address is not supplied, then jail IP is choosed and in_pcbbind() is called.
Since udp_output() does not save local addr after call to in_pcbconnect_setup(),
in_pcbbind() is called for each packet, and this is incorrect.
So, we shall treat jailed sockets specially in udp_output(), we will save
their local address.
This fixes a long standing bug with broken sendto() system call in jails.
PR: kern/26506
Reviewed by: rwatson
MFC after: 2 weeks
loopback interface. Nobody have explained me sense of this check.
It breaks connect() system call to a destination address which is
loopback routed (e.g. blackholed).
Reviewed by: silence on net@
MFC after: 2 weeks
a socket from a regular socket to a listening socket able to accept new
connections. As part of this state transition, solisten() calls into the
protocol to update protocol-layer state. There were several bugs in this
implementation that could result in a race wherein a TCP SYN received
in the interval between the protocol state transition and the shortly
following socket layer transition would result in a panic in the TCP code,
as the socket would be in the TCPS_LISTEN state, but the socket would not
have the SO_ACCEPTCONN flag set.
This change does the following:
- Pushes the socket state transition from the socket layer solisten() to
to socket "library" routines called from the protocol. This permits
the socket routines to be called while holding the protocol mutexes,
preventing a race exposing the incomplete socket state transition to TCP
after the TCP state transition has completed. The check for a socket
layer state transition is performed by solisten_proto_check(), and the
actual transition is performed by solisten_proto().
- Holds the socket lock for the duration of the socket state test and set,
and over the protocol layer state transition, which is now possible as
the socket lock is acquired by the protocol layer, rather than vice
versa. This prevents additional state related races in the socket
layer.
This permits the dual transition of socket layer and protocol layer state
to occur while holding locks for both layers, making the two changes
atomic with respect to one another. Similar changes are likely require
elsewhere in the socket/protocol code.
Reported by: Peter Holm <peter@holm.cc>
Review and fixes from: emax, Antoine Brodin <antoine.brodin@laposte.net>
Philosophical head nod: gnn
reported to the sender - in the case where the sender sends data
outside the window (as WinXP does :().
Reported by: Sam Jensen <sam at wand dot net dot nz>
Submitted by: Mohan Srinivasan
Remove the SACK "initburst" sysctl.
- Fix bugs in SACK dupack and partialack handling that can cause
large bursts while in SACK recovery.
Submitted by: Mohan Srinivasan
o Use SYSCTL_IN() macro instead of direct call of copyin(9).
Submitted by: ume
o Move sysctl_drop() implementation to sys/netinet/tcp_subr.c where
most of tcp sysctls live.
o There are net.inet[6].tcp[6].getcred sysctls already, no needs in
a separate struct tcp_ident_mapping.
Suggested by: ume
utility:
The tcpdrop command drops the TCP connection specified by the
local address laddr, port lport and the foreign address faddr,
port fport.
Obtained from: OpenBSD
Reviewed by: rwatson (locking), ru (man page), -current
MFC after: 1 month
a UMA zone instead. This should eliminate a bit of the locking
overhead associated with with malloc and reduce the memory
consumption associated with each new state.
Reviewed by: rwatson, andre
Silence on: ipfw@
MFC after: 1 week
second; since the default hz has changed to 1000 times a second,
this resulted in unecessary work being performed.
MFC after: 2 weeks
Discussed with: phk, cperciva
General head nod: silby
- ip_fw_chk() returns action as function return value. Field retval is
removed from args structure. Action is not flag any more. It is one
of integer constants.
- Any action-specific cookies are returned either in new "cookie" field
in args structure (dummynet, future netgraph glue), or in mbuf tag
attached to packet (divert, tee, some future action).
o Convert parsing of return value from ip_fw_chk() in ipfw_check_{in,out}()
to a switch structure, so that the functions are more readable, and a future
actions can be added with less modifications.
Approved by: andre
MFC after: 2 months
of len in tcp_output(), in the case where the FIN has already been
transmitted. The mis-computation of len is because of a gcc
optimization issue, which this change works around.
Submitted by: Mohan Srinivasan
that the RFC 793 specification for accepting RST packets should be
following. When followed, this makes one vulnerable to the attacks
described in "slipping in the window", but it may be necessary in
some odd circumstances.
connection rates, which is causing problems for some users.
To retain the security advantage of random ports and ensure
correct operation for high connection rate users, disable
port randomization during periods of high connection rates.
Whenever the connection rate exceeds randomcps (10 by default),
randomization will be disabled for randomtime (45 by default)
seconds. These thresholds may be tuned via sysctl.
Many thanks to Igor Sysoev, who proved the necessity of this
change and tested many preliminary versions of the patch.
MFC After: 20 seconds
cases for tcp_input():
While it is true that the pcbinfo lock provides a pseudo-reference to
inpcbs, both the inpcb and pcbinfo locks are required to free an
un-referenced inpcb. As such, we can release the pcbinfo lock as
long as the inpcb remains locked with the confidence that it will not
be garbage-collected. This leads to a less conservative locking
strategy that should reduce contention on the TCP pcbinfo lock.
Discussed with: sam
multiple MIB entries using sysctl in short order, which might
result in unexpected values for tcp_maxidle being generated by
tcp_slowtimo. In practice, this will not happen, or at least,
doesn't require an explicit comment.
MFC after: 2 weeks
Andre:
First lets get major new features into the kernel in a clean and nice way,
and then start optimizing. In this case we don't have any obfusication that
makes later profiling and/or optimizing difficult in any way.
Requested by: csjp, sam
mechanism used by pfil. This shared locking mechanism will remove
a nasty lock order reversal which occurs when ucred based rules
are used which results in hard locks while mpsafenet=1.
So this removes the debug.mpsafenet=0 requirement when using
ucred based rules with IPFW.
It should be noted that this locking mechanism does not guarantee
fairness between read and write locks, and that it will favor
firewall chain readers over writers. This seemed acceptable since
write operations to firewall chains protected by this lock tend to
be less frequent than reads.
Reviewed by: andre, rwatson
Tested by: myself, seanc
Silence on: ipfw@
MFC after: 1 month
tcpip_fillheaders()
tcp_discardcb()
tcp_close()
tcp_notify()
tcp_new_isn()
tcp_xmit_bandwidth_limit()
Fix a locking comment in tcp_twstart(): the pcbinfo will be locked (and
is asserted).
MFC after: 2 weeks
inp->inp_moptions pointer, so that ip_getmoptions() can perform
necessary locking when doing non-atomic reads.
Lock the inpcb by default to copy any data to local variables, then
unlock before performing sooptcopyout().
MFC after: 2 weeks
modifications to the inpcb IP options mbuf:
- Lock the inpcb before passing it into ip_pcbopts() in order to prevent
simulatenous reads and read-modify-writes that could result in races.
- Pass the inpcb reference into ip_pcbopts() instead of the option chain
pointer in the inpcb.
- Assert the inpcb lock in ip_pcbots.
- Convert one or two uses of a pointer as a boolean or an integer
comparison to a comparison with NULL for readability.
pointer updates: test available space while holding the socket buffer
mutex, and continue to hold until until the pointer update has been
performed.
MFC after: 2 weeks
This socket option allows processes query a TCP socket for some low
level transmission details, such as the current send, bandwidth, and
congestion windows. Linux provides a 'struct tcpinfo' structure
containing various variables, rather than separate socket options;
this makes the API somewhat fragile as it makes it dificult to add
new entries of interest as requirements and implementation evolve.
As such, I've included a large pad at the end of the structure.
Right now, relatively few of the Linux API fields are filled in, and
some contain no logical equivilent on FreeBSD. I've include __'d
entries in the structure to make it easier to figure ou what is and
isn't omitted. This API/ABI should be considered unstable for the
time being.
window was 0 bytes in size. This may have been the cause of unsolved
"connection not closing" reports over the years.
Thanks to Michiel Boland for providing the fix and providing a concise
test program for the problem.
Submitted by: Michiel Boland
MFC after: 2 weeks
contents of the tcpcb are read and modified in volume.
In tcp_input(), replace th comparison with 0 with a comparison with
NULL.
At the 'findpcb', 'dropafterack', and 'dropwithreset' labels in
tcp_input(), assert 'headlocked'. Try to improve consistency between
various assertions regarding headlocked to be more informative.
MFC after: 2 weeks
structure, so assert the inpcb lock associated with the tcptw.
Also assert the tcbinfo lock, as tcp_timewait() may call
tcp_twclose() or tcp_2msl_rest(), which require it. Since
tcp_timewait() is already called with that lock from tcp_input(),
this doesn't change current locking, merely documents reasons for
it.
In tcp_twstart(), assert the tcbinfo lock, as tcp_timer_2msl_rest()
is called, which requires that lock.
In tcp_twclose(), assert the tcbinfo lock, as tcp_timer_2msl_stop()
is called, which requires that lock.
Document the locking strategy for the time wait queues in tcp_timer.c,
which consists of protecting the time wait queues in the same manner
as the tcbinfo structure (using the tcbinfo lock).
In tcp_timer_2msl_reset(), assert the tcbinfo lock, as the time wait
queues are modified.
In tcp_timer_2msl_stop(), assert the tcbinfo lock, as the time wait
queues may be modified.
In tcp_timer_2msl_tw(), assert the tcbinfo lock, as the time wait
queues may be modified.
MFC after: 2 weeks
but unlikely races that could be corrected by having tcp_keepcnt
and tcp_keepintvl modifications go through handler functions via
sysctl, but probably is not worth doing. Updates to multiple
sysctls within evaluation of a single addition are unlikely.
Annotate that tcp_canceltimers() is currently unused.
De-spl tcp_timer_delack().
De-spl tcp_timer_2msl().
MFC after: 2 weeks
on the tcpcb, but also calls into tcp_close() and tcp_twrespond().
Annotate that tcp_twrecycleable() requires the inpcb lock because it does
a series of non-atomic reads of the tcpcb, but is currently called
without the inpcb lock by the caller. This is a bug.
Assert the inpcb lock in tcp_twclose() as it performs a read-modify-write
of the timewait structure/inpcb, and calls in_pcbdetach() which requires
the lock.
Assert the inpcb lock in tcp_twrespond(), as it performs multiple
non-atomic reads of the tcptw and inpcb structures, as well as calling
mac_create_mbuf_from_inpcb(), tcpip_fillheaders(), which require the
inpcb lock.
MFC after: 2 weeks
protects access to the ISN state variables.
Acquire the tcbinfo write lock in tcp_isn_tick() to synchronize
timer-driven isn bumping.
Staticize internal ISN variables since they're not used outside of
tcp_subr.c.
MFC after: 2 weeks
from divert sockets.
- Remove div_disconnect() method, since it shouldn't be called now.
- Remove div_abort() method. It was never called directly, since protocol
doesn't have listen queue. It was called only from div_disconnect(),
which is removed now.
Reviewed by: rwatson, maxim
Approved by: julian (mentor)
MT5 after: 1 week
MT4 after: 1 month
after allowing more than one address with the same prefix.
Reported by: Vladimir Grebenschikov <vova NO fbsd SPAM ru>
Submitted by: ru (also NetBSD rev. 1.83)
Pointyhat to: mlaier
and has been broken twice:
- in the beginning of div_output() replace KASSERT with assignment, as
it was in rev. 1.83. [1] [to be MFCed]
- refactor changes introduced in rev. 1.100: do not prepend a new tag
unconditionally. Before doing this check whether we have one. [2]
A small note for all hacking in this area:
when divert socket is not a real userland, but ng_ksocket(4), we receive
_the same_ mbufs, that we transmitted to socket. These mbufs have rcvif,
the tags we've put on them. And we should treat them correctly.
Discussed with: mlaier [1]
Silence from: green [2]
Reviewed by: maxim
Approved by: julian (mentor)
MFC after: 1 week
This makes it possible to have more than one address with the same prefix.
The first address added is used for the route. On deletion of an address
with IFA_ROUTE set, we try to find a "fallback" address and hand over the
route if possible.
I plan to MFC this in 4 weeks, hence I keep the - now obsolete - argument to
in_ifscrub as it must be considered KAPI as it is not static in in.c. I will
clean this after the MFC.
Discussed on: arch, net
Tested by: many testers of the CARP patches
Nits from: ru, Andrea Campi <andrea+freebsd_arch webcom it>
Obtained from: WIDE via OpenBSD
MFC after: 1 month
retain the pcbinfo lock until we're done using a pcb in the in-bound
path, as the pcbinfo lock acts as a pseuo-reference to prevent the pcb
from potentially being recycled. Clean up assertions and make sure to
assert that the pcbinfo is locked at the head of code subsections where
it is needed. Free the mbuf at the end of tcp_input after releasing
any held locks to reduce the time the locks are held.
MFC after: 3 weeks
destined for a blackhole route.
This also means that blackhole routes do not need to be bound to lo(4)
or disc(4) interfaces for the net.inet.ip.fastforwarding=1 case.
Submitted by: james at towardex dot com
Sponsored by: eXtensible Open Router Project <URL:http://www.xorp.org/>
MFC after: 3 weeks
udp_in6, and udp_ip6 to pass socket address state between udp_input(),
udp_append(), and soappendaddr_locked(). While file in the default
configuration, when running with multiple netisrs or direct ithread
dispatch, this can result in races wherein user processes using
recvmsg() get back the wrong source IP/port. To correct this and
related races:
- Eliminate udp_ip6, which is believed to be generated but then never
used. Eliminate ip_2_ip6_hdr() as it is now unneeded.
- Eliminate setting, testing, and existence of 'init' status fields
for the IPv6 structures. While with multiple UDP delivery this
could lead to amortization of IPv4 -> IPv6 conversion when
delivering an IPv4 UDP packet to an IPv6 socket, it added
substantial complexity and side effects.
- Move global structures into the stack, declaring udp_in in
udp_input(), and udp_in6 in udp_append() to be used if a conversion
is required. Pass &udp_in into udp_append().
- Re-annotate comments to reflect updates.
With this change, UDP appears to operate correctly in the presence of
substantial inbound processing parallelism. This solution avoids
introducing additional synchronization, but does increase the
potential stack depth.
Discovered by: kris (Bug Magnet)
MFC after: 3 weeks
A complete rationale and discussion is given in this message
and the resulting discussion:
http://docs.freebsd.org/cgi/mid.cgi?4177C8AD.6060706
Note that this commit removes only the functional part of T/TCP
from the tcp_* related functions in the kernel. Other features
introduced with RFC1644 are left intact (socket layer changes,
sendmsg(2) on connection oriented protocols) and are meant to
be reused by a simpler and less intrusive reimplemention of the
previous T/TCP functionality.
Discussed on: -arch
under high load: only set function state to loop and continuing sending
if there is no data left to send.
RELENG_5_3 candidate.
Feet provided: Peter Losher <Peter underscore Losher at isc dot org>
Diagnosed by: Aniel Hartmeier <daniel at benzedrine dot cx>
Submitted by: mohan <mohans at yahoo-inc dot com>
protocols: it is possible for sockets to be created and attached
to the divert protocol between the test for sockets present and
successful unload of the registration handler. We will need to
explore more mature APIs for unregistering the protocol and then
draining consumers, or an atomic test-and-unregister mechanism.
of protocols. The call to divert_packet() is done through a function pointer. All
semantics of IPDIVERT remain intact. If IPDIVERT is not loaded ipfw will refuse to
install divert rules and natd will complain about 'protocol not supported'. Once
it is loaded both will work and accept rules and open the divert socket. The module
can only be unloaded if no divert sockets are open. It does not close any divert
sockets when an unload is requested but will return EBUSY instead.
protocols in inetsw[] and define initially eight spacer slots.
Remove conflicting declaration 'struct pr_usrreqs nousrreqs'. It is
now declared and initialized in kern/uipc_domain.c.
With pr_proto_register() it has become possible to dynamically load protocols
within the PF_INET domain. However the PF_INET domain has a second important
structure called ip_protox[] that is derived from the 'struct protosw inetsw[]'
and takes care of the de-multiplexing of the various protocols that ride on
top of IP packets.
The functions ipproto_[un]register() allow to dynamically adjust the ip_protox[]
array mux in a consistent and easy way. To register a protocol within
ip_protox[] the existence of a corresponding and matching protocol definition
in inetsw[] is required. The function does not allow to overwrite an already
registered protocol. The unregister function simply replaces the mux slot with
the default index pointer to IPPROTO_RAW as it was previously.
(sorele()/sotryfree()):
- This permits the caller to acquire the accept mutex before the socket
mutex, avoiding sofree() having to drop the socket mutex and re-order,
which could lead to races permitting more than one thread to enter
sofree() after a socket is ready to be free'd.
- This also covers clearing of the so_pcb weak socket reference from
the protocol to the socket, preventing races in clearing and
evaluation of the reference such that sofree() might be called more
than once on the same socket.
This appears to close a race I was able to easily trigger by repeatedly
opening and resetting TCP connections to a host, in which the
tcp_close() code called as a result of the RST raced with the close()
of the accepted socket in the user process resulting in simultaneous
attempts to de-allocate the same socket. The new locking increases
the overhead for operations that may potentially free the socket, so we
will want to revise the synchronization strategy here as we normalize
the reference counting model for sockets. The use of the accept mutex
in freeing of sockets that are not listen sockets is primarily
motivated by the potential need to remove the socket from the
incomplete connection queue on its parent (listen) socket, so cleaning
up the reference model here may allow us to substantially weaken the
synchronization requirements.
RELENG_5_3 candidate.
MFC after: 3 days
Reviewed by: dwhite
Discussed with: gnn, dwhite, green
Reported by: Marc UBM Bocklet <ubm at u-boot-man dot de>
Reported by: Vlad <marchenko at gmail dot com>
in udp_input(), since the udbinfo lock is used to prevent removal of
the inpcb while in use (i.e., as a form of reference count) in the
in-bound path.
RELENG_5 candidate.
it isn't printed if the IP address in question is '0.0.0.0', which is
used by nodes performing DHCP lookup, and so constitute a false
positive as a report of misconfiguration.
processes in jail could create raw sockets, additional access control
checks were added to raw IP sockets to limit the ways in which those
sockets could be used. Specifically, only the socket option IP_HDRINCL
was permitted in rip_ctloutput(). Other socket options were protected
by a call to suser(). This change was required to prevent processes
in a Jail from modifying system properties such as multicast routing
and firewall rule sets.
However, it also introduced a regression: processes that create a raw
socket with root privilege, but then downgraded credential (i.e., a
daemon giving up root, or a setuid process switching back to the real
uid) could no longer issue other unprivileged generic IP socket option
operations, such as IP_TOS, IP_TTL, and the multicast group membership
options, which prevented multicast routing daemons (and some other
tools) from operating correctly.
This change pushes the access control decision down to the granularity
of individual socket options, rather than all socket options, on raw
IP sockets. When rip_ctloutput() doesn't implement an option, it will
now pass the request directly to in_control() without an access
control check. This should restore the functionality of the generic
IP socket options for raw sockets in the above-described scenarios,
which may be confirmed with the ipsockopt regression test.
RELENG_5 candidate.
Reviewed by: csjp
reaching into the socket buffer. This prevents a number of potential
races, including dereferencing of sb_mb while unlocked leading to
a NULL pointer deref (how I found it). Potentially this might also
explain other "odd" TCP behavior on SMP boxes (although haven't
seen it reported).
RELENG_5 candidate.
callouts as non-CALLOUT_MPSAFE. Otherwise, they may trigger an
assertion regarding Giant if they enter other parts of the stack from
the callout.
MFC after: 3 days
Reported by: Dikshie < dikshie at ppk dot itb dot ac dot id >
