its compiling on XR now :-D ** Reviewed by: gnn
2007-05-08 00:21:05 +00:00
*totaddr = i;
break;
}
sa = (struct sockaddr *)((caddr_t)sa + incr);
}
return ((struct sctp_tcb *)NULL);
}
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
/*
* sctp_bindx(ADD) for one address.
* assumes all arguments are valid/checked by caller.
*/
void
sctp_bindx_add_address(struct socket *so, struct sctp_inpcb *inp,
struct sockaddr *sa, sctp_assoc_t assoc_id,
uint32_t vrf_id, int *error, void *p)
{
struct sockaddr *addr_touse;
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#if defined(INET) && defined(INET6)
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
struct sockaddr_in sin;
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#endif
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
/* see if we're bound all already! */
if (inp->sctp_flags & SCTP_PCB_FLAGS_BOUNDALL) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
*error = EINVAL;
return;
}
addr_touse = sa;
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#ifdef INET6
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
if (sa->sa_family == AF_INET6) {
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#ifdef INET
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
struct sockaddr_in6 *sin6;
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#endif
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
if (sa->sa_len != sizeof(struct sockaddr_in6)) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
*error = EINVAL;
return;
}
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
if ((inp->sctp_flags & SCTP_PCB_FLAGS_BOUND_V6) == 0) {
/* can only bind v6 on PF_INET6 sockets */
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
*error = EINVAL;
return;
}
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#ifdef INET
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
sin6 = (struct sockaddr_in6 *)addr_touse;
if (IN6_IS_ADDR_V4MAPPED(&sin6->sin6_addr)) {
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
if ((inp->sctp_flags & SCTP_PCB_FLAGS_BOUND_V6) &&
SCTP_IPV6_V6ONLY(inp)) {
/* can't bind v4-mapped on PF_INET sockets */
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
*error = EINVAL;
return;
}
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
in6_sin6_2_sin(&sin, sin6);
addr_touse = (struct sockaddr *)&sin;
}
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#endif
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
}
#endif
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#ifdef INET
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
if (sa->sa_family == AF_INET) {
if (sa->sa_len != sizeof(struct sockaddr_in)) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
*error = EINVAL;
return;
}
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
if ((inp->sctp_flags & SCTP_PCB_FLAGS_BOUND_V6) &&
SCTP_IPV6_V6ONLY(inp)) {
/* can't bind v4 on PF_INET sockets */
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
*error = EINVAL;
return;
}
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
}
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#endif
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
if (inp->sctp_flags & SCTP_PCB_FLAGS_UNBOUND) {
if (p == NULL) {
/* Can't get proc for Net/Open BSD */
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
*error = EINVAL;
return;
}
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
*error = sctp_inpcb_bind(so, addr_touse, NULL, p);
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
return;
}
/*
* No locks required here since bind and mgmt_ep_sa all do their own
* locking. If we do something for the FIX: below we may need to
* lock in that case.
*/
if (assoc_id == 0) {
/* add the address */
struct sctp_inpcb *lep;
- Update bindx address checking to properly screen out address per the socket api, adding port validation. We allow port 0 or the already bound port number and no others. Approved by: re@freebsd.org (Ken Smith)
2007-06-25 19:05:26 +00:00
struct sockaddr_in *lsin = (struct sockaddr_in *)addr_touse;
/* validate the incoming port */
if ((lsin->sin_port != 0) &&
(lsin->sin_port != inp->sctp_lport)) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- Update bindx address checking to properly screen out address per the socket api, adding port validation. We allow port 0 or the already bound port number and no others. Approved by: re@freebsd.org (Ken Smith)
2007-06-25 19:05:26 +00:00
*error = EINVAL;
return;
} else {
/* user specified 0 port, set it to existing port */
lsin->sin_port = inp->sctp_lport;
}
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
lep = sctp_pcb_findep(addr_touse, 1, 0, vrf_id);
if (lep != NULL) {
/*
* We must decrement the refcount since we have the
* ep already and are binding. No remove going on
* here.
*/
Adds support for the SCTP_PORT_REUSE option Fixes a refcount bug found in the process Obtained from: With the help of Michael Tuexen
2008-07-31 11:08:30 +00:00
SCTP_INP_DECR_REF(lep);
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
}
if (lep == inp) {
/* already bound to it.. ok */
return;
} else if (lep == NULL) {
((struct sockaddr_in *)addr_touse)->sin_port = 0;
*error = sctp_addr_mgmt_ep_sa(inp, addr_touse,
SCTP_ADD_IP_ADDRESS,
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
vrf_id, NULL);
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
} else {
*error = EADDRINUSE;
}
if (*error)
return;
} else {
/*
* FIX: decide whether we allow assoc based bindx
*/
}
}
/*
* sctp_bindx(DELETE) for one address.
* assumes all arguments are valid/checked by caller.
*/
void
Fix unused parameter warnings. While there, fix some whitespace issues. MFC after: 3 months.
2011-12-17 19:21:40 +00:00
sctp_bindx_delete_address(struct sctp_inpcb *inp,
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
struct sockaddr *sa, sctp_assoc_t assoc_id,
uint32_t vrf_id, int *error)
{
struct sockaddr *addr_touse;
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#if defined(INET) && defined(INET6)
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
struct sockaddr_in sin;
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#endif
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
/* see if we're bound all already! */
if (inp->sctp_flags & SCTP_PCB_FLAGS_BOUNDALL) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
*error = EINVAL;
return;
}
addr_touse = sa;
#ifdef INET and INET6 consistently. This also fixes a bug, where it was done wrong. MFC after: 3 days
2012-07-15 11:04:49 +00:00
#ifdef INET6
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
if (sa->sa_family == AF_INET6) {
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#ifdef INET
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
struct sockaddr_in6 *sin6;
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#endif
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
if (sa->sa_len != sizeof(struct sockaddr_in6)) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
*error = EINVAL;
return;
}
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
if ((inp->sctp_flags & SCTP_PCB_FLAGS_BOUND_V6) == 0) {
/* can only bind v6 on PF_INET6 sockets */
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
*error = EINVAL;
return;
}
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#ifdef INET
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
sin6 = (struct sockaddr_in6 *)addr_touse;
if (IN6_IS_ADDR_V4MAPPED(&sin6->sin6_addr)) {
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
if ((inp->sctp_flags & SCTP_PCB_FLAGS_BOUND_V6) &&
SCTP_IPV6_V6ONLY(inp)) {
/* can't bind mapped-v4 on PF_INET sockets */
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
*error = EINVAL;
return;
}
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
in6_sin6_2_sin(&sin, sin6);
addr_touse = (struct sockaddr *)&sin;
}
Fix the support of mapped IPv4 addresses. Thanks to Mark Bonnekessel and Markus Boese for making me aware of the problems. MFC after: 1 week
2014-12-06 20:00:08 +00:00
#endif
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
}
#endif
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#ifdef INET
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
if (sa->sa_family == AF_INET) {
if (sa->sa_len != sizeof(struct sockaddr_in)) {
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
*error = EINVAL;
return;
}
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
if ((inp->sctp_flags & SCTP_PCB_FLAGS_BOUND_V6) &&
SCTP_IPV6_V6ONLY(inp)) {
/* can't bind v4 on PF_INET sockets */
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTPUTIL, EINVAL);
- fix bindx to check addresses against socket's protocol family
2007-06-13 14:39:41 +00:00
*error = EINVAL;
return;
}
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
}
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#endif
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
/*
* No lock required mgmt_ep_sa does its own locking. If the FIX:
* below is ever changed we may need to lock before calling
* association level binding.
*/
if (assoc_id == 0) {
/* delete the address */
*error = sctp_addr_mgmt_ep_sa(inp, addr_touse,
SCTP_DEL_IP_ADDRESS,
- Fix so ifn's are properly deleted when the ref count goes to 0. - Fix so VRF's will clean themselves up when no references are around. - Allow sctp_ifa to be passed into inpcb_bind, addr_mgmt_ep_sa to bypass normal validation checks. - turn auto-asconf off for subset bound sockets - Moves all logging to use KTR. This gets rid of most of the logging #ifdef's with a few exceptions reducing the number of config options for SCTP.
2007-06-14 22:59:04 +00:00
vrf_id, NULL);
- Restructure so bindx functions are not done inline to socket option but are a seperate call that can be re-used if needed. - 64 bit issues o re-arrange cookie so it is better 64 bit aligned o For wire level things we need the packed attribute.
2007-06-12 11:21:00 +00:00
} else {
/*
* FIX: decide whether we allow assoc based bindx
*/
}
}
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
/*
* returns the valid local address count for an assoc, taking into account
* all scoping rules
*/
int
sctp_local_addr_count(struct sctp_tcb *stcb)
{
Changes from upstream to improve compilation when INET or INET6 or none of them is defined. MFC after: 3 days
2013-11-02 20:12:19 +00:00
int loopback_scope;
#if defined(INET)
int ipv4_local_scope, ipv4_addr_legal;
#endif
#if defined (INET6)
int local_scope, site_scope, ipv6_addr_legal;
#endif
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
struct sctp_vrf *vrf;
struct sctp_ifn *sctp_ifn;
struct sctp_ifa *sctp_ifa;
int count = 0;
/* Turn on all the appropriate scopes */
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
loopback_scope = stcb->asoc.scope.loopback_scope;
Changes from upstream to improve compilation when INET or INET6 or none of them is defined. MFC after: 3 days
2013-11-02 20:12:19 +00:00
#if defined(INET)
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
ipv4_local_scope = stcb->asoc.scope.ipv4_local_scope;
Changes from upstream to improve compilation when INET or INET6 or none of them is defined. MFC after: 3 days
2013-11-02 20:12:19 +00:00
ipv4_addr_legal = stcb->asoc.scope.ipv4_addr_legal;
#endif
#if defined(INET6)
Cleanup the handling of address scopes. Announce in the INIT/INIT-ACK only the supported address types. While there, do some whitespace cleanups. MFC after: 1 week
2013-02-09 17:26:14 +00:00
local_scope = stcb->asoc.scope.local_scope;
site_scope = stcb->asoc.scope.site_scope;
ipv6_addr_legal = stcb->asoc.scope.ipv6_addr_legal;
Changes from upstream to improve compilation when INET or INET6 or none of them is defined. MFC after: 3 days
2013-11-02 20:12:19 +00:00
#endif
- The address lock is changed to a rwlock. This also involves macro changes to have a RLOCK and a WLOCK and placing the correct version within the code. - The INP-INFO lock is changed to a rwlock. - When sctp_shutdown() is called on Mac OS X, the socket lock is held. So call sctp_chunk_output with SCTP_SO_LOCKED and not SCTP_SO_NOT_LOCKED. - Add SCTP_IPI_ADDR_[RW]LOCK and SCTP_IPI_ADDR_[RW]UNLOCK for Mac OS X. - u_int64_t -> uint64_t - add missing addr unlock for error return path Approved by: re@freebsd.org (K Smith)
2007-09-18 15:16:39 +00:00
SCTP_IPI_ADDR_RLOCK();
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
vrf = sctp_find_vrf(stcb->asoc.vrf_id);
if (vrf == NULL) {
/* no vrf, no addresses */
- The address lock is changed to a rwlock. This also involves macro changes to have a RLOCK and a WLOCK and placing the correct version within the code. - The INP-INFO lock is changed to a rwlock. - When sctp_shutdown() is called on Mac OS X, the socket lock is held. So call sctp_chunk_output with SCTP_SO_LOCKED and not SCTP_SO_NOT_LOCKED. - Add SCTP_IPI_ADDR_[RW]LOCK and SCTP_IPI_ADDR_[RW]UNLOCK for Mac OS X. - u_int64_t -> uint64_t - add missing addr unlock for error return path Approved by: re@freebsd.org (K Smith)
2007-09-18 15:16:39 +00:00
SCTP_IPI_ADDR_RUNLOCK();
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
return (0);
}
if (stcb->sctp_ep->sctp_flags & SCTP_PCB_FLAGS_BOUNDALL) {
/*
* bound all case: go through all ifns on the vrf
*/
LIST_FOREACH(sctp_ifn, &vrf->ifnlist, next_ifn) {
if ((loopback_scope == 0) &&
SCTP_IFN_IS_IFT_LOOP(sctp_ifn)) {
continue;
}
LIST_FOREACH(sctp_ifa, &sctp_ifn->ifalist, next_ifa) {
if (sctp_is_addr_restricted(stcb, sctp_ifa))
continue;
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
switch (sctp_ifa->address.sa.sa_family) {
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#ifdef INET
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
case AF_INET:
if (ipv4_addr_legal) {
struct sockaddr_in *sin;
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
Use union sctp_sockstore instead of struct sockaddr_storage. This eliminiates some warnings when building in userland. Thanks to Patrick Laimbock for reporting this issue. Remove also some unnecessary casts. There should be no functional change. MFC after: 1 week
2014-09-07 09:06:26 +00:00
sin = &sctp_ifa->address.sin;
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
if (sin->sin_addr.s_addr == 0) {
/*
* skip unspecified
* addrs
*/
continue;
}
Honor jails for unbound SCTP sockets when selecting source addresses, reporting IP-addresses to the peer during the handshake, adding addresses to the host, reporting the addresses via the sysctl interface (used by netstat, for example) and reporting the addresses to the application via socket options. This issue was reported by Bernd Walter. MFC after: 3 days
2014-06-20 13:26:49 +00:00
if (prison_check_ip4(stcb->sctp_ep->ip_inp.inp.inp_cred,
&sin->sin_addr) != 0) {
continue;
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
if ((ipv4_local_scope == 0) &&
(IN4_ISPRIVATE_ADDRESS(&sin->sin_addr))) {
continue;
}
/* count this one */
count++;
} else {
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
continue;
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
break;
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
#endif
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
#ifdef INET6
case AF_INET6:
if (ipv6_addr_legal) {
struct sockaddr_in6 *sin6;
Use union sctp_sockstore instead of struct sockaddr_storage. This eliminiates some warnings when building in userland. Thanks to Patrick Laimbock for reporting this issue. Remove also some unnecessary casts. There should be no functional change. MFC after: 1 week
2014-09-07 09:06:26 +00:00
sin6 = &sctp_ifa->address.sin6;
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
if (IN6_IS_ADDR_UNSPECIFIED(&sin6->sin6_addr)) {
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
continue;
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
}
Honor jails for unbound SCTP sockets when selecting source addresses, reporting IP-addresses to the peer during the handshake, adding addresses to the host, reporting the addresses via the sysctl interface (used by netstat, for example) and reporting the addresses to the application via socket options. This issue was reported by Bernd Walter. MFC after: 3 days
2014-06-20 13:26:49 +00:00
if (prison_check_ip6(stcb->sctp_ep->ip_inp.inp.inp_cred,
&sin6->sin6_addr) != 0) {
continue;
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
if (IN6_IS_ADDR_LINKLOCAL(&sin6->sin6_addr)) {
if (local_scope == 0)
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
continue;
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
if (sin6->sin6_scope_id == 0) {
if (sa6_recoverscope(sin6) != 0)
/*
*
* bad
*
* li
* nk
*
* loc
* al
*
* add
* re
* ss
* */
continue;
}
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
if ((site_scope == 0) &&
(IN6_IS_ADDR_SITELOCAL(&sin6->sin6_addr))) {
continue;
}
/* count this one */
count++;
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
}
Allow SCTP to compile without INET6. PR: 116816 Obtained from tuexen@fh-muenster.de: MFC after: 2 weeks
2008-04-16 17:24:18 +00:00
break;
#endif
default:
/* TSNH */
break;
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
}
}
}
} else {
/*
* subset bound case
*/
struct sctp_laddr *laddr;
LIST_FOREACH(laddr, &stcb->sctp_ep->sctp_addr_list,
sctp_nxt_addr) {
if (sctp_is_addr_restricted(stcb, laddr->ifa)) {
continue;
}
/* count this one */
count++;
}
}
- The address lock is changed to a rwlock. This also involves macro changes to have a RLOCK and a WLOCK and placing the correct version within the code. - The INP-INFO lock is changed to a rwlock. - When sctp_shutdown() is called on Mac OS X, the socket lock is held. So call sctp_chunk_output with SCTP_SO_LOCKED and not SCTP_SO_NOT_LOCKED. - Add SCTP_IPI_ADDR_[RW]LOCK and SCTP_IPI_ADDR_[RW]UNLOCK for Mac OS X. - u_int64_t -> uint64_t - add missing addr unlock for error return path Approved by: re@freebsd.org (K Smith)
2007-09-18 15:16:39 +00:00
SCTP_IPI_ADDR_RUNLOCK();
- take out a needless panic under invariants for sctp_output.c - Fix addrs's error checking of sctp_sendx(3) when addrcnt is less than SCTP_SMALL_IOVEC_SIZE - re-add back inpcb_bind local address check bypass capability - Fix it so sctp_opt_info is independant of assoc_id postion. - Fix cookie life set to use MSEC_TO_TICKS() macro. - asconf changes o More comment changes/clarifications related to the old local address "not" list which is now an explicit restricted list. o Rename some functions for clarity: - sctp_add/del_local_addr_assoc to xxx_local_addr_restricted() - asconf related iterator functions to sctp_asconf_iterator_xxx() o Fix bug when the same address is deleted and added (and removed from the asconf queue) where the ifa is "freed" twice refcount wise, possibly freeing it completely. o Fix bug in output where the first ASCONF would not go out after the last address is changed (e.g. only goes out when retransmitted). o Fix bug where multiple ASCONFs can be bundled in the same packet with the and with the same serial numbers. o Fix asconf stcb iterator to not send ASCONF until after all work queue entries have been processed. o Change behavior so that when the last address is deleted (auto asconf on a bound all endpoint) no action is taken until an address is added; at that time, an ASCONF add+delete is sent (if the assoc is still up). o Fix local address counting so that address scoping is taken into account. o #ifdef SCTP_TIMER_BASED_ASCONF the old timer triggered sending of ASCONF (after an RTO). The default now is to send ASCONF immediately (except for the case of changing/deleting the last usable address). Approved by: re(ken smith)@freebsd.org
2007-07-24 20:06:02 +00:00
return (count);
}
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
#if defined(SCTP_LOCAL_TRACE_BUF)
void
- DF bit was on for COOKIE-ECHO chunks. This is incorrect and should be OFF letting IP fragment large cookie-echos. - Rename sysctl variable logging to log_level. - Fix description of sysctl variable stats. - Add sysctl variable log to make sctp_log readable via sysctl mechanism (this is by compile switch and targets non KTR platforms or when someone wants to do performance wise tracing). - Removed debug code Approved by: re@freebsd.org (B Mah)
2007-09-13 14:43:54 +00:00
sctp_log_trace(uint32_t subsys, const char *str SCTP_UNUSED, uint32_t a, uint32_t b, uint32_t c, uint32_t d, uint32_t e, uint32_t f)
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
{
- DF bit was on for COOKIE-ECHO chunks. This is incorrect and should be OFF letting IP fragment large cookie-echos. - Rename sysctl variable logging to log_level. - Fix description of sysctl variable stats. - Add sysctl variable log to make sctp_log readable via sysctl mechanism (this is by compile switch and targets non KTR platforms or when someone wants to do performance wise tracing). - Removed debug code Approved by: re@freebsd.org (B Mah)
2007-09-13 14:43:54 +00:00
uint32_t saveindex, newindex;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
do {
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
saveindex = SCTP_BASE_SYSCTL(sctp_log).index;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
if (saveindex >= SCTP_MAX_LOGGING_SIZE) {
newindex = 1;
} else {
newindex = saveindex + 1;
}
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
} while (atomic_cmpset_int(&SCTP_BASE_SYSCTL(sctp_log).index, saveindex, newindex) == 0);
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
if (saveindex >= SCTP_MAX_LOGGING_SIZE) {
saveindex = 0;
}
- Macro-izes the packed declaration in all headers. - Vimage prep - these are major restructures to move all global variables to be accessed via a macro or two. The variables all go into a single structure. - Asconf address addition tweaks (add_or_del Interfaces) - Fix rwnd calcualtion to be more conservative. - Support SACK_IMMEDIATE flag to skip delayed sack by demand of peer. - Comment updates in the sack mapping calculations - Invarients panic added. - Pre-support for UDP tunneling (we can do this on MAC but will need added support from UDP to get a "pipe" of UDP packets in. - clear trace buffer sysctl added when local tracing on. Note the majority of this huge patch is all the vimage prep stuff :-)
2008-06-14 07:58:05 +00:00
SCTP_BASE_SYSCTL(sctp_log).entry[saveindex].timestamp = SCTP_GET_CYCLECOUNT;
SCTP_BASE_SYSCTL(sctp_log).entry[saveindex].subsys = subsys;
SCTP_BASE_SYSCTL(sctp_log).entry[saveindex].params[0] = a;
SCTP_BASE_SYSCTL(sctp_log).entry[saveindex].params[1] = b;
SCTP_BASE_SYSCTL(sctp_log).entry[saveindex].params[2] = c;
SCTP_BASE_SYSCTL(sctp_log).entry[saveindex].params[3] = d;
SCTP_BASE_SYSCTL(sctp_log).entry[saveindex].params[4] = e;
SCTP_BASE_SYSCTL(sctp_log).entry[saveindex].params[5] = f;
- Fix address add handling to clear cached routes and source addresses when peer acks the add in case the routing table changes. - Fix sctp_lower_sosend to send shutdown chunk for mbuf send case when sndlen = 0 and sinfoflag = SCTP_EOF - Fix sctp_lower_sosend for SCTP_ABORT mbuf send case with null data, So that it does not send the "null" data mbuf out and cause it to get freed twice. - Fix so auto-asconf sysctl actually effect the socket's asconf state. - Do not allow SCTP_AUTO_ASCONF option to be used on subset bound sockets. - Memset bug in sctp_output.c (arguments were reversed) submitted found and reported by Dave Jones (davej@codemonkey.org.uk). - PD-API point needs to be invoked >= not just > to conform to socket api draft this fixes sctp_indata.c in the two places need to be >=. - move M_NOTIFICATION to use M_PROTO5. - PEER_ADDR_PARAMS did not fail properly if you specify an address that is not in the association with a valid assoc_id. This meant you got or set the stcb level values instead of the destination you thought you were going to get/set. Now validate if the stcb is non-null and the net is NULL that the sa_family is set and the address is unspecified otherwise return an error. - The thread based iterator could crash if associations were freed at the exact time it was running. rework the worker thread to use the increment/decrement to prevent this and no longer use the markers that the timer based iterator uses. - Fix the memleak in sctp_add_addr_to_vrf() for the case when it is detected that ifa is already pointing to a ifn. - Fix it so that if someone is so insane that they drop the send window below the minimal add mark, they still can send. - Changed all state for associations to use mask safe macro. - During front states in association freeing in sctp_inpcbfree, we had a locking problem where locks were not in place where they should have been. - Free association calls were not testing the return value in sctp_inpcb_free() properly... others should be cast void returns where we don't care about the return value. - If a reference count is held on an assoc, even from the "force free" we should not do the actual free.. but instead let the timer free it. - When we enter sctp_input(), if the SCTP_ASOC_ABOUT_TO_BE_FREED flag is set, we must NOT process the packet but handle it like ootb. This is because while freeing an assoc we release the locks to get all the higher order locks so we can purge all the hash tables. This leaves a hole if a packet comes in just at that point. Now sctp_common_input_processing() will call the ootb code in such a case. - Change MBUF M_NOTIFICATION to use M_PROTO5 (per Sam L). This makes it so we don't have a conflict (I think this is a covertity change). We made this change AFTER some conversation and looking to make sure that M_PROTO5 does not have a problem between SCTP and the 802.11 stuff (which is the only other place its used). - Fixed lock order reversal and missing atomic protection around locked_tcb during association lookup and the 1-2-1 model. - Added debug to source address selection. - V6 output must always do checksum even for loopback. - Remove more locks around inp that are not needed for an atomically added/subtracted ref count. - slight optimization in the way we zero the array in sctp_sack_check() - It was possible to respond to a ABORT() with bad checksum with a PKT-DROP. This lead to a PKT-DROP/ABORT war. Add code to NOT send a PKT-DROP to any ABORT(). - Add an option for local logging (useful for macintosh or when you need better performing during debugging). Note no commands are here to get the log info, you must just use kgdb. - The timer code needs to be aware of if it needs to call sctp_sack_check() to slide the maps and adjust the cum-ack. This is because it may be out of sync cum-ack wise. - Added threshold managment logging. - If the user picked just the right size, that just filled the send window minus one mtu, we would enter a forever loop not copying and at the same time not blocking. Change from < to <= solves this. - Sysctl added to control the fragment interleave level which defaults to 1. - My rwnd control was not being used to control the rwnd properly (we did not add and subtract to it :-() this is now fixed so we handle small messages (1 byte etc) better to bring our rwnd down more slowly. Approved by: re@freebsd.org (Bruce Mah)
2007-08-24 00:53:53 +00:00
}
#endif
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
static void
Fix several problems with Stream Reset. 1) We were not handling (or sending) the IN_PROGRESS case if the other side (or our side) was not able to reset (awaiting more data). 2) We would improperly send a stream-reset when we should not. Not waiting until the TSN had been assigned when data was inqueue. Reviewed by: tuexen
2015-07-22 11:30:37 +00:00
sctp_recv_udp_tunneled_packet(struct mbuf *m, int off, struct inpcb *inp,
Add context pointer and source address to the UDP tunnel callback These are needed for the forthcoming vxlan implementation. The context pointer means we do not have to use a spare pointer field in the inpcb, and the source address is required to populate vxlan's forwarding table. While I highly doubt there is an out of tree consumer of the UDP tunneling callback, this change may be a difficult to eventually MFC. Phabricator: https://reviews.freebsd.org/D383 Reviewed by: gnn
2014-10-10 06:08:59 +00:00
const struct sockaddr *sa SCTP_UNUSED, void *ctx SCTP_UNUSED)
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
{
struct ip *iph;
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
#ifdef INET6
struct ip6_hdr *ip6;
#endif
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
struct mbuf *sp, *last;
struct udphdr *uhdr;
Cleanup the UDP decapsulation code. MFC after: 3 days
2012-06-18 17:09:39 +00:00
uint16_t port;
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
if ((m->m_flags & M_PKTHDR) == 0) {
/* Can't handle one that is not a pkt hdr */
goto out;
}
Cleanup the UDP decapsulation code. MFC after: 3 days
2012-06-18 17:09:39 +00:00
/* Pull the src port */
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
iph = mtod(m, struct ip *);
uhdr = (struct udphdr *)((caddr_t)iph + off);
port = uhdr->uh_sport;
Cleanup the UDP decapsulation code. MFC after: 3 days
2012-06-18 17:09:39 +00:00
/*
* Split out the mbuf chain. Leave the IP header in m, place the
* rest in the sp.
*/
Mechanically substitute flags from historic mbuf allocator with malloc(9) flags within sys. Exceptions: - sys/contrib not touched - sys/mbuf.h edited manually
2012-12-05 08:04:20 +00:00
sp = m_split(m, off, M_NOWAIT);
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
if (sp == NULL) {
/* Gak, drop packet, we can't do a split */
goto out;
}
Cleanup the UDP decapsulation code. MFC after: 3 days
2012-06-18 17:09:39 +00:00
if (sp->m_pkthdr.len < sizeof(struct udphdr) + sizeof(struct sctphdr)) {
/* Gak, packet can't have an SCTP header in it - too small */
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
m_freem(sp);
goto out;
}
Cleanup the UDP decapsulation code. MFC after: 3 days
2012-06-18 17:09:39 +00:00
/* Now pull up the UDP header and SCTP header together */
sp = m_pullup(sp, sizeof(struct udphdr) + sizeof(struct sctphdr));
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
if (sp == NULL) {
/* Gak pullup failed */
goto out;
}
Cleanup the UDP decapsulation code. MFC after: 3 days
2012-06-18 17:09:39 +00:00
/* Trim out the UDP header */
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
m_adj(sp, sizeof(struct udphdr));
/* Now reconstruct the mbuf chain */
Cleanup the UDP decapsulation code. MFC after: 3 days
2012-06-18 17:09:39 +00:00
for (last = m; last->m_next; last = last->m_next);
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
last->m_next = sp;
m->m_pkthdr.len += sp->m_pkthdr.len;
When receiving an SCTP/UDP packet and the interface performed the UDP checksum computation and signals that it was OK, clear this bit when passing the packet to SCTP. Since the bits indicating a valid UDP checksum and a valid SCTP checksum are the same, the SCTP stack would assume that also an SCTP checksum check has been performed. MFC after: 1 week
2015-11-26 09:25:20 +00:00
/*
* The CSUM_DATA_VALID flags indicates that the HW checked the UDP
* checksum and it was valid. Since CSUM_DATA_VALID ==
* CSUM_SCTP_VALID this would imply that the HW also verified the
* SCTP checksum. Therefore, clear the bit.
*/
SCTPDBG(SCTP_DEBUG_CRCOFFLOAD,
"sctp_recv_udp_tunneled_packet(): Packet of length %d received on %s with csum_flags 0x%b.\n",
m->m_pkthdr.len,
if_name(m->m_pkthdr.rcvif),
(int)m->m_pkthdr.csum_flags, CSUM_BITS);
m->m_pkthdr.csum_flags &= ~CSUM_DATA_VALID;
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
iph = mtod(m, struct ip *);
switch (iph->ip_v) {
Improve compilation of SCTP code without INET support. Some bugs where fixed while doing this: * ASCONF-ACK messages might use wrong port number when using IPv6. * Checking for additional addresses takes the correct address into account and also does not do more comparisons than necessary. This patch is based on one received from bz@ who was sponsored by The FreeBSD Foundation and iXsystems. MFC after: 1 week
2011-04-30 11:18:16 +00:00
#ifdef INET
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
case IPVERSION:
Use ntohs() and htons() in correct order. However, this doesn't change functionality.
2012-10-29 20:42:48 +00:00
iph->ip_len = htons(ntohs(iph->ip_len) - sizeof(struct udphdr));
Cleanup the UDP decapsulation code. MFC after: 3 days
2012-06-18 17:09:39 +00:00
sctp_input_with_port(m, off, port);
break;
Improve compilation of SCTP code without INET support. Some bugs where fixed while doing this: * ASCONF-ACK messages might use wrong port number when using IPv6. * Checking for additional addresses takes the correct address into account and also does not do more comparisons than necessary. This patch is based on one received from bz@ who was sponsored by The FreeBSD Foundation and iXsystems. MFC after: 1 week
2011-04-30 11:18:16 +00:00
#endif
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
#ifdef INET6
case IPV6_VERSION >> 4:
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
ip6 = mtod(m, struct ip6_hdr *);
ip6->ip6_plen = htons(ntohs(ip6->ip6_plen) - sizeof(struct udphdr));
sctp6_input_with_port(&m, &off, port);
Cleanup the UDP decapsulation code. MFC after: 3 days
2012-06-18 17:09:39 +00:00
break;
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
#endif
default:
Cleanup the UDP decapsulation code. MFC after: 3 days
2012-06-18 17:09:39 +00:00
goto out;
break;
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
}
return;
out:
m_freem(m);
}
- Adds support for the multi-asconf (From Kozuka-san) - Adds some prepwork (Not all yet) for vimage in particular support the delete the sctppcbinfo.xx structs. There is still a leak in here if it were to be called plus we stil need the regrouping (From Me and Michael Tuexen) - Adds support for UDP tunneling. For BSD there is no socket yet setup so its disabled, but major argument changes are in here to emcompass the passing of the port number (zero when you don't have a udp tunnel, the default for BSD). Will add some hooks in UDP here shortly (discussed with Robert) that will allow easy tunneling. (Mainly from Peter Lei and Michael Tuexen with some BSD work from me :-D) - Some ease for windows, evidently leave is reserved by their compile move label leave: -> out: MFC after: 1 week
2008-05-20 13:47:46 +00:00
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
void
- Adds support for the multi-asconf (From Kozuka-san) - Adds some prepwork (Not all yet) for vimage in particular support the delete the sctppcbinfo.xx structs. There is still a leak in here if it were to be called plus we stil need the regrouping (From Me and Michael Tuexen) - Adds support for UDP tunneling. For BSD there is no socket yet setup so its disabled, but major argument changes are in here to emcompass the passing of the port number (zero when you don't have a udp tunnel, the default for BSD). Will add some hooks in UDP here shortly (discussed with Robert) that will allow easy tunneling. (Mainly from Peter Lei and Michael Tuexen with some BSD work from me :-D) - Some ease for windows, evidently leave is reserved by their compile move label leave: -> out: MFC after: 1 week
2008-05-20 13:47:46 +00:00
sctp_over_udp_stop(void)
{
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
/*
* This function assumes sysctl caller holds sctp_sysctl_info_lock()
* for writting!
*/
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
#ifdef INET
if (SCTP_BASE_INFO(udp4_tun_socket) != NULL) {
soclose(SCTP_BASE_INFO(udp4_tun_socket));
SCTP_BASE_INFO(udp4_tun_socket) = NULL;
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
}
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
#endif
#ifdef INET6
if (SCTP_BASE_INFO(udp6_tun_socket) != NULL) {
soclose(SCTP_BASE_INFO(udp6_tun_socket));
SCTP_BASE_INFO(udp6_tun_socket) = NULL;
}
#endif
- Adds support for the multi-asconf (From Kozuka-san) - Adds some prepwork (Not all yet) for vimage in particular support the delete the sctppcbinfo.xx structs. There is still a leak in here if it were to be called plus we stil need the regrouping (From Me and Michael Tuexen) - Adds support for UDP tunneling. For BSD there is no socket yet setup so its disabled, but major argument changes are in here to emcompass the passing of the port number (zero when you don't have a udp tunnel, the default for BSD). Will add some hooks in UDP here shortly (discussed with Robert) that will allow easy tunneling. (Mainly from Peter Lei and Michael Tuexen with some BSD work from me :-D) - Some ease for windows, evidently leave is reserved by their compile move label leave: -> out: MFC after: 1 week
2008-05-20 13:47:46 +00:00
}
Some more cleanups related to an kernel without INET. MFC after: 1 week
2011-05-02 15:53:00 +00:00
int
- Adds support for the multi-asconf (From Kozuka-san) - Adds some prepwork (Not all yet) for vimage in particular support the delete the sctppcbinfo.xx structs. There is still a leak in here if it were to be called plus we stil need the regrouping (From Me and Michael Tuexen) - Adds support for UDP tunneling. For BSD there is no socket yet setup so its disabled, but major argument changes are in here to emcompass the passing of the port number (zero when you don't have a udp tunnel, the default for BSD). Will add some hooks in UDP here shortly (discussed with Robert) that will allow easy tunneling. (Mainly from Peter Lei and Michael Tuexen with some BSD work from me :-D) - Some ease for windows, evidently leave is reserved by their compile move label leave: -> out: MFC after: 1 week
2008-05-20 13:47:46 +00:00
sctp_over_udp_start(void)
{
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
uint16_t port;
int ret;
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
#ifdef INET
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
struct sockaddr_in sin;
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
#endif
#ifdef INET6
struct sockaddr_in6 sin6;
#endif
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
/*
* This function assumes sysctl caller holds sctp_sysctl_info_lock()
* for writting!
*/
port = SCTP_BASE_SYSCTL(sctp_udp_tunneling_port);
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
if (ntohs(port) == 0) {
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
/* Must have a port set */
return (EINVAL);
}
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
#ifdef INET
if (SCTP_BASE_INFO(udp4_tun_socket) != NULL) {
/* Already running -- must stop first */
return (EALREADY);
}
#endif
#ifdef INET6
if (SCTP_BASE_INFO(udp6_tun_socket) != NULL) {
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
/* Already running -- must stop first */
return (EALREADY);
}
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
#endif
#ifdef INET
if ((ret = socreate(PF_INET, &SCTP_BASE_INFO(udp4_tun_socket),
SOCK_DGRAM, IPPROTO_UDP,
curthread->td_ucred, curthread))) {
sctp_over_udp_stop();
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
return (ret);
}
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
/* Call the special UDP hook. */
if ((ret = udp_set_kernel_tunneling(SCTP_BASE_INFO(udp4_tun_socket),
Add context pointer and source address to the UDP tunnel callback These are needed for the forthcoming vxlan implementation. The context pointer means we do not have to use a spare pointer field in the inpcb, and the source address is required to populate vxlan's forwarding table. While I highly doubt there is an out of tree consumer of the UDP tunneling callback, this change may be a difficult to eventually MFC. Phabricator: https://reviews.freebsd.org/D383 Reviewed by: gnn
2014-10-10 06:08:59 +00:00
sctp_recv_udp_tunneled_packet, NULL))) {
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
sctp_over_udp_stop();
return (ret);
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
}
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
/* Ok, we have a socket, bind it to the port. */
memset(&sin, 0, sizeof(struct sockaddr_in));
sin.sin_len = sizeof(struct sockaddr_in);
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
sin.sin_family = AF_INET;
sin.sin_port = htons(port);
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
if ((ret = sobind(SCTP_BASE_INFO(udp4_tun_socket),
(struct sockaddr *)&sin, curthread))) {
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
sctp_over_udp_stop();
return (ret);
}
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
#endif
#ifdef INET6
if ((ret = socreate(PF_INET6, &SCTP_BASE_INFO(udp6_tun_socket),
SOCK_DGRAM, IPPROTO_UDP,
curthread->td_ucred, curthread))) {
sctp_over_udp_stop();
return (ret);
}
/* Call the special UDP hook. */
if ((ret = udp_set_kernel_tunneling(SCTP_BASE_INFO(udp6_tun_socket),
Add context pointer and source address to the UDP tunnel callback These are needed for the forthcoming vxlan implementation. The context pointer means we do not have to use a spare pointer field in the inpcb, and the source address is required to populate vxlan's forwarding table. While I highly doubt there is an out of tree consumer of the UDP tunneling callback, this change may be a difficult to eventually MFC. Phabricator: https://reviews.freebsd.org/D383 Reviewed by: gnn
2014-10-10 06:08:59 +00:00
sctp_recv_udp_tunneled_packet, NULL))) {
Add support for SCTP/UDP/IPV6. This completes the support of http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps MFC after: 1 week
2012-11-17 20:04:04 +00:00
sctp_over_udp_stop();
return (ret);
}
/* Ok, we have a socket, bind it to the port. */
memset(&sin6, 0, sizeof(struct sockaddr_in6));
sin6.sin6_len = sizeof(struct sockaddr_in6);
sin6.sin6_family = AF_INET6;
sin6.sin6_port = htons(port);
if ((ret = sobind(SCTP_BASE_INFO(udp6_tun_socket),
(struct sockaddr *)&sin6, curthread))) {
sctp_over_udp_stop();
return (ret);
}
#endif
- Cleanup checksum code. - Prepare for CRC offloading, add MIB counters (RS/MT). - Bugfix: Disable CRC computation for IPv6 addresses with local scope (MT). - Bugfix: Handle close() with SO_LINGER correctly when notifications are generated during the close() call(MT). - Bugfix: Generate DRY event when sender is dry during subscription. Only for 1-to-1 style sockets (RS/MT) - Bugfix: Put vtags for the correct amount of time into time-wait (MT). - Bugfix: Clear vtag entries correctly on expiration (MT). - Bugfix: shutdown() indicates ENOTCONN when called for unconnected 1-to-1 style sockets (MT). - Bugfix: In sctp Auth code (PL). - Add support for devices that support SCTP csum offload (igb). - Add missing sctp_associd to mib sysctl xsctp_tcb structure (RS) Obtained from: With help from Peter Lei and Michael Tuexen
2009-02-03 11:04:03 +00:00
return (0);
- Adds support for the multi-asconf (From Kozuka-san) - Adds some prepwork (Not all yet) for vimage in particular support the delete the sctppcbinfo.xx structs. There is still a leak in here if it were to be called plus we stil need the regrouping (From Me and Michael Tuexen) - Adds support for UDP tunneling. For BSD there is no socket yet setup so its disabled, but major argument changes are in here to emcompass the passing of the port number (zero when you don't have a udp tunnel, the default for BSD). Will add some hooks in UDP here shortly (discussed with Robert) that will allow easy tunneling. (Mainly from Peter Lei and Michael Tuexen with some BSD work from me :-D) - Some ease for windows, evidently leave is reserved by their compile move label leave: -> out: MFC after: 1 week
2008-05-20 13:47:46 +00:00
}
Reference in New Issue Copy Permalink