Commit Graph

404 Commits

Author SHA1 Message Date
Ruslan Ermilov
9573c7e9cf Whitespace nit. 2004-05-19 11:35:03 +00:00
Ruslan Ermilov
171e08dc7f Maintain statistics about the received frames. 2004-05-19 11:26:33 +00:00
Archie Cobbs
1a292b8015 Allow ng_ether "lower" and "orphans" hooks to be connected at the same time.
Reviewed by:	julian
PR:		kern/63317
2004-05-16 19:31:35 +00:00
Maksim Yevmenkin
73a4a9a759 Mode few Bluetooth defines into system include files
Reviewed by:	imp
2004-05-10 02:24:56 +00:00
John Polstra
0a4a8041ba Fix a memory leak in ng_get_string_token. A dynamically-allocated
buffer wasn't freed if the function failed.

MFC after:	3 days
2004-04-29 01:37:11 +00:00
Maksim Yevmenkin
b84b10f92f Address few style issues pointed out by bde
Reviewed by:	bde, ru
2004-04-27 16:38:15 +00:00
Archie Cobbs
922ee196d9 Add 'enableWindowing' configuration knob to the ng_pptpgre(4) netgraph node.
Submitted by:	Michael Bretterklieber <mbretter@a-quadrat.at>
MFC after:	2 weeks
2004-04-26 14:26:54 +00:00
Maksim Yevmenkin
48bd0712bc Make sure RFCOMM multiplexor channel does not hang in DISCONNECTING
state. Apparently it happens when both devices try to disconnect RFCOMM
multiplexor channel at the same time.

The scenario is as follows:

- local device initiates RFCOMM connection to the remote device. This
  creates both RFCOMM multiplexor channel and data channel;

- remote device terminates RFCOMM data channel (inactivity timeout);

- local device acknowledges RFCOMM data channel termination. Because
  there is no more active data channels and local device has initiated
  connection it terminates RFCOMM multiplexor channel;

- remote device does not acknowledges RFCOMM multiplexor channel
  termination. Instead it sends its own request to terminate RFCOMM
  multiplexor channel. Even though local device acknowledges RFCOMM
  multiplexor channel termination the remote device still keeps
  L2CAP connection open.

Because of hanging RFCOMM multiplexor channel subsequent RFCOMM
connections between local and remote devices will fail.

Reported by:	Johann Hugo <jhugo@icomtek.csir.co.za>
2004-04-23 20:21:17 +00:00
Luigi Rizzo
212b6d5244 + rename and document an unused field in struct arpcom (field is still
there so there are no ABI changes);
+ replace 5 redefinitions of the IPF2AC macro with one in if_arp.h

Eventually (but before freezing the ABI) we need to get rid of
struct arpcom (initially with the help of some smart #defines
to avoid having to touch each and every driver, see below).

Apart from the struct ifnet, struct arpcom now only stores a copy
of the MAC address (ac_enaddr, but we already have another copy in
the struct ifnet -- if_addrhead), and a netgraph-specific field
which is _always_ accessed through the ifp, so it might well go
into the struct ifnet too (where, besides, there is already an entry
for AF_NETGRAPH data...)

Too bad ac_enaddr is widely referenced by all drivers. But
this can be fixed as follows:

#define ac_enaddr       ac_if.the_original_ac_enaddr_in_struct_ifnet

(note that the right hand side would likely be a pointer rather than
the base address of an array.)
2004-04-18 01:15:32 +00:00
Ruslan Ermilov
dfa515f294 Don't give up if sending to one link fails, continue.
Suggested by:	jmallett
2004-04-17 23:52:57 +00:00
Ruslan Ermilov
b462702f94 A simple packet distribution node type that acts like an Ethernet hub. 2004-04-17 12:42:17 +00:00
Luigi Rizzo
7affdebea2 Consistently use ifaddr_byindex() to access the link-level address
of an interface. No functional change.

On passing, comment an useless invocation of TAILQ_INIT(&ifp->if_addrhead)
which could probably be removed in the interest of clarity.
2004-04-16 08:15:37 +00:00
Maksim Yevmenkin
4ae439a316 Make sure Bluetooth stuff can be compiled on amd64
Submitted by:	ps
2004-04-09 23:01:42 +00:00
Archie Cobbs
2c9027fcba Rename internal structure to fix cut & paste error.
Submitted by:	Bjoern A. Zeeb <bzeeb+freebsd@zabbadoz.net>
MFC After:	3 days
2004-04-04 21:33:09 +00:00
Roman Kurakin
8b96678536 sppp (4) to netgraph (4) node. As always: I'l connect it to the
system after extra check.

Approved by:	imp (mentor)
Approved by:	julian (in general)
2004-03-24 17:24:01 +00:00
Julian Elischer
ba9c6c61b2 Don't duplicate a packet, just to throw it away when we discover the
duplicat hook is not connected.
2004-03-24 08:53:45 +00:00
Julian Elischer
0598ef01ee Dont count bytes being sent to a disconnected hook.
MFC after:	3 days
2004-03-22 01:15:28 +00:00
David Malone
f6fee71d57 Make the vaule of PTT_RELAY_SID match the RFC. This should help PPPoE
users that are working with relayed PPPoE.

Submitted by:	Bodo Rüskamp <jordbaer@mac.com>
PR:		44936
Approved by:	julian
MFC after:	1 week
2004-03-18 12:34:14 +00:00
Nate Lawson
5f96beb9e0 Convert callers to the new bus_alloc_resource_any(9) API.
Submitted by:	Mark Santcroos <marks@ripe.net>
Reviewed by:	imp, dfr, bde
2004-03-17 17:50:55 +00:00
Benno Rice
bde778e9f2 Add a netgraph node to handle ATM LLC encapsulation. This currently handles
ethernet (tested) and FDDI (not tested).  The main use for this is on ADSL (or
other ATM) connections where bridged ethernet is used, PPPoE being a prime
example.

There is no manual page as yet, I will write one shortly.

Reviewed by:	harti
2004-03-08 10:54:35 +00:00
Ruslan Ermilov
1164db57e4 Implemented the "getifindex" control message.
PR:		kern/63864
Submitted by:	Gleb Smirnoff
2004-03-07 23:00:44 +00:00
Ruslan Ermilov
407ea29095 Netgraph node type for IEEE 802.1Q VLAN tagging. 2004-03-01 17:22:16 +00:00
Robert Watson
746e5bf09b Rename dup_sockaddr() to sodupsockaddr() for consistency with other
functions in kern_socket.c.

Rename the "canwait" field to "mflags" and pass M_WAITOK and M_NOWAIT
in from the caller context rather than "1" or "0".

Correct mflags pass into mac_init_socket() from previous commit to not
include M_ZERO.

Submitted by:	sam
2004-03-01 03:14:23 +00:00
Poul-Henning Kamp
917a7daac3 Unconst. Somebody didn't compile LINT. 2004-02-24 22:16:40 +00:00
Colin Percival
3f54070bd2 Check that a pointer is non-NULL before dereferencing it, not after.
Reported by:	"Ted Unangst" <tedu@coverity.com>
Approved by:	rwatson (mentor)
2004-02-24 18:01:43 +00:00
Poul-Henning Kamp
dc08ffec87 Device megapatch 4/6:
Introduce d_version field in struct cdevsw, this must always be
initialized to D_VERSION.

Flip sense of D_NOGIANT flag to D_NEEDGIANT, this involves removing
four D_NOGIANT flags and adding 145 D_NEEDGIANT flags.
2004-02-21 21:10:55 +00:00
Poul-Henning Kamp
c9c7976f7f Device megapatch 1/6:
Free approx 86 major numbers with a mostly automatically generated patch.

A number of strategic drivers have been left behind by caution, and a few
because they still (ab)use their major number.
2004-02-21 19:42:58 +00:00
Pawel Jakub Dawidek
3c976c3f95 Backout previous change, it breaks build and it is not needed
layering violation. As pointed out, there is much better way to do this.
Sorry guys, I need to find a better way to force reviews.

Requested by:	harti, julian, scottl (mentor)
Pointy hat to:	pjd
2004-02-20 08:26:27 +00:00
Pawel Jakub Dawidek
51e9da0539 Add new failure detection algorithm.
It works as follows:
In every 'interval' seconds defined links are checked.
If they are non-active they will not be used by to data transfer.

No response from:	julian, archie
Silent on:		net@
Approved by:		scottl (mentor)
2004-02-19 17:04:23 +00:00
Pawel Jakub Dawidek
d592e95ba7 Export private structure owned by ng_ether(4) module outside.
It'll is required by new failure detection algorithm for ng_one2many(4).

No response from:	julian, archie
Silent on:		net@
Approved by:		scottl (mentor)
2004-02-19 16:58:01 +00:00
Hartmut Brandt
a1adb510b2 Use the official ng_timeout function to trigger sending. This means,
that we can get rid of of all the spl*() calls, because ng_timeout
handles the locking issues.
2004-01-30 15:34:57 +00:00
Ruslan Ermilov
4296e6f8b8 Correct the description of the net.graph.recvspace sysctl. 2004-01-27 22:02:01 +00:00
Hartmut Brandt
4e0dea9a30 Don't confuse NULL and 0, use 0 where an integer is expected. 2004-01-27 10:46:33 +00:00
Hartmut Brandt
76bd585721 Style: add __FBSDID, relocate some { that were on the wrong line,
correct some indendation, change __FUNCTION__ to __func__ and remove
a local KASSERT definition.
2004-01-27 10:45:37 +00:00
Hartmut Brandt
bbee16c0b9 Replace deprecated NG_NODELEN with the new NG_NODESIZ. There is one
problem here still to be solved: the sockaddr_hci has still a 16 byte
field for the node name. The code currently does not correctly use the
length field in the sockaddr to handle the address length, so
node names get truncated to 15 characters when put into a sockaddr_hci.
2004-01-26 15:19:43 +00:00
Hartmut Brandt
211326affc Get rid of the old *LEN constants in favour of the new *SIZ constants
that also include the trailing \0 byte.
2004-01-26 14:57:49 +00:00
Hartmut Brandt
46005fe0eb Define the new command NGM_SOURCE_START_NOW to allow generation of
traffic for non-ethernet hooks. This commit should have been packaged
with the commit to ng_source.c.
2004-01-26 14:54:39 +00:00
Hartmut Brandt
f5d15522f7 Make ng_source to work with non-ethernet interfaces. We do this by
introducing a START_NOW command. This command does not send
and GET_IFINDEX message downstream (to wait for the response from
the ETHERNET node), but directly starts the sending process. This allows
one to generate traffic as input for any hook on any node.
2004-01-26 14:53:16 +00:00
Hartmut Brandt
2cafef3ea4 Declare a function to silence a warning. 2004-01-26 14:48:21 +00:00
Hartmut Brandt
81a4ef8131 Should use the non-locking versions of the ifqueue macros to
fiddle around with private queues, because their mutex is not
needed. All this processing should be protected by the netgraph
locking.
2004-01-26 14:46:35 +00:00
Hartmut Brandt
b1b70498ba Replace a call to bzero() with an M_ZERO flag. Replace the MALLOC() with
malloc().
2004-01-26 14:44:36 +00:00
Hartmut Brandt
4321c5077c The version in the type description must be the ABI version, not
the netgraph version.

Correct the return type of a function: it wants to return an error
code, so it cannot be void.
2004-01-26 14:14:09 +00:00
Hartmut Brandt
87e2c66a6a Get rid of the deprecated *LEN constants in favour of the new
*SIZ constants that include the trailing \0 byte.
2004-01-26 14:05:31 +00:00
Hartmut Brandt
2c858ebee0 Get rid of the deprecated *LEN constants and use the new *SIZ
(that include the trailing \0) constants instead.
2004-01-26 12:24:07 +00:00
Ruslan Ermilov
d2030b65a6 Allow the socket buffer sizes to be controlled via sysctl(8).
MFC after:	3 days
2004-01-23 14:35:44 +00:00
Poul-Henning Kamp
58d120f6d3 Simplify timing gymnastics a bit. 2004-01-21 19:20:57 +00:00
Brian Feldman
7586b25ce3 Add an "ethernet" hook to the rfc1490 netgraph module. It will send
and receive FCS-less RFC1490-"bridged" Ethernet packets that are
currently just ignored.
2004-01-14 00:39:28 +00:00
Brian Feldman
b712e9ec66 Also, don't crash in the netgraph disconnect node if the interface is
detached from the other direction.
2004-01-09 02:03:24 +00:00
Brian Feldman
018df1c369 Don't try to ifpromisc() on an interface that's deleted already. 2004-01-09 00:41:45 +00:00
Sam Leffler
437ffe1823 o eliminate widespread on-stack mbuf use for bpf by introducing
a new bpf_mtap2 routine that does the right thing for an mbuf
  and a variable-length chunk of data that should be prepended.
o while we're sweeping the drivers, use u_int32_t uniformly when
  when prepending the address family (several places were assuming
  sizeof(int) was 4)
o return M_ASSERTVALID to BPF_MTAP* now that all stack-allocated
  mbufs have been eliminated; this may better be moved to the bpf
  routines

Reviewed by:	arch@ and several others
2003-12-28 03:56:00 +00:00