linux.git/net/ipv4, branch v2.6.34

tcp: fix MD5 (RFC2385) support

2010-05-16T07:34:04+00:00

TCP MD5 support uses percpu data for temporary storage. It currently
disables preemption so that same storage cannot be reclaimed by another
thread on same cpu.

We also have to make sure a softirq handler wont try to use also same
context. Various bug reports demonstrated corruptions.

Fix is to disable preemption and BH.

Reported-by: Bhaskar Dutta 
Signed-off-by: Eric Dumazet 
Signed-off-by: David S. Miller

net: Fix FDDI and TR config checks in ipv4 arp and LLC.

2010-05-10T11:59:07+00:00

Need to check both CONFIG_FOO and CONFIG_FOO_MODULE

Signed-off-by: David S. Miller

IPv4: unresolved multicast route cleanup

2010-05-10T11:47:49+00:00

Fixes the expiration timer for unresolved multicast route entries.
In case new multicast routing requests come in faster than the 
expiration timeout occurs (e.g. zap through multicast TV streams), the 
timer is prevented from being called at time for already existing entries.

As the single timer is resetted to default whenever a new entry is made, 
the timeout for existing unresolved entires are missed and/or not 
updated. As a consequence new requests are denied when the limit of 
unresolved entries has been reached because old entries live longer than 
they are supposed to.

The solution is to reset the timer only for the first unresolved entry 
in the multicast routing cache. All other timers are already set and 
updated correctly within the timer function itself by now.

Signed-off by: Andreas Meissner 
Signed-off-by: David S. Miller

ipv4: udp: fix short packet and bad checksum logging

2010-05-07T04:49:59+00:00

commit 2783ef23 moved the initialisation of saddr and daddr after
pskb_may_pull() to avoid a potential data corruption.  Unfortunately
also placing it after the short packet and bad checksum error paths,
where these variables are used for logging.  The result is bogus
output like

[92238.389505] UDP: short packet: From 2.0.0.0:65535 23715/178 to 0.0.0.0:65535

Moving the saddr and daddr initialisation above the error paths, while still
keeping it after the pskb_may_pull() to keep the fix from commit 2783ef23.

Signed-off-by: Bjørn Mork 
Cc: stable@kernel.org
Acked-by: Eric Dumazet 
Signed-off-by: David S. Miller

Revert "tcp: bind() fix when many ports are bound"

2010-04-28T18:25:59+00:00

This reverts two commits:

fda48a0d7a8412cedacda46a9c0bf8ef9cd13559
tcp: bind() fix when many ports are bound

and a follow-on fix for it:

6443bb1fc2050ca2b6585a3fa77f7833b55329ed
ipv6: Fix inet6_csk_bind_conflict()

It causes problems with binding listening sockets when time-wait
sockets from a previous instance still are alive.

It's too late to keep fiddling with this so late in the -rc
series, and we'll deal with it in net-next-2.6 instead.

Signed-off-by: David S. Miller

tcp: bind() fix when many ports are bound

2010-04-23T02:06:06+00:00

Port autoselection done by kernel only works when number of bound
sockets is under a threshold (typically 30000).

When this threshold is over, we must check if there is a conflict before
exiting first loop in inet_csk_get_port()

Change inet_csk_bind_conflict() to forbid two reuse-enabled sockets to
bind on same (address,port) tuple (with a non ANY address)

Same change for inet6_csk_bind_conflict()

Reported-by: Gaspar Chilingarov 
Signed-off-by: Eric Dumazet 
Acked-by: Evgeniy Polyakov 
Signed-off-by: David S. Miller

ip: Fix ip_dev_loopback_xmit()

2010-04-15T21:25:22+00:00

Eric Paris got following trace with a linux-next kernel

[   14.203970] BUG: using smp_processor_id() in preemptible [00000000]
code: avahi-daemon/2093
[   14.204025] caller is netif_rx+0xfa/0x110
[   14.204035] Call Trace:
[   14.204064]  [] debug_smp_processor_id+0x105/0x110
[   14.204070]  [] netif_rx+0xfa/0x110
[   14.204090]  [] ip_dev_loopback_xmit+0x71/0xa0
[   14.204095]  [] ip_mc_output+0x192/0x2c0
[   14.204099]  [] ip_local_out+0x20/0x30
[   14.204105]  [] ip_push_pending_frames+0x28d/0x3d0
[   14.204119]  [] udp_push_pending_frames+0x14c/0x400
[   14.204125]  [] udp_sendmsg+0x39c/0x790
[   14.204137]  [] inet_sendmsg+0x45/0x80
[   14.204149]  [] sock_sendmsg+0xf1/0x110
[   14.204189]  [] sys_sendmsg+0x20c/0x380
[   14.204233]  [] system_call_fastpath+0x16/0x1b

While current linux-2.6 kernel doesnt emit this warning, bug is latent
and might cause unexpected failures.

ip_dev_loopback_xmit() runs in process context, preemption enabled, so
must call netif_rx_ni() instead of netif_rx(), to make sure that we
process pending software interrupt.

Same change for ip6_dev_loopback_xmit()

Reported-by: Eric Paris 
Signed-off-by: Eric Dumazet 
Signed-off-by: David S. Miller

fib: suppress lockdep-RCU false positive in FIB trie.

2010-04-14T23:13:29+00:00

Followup of commit 634a4b20

Allow tnode_get_child_rcu() to be called either under rcu_read_lock()
protection or with RTNL held.

Signed-off-by: Eric Dumazet 
Signed-off-by: Paul E. McKenney 
Signed-off-by: David S. Miller

Merge branch 'master' of /home/davem/src/GIT/linux-2.6/

2010-04-11T09:44:30+00:00

Revert "tcp: Set CHECKSUM_UNNECESSARY in tcp_init_nondata_skb"

2010-04-11T09:40:49+00:00

This reverts commit 2626419ad5be1a054d350786b684b41d23de1538.

It causes regressions for people with IGB cards.  Connection
requests don't complete etc.  The true cause of the issue is
still not known, but we should sort this out in net-next-2.6
not net-2.6

Signed-off-by: David S. Miller