linux-stable.git/net/core/dev.c, branch linux-2.6.36.y

vlan: Avoid hwaccel vlan packets when vid not used.

2010-12-09T21:33:29+00:00

[This patch applies only to 2.6.36 stable.  The problem was introduced
in that release and is already fixed by larger changes to the vlan
code in 2.6.37.]

Normally hardware accelerated vlan packets are quickly dropped if
there is no corresponding vlan device configured.  The one exception
is promiscuous mode, where we allow all of these packets through so
they can be picked up by tcpdump.  However, this behavior causes a
crash if we actually try to receive these packets.  This fixes that
crash by ignoring packets with vids not corresponding to a configured
device in the vlan hwaccel routines and then dropping them before they
get to consumers in the network stack.


Reported-by: Ben Greear 
Tested-by: Nikola Ciprich 
Signed-off-by: Jesse Gross 
Acked-by: David Miller 
Signed-off-by: Greg Kroah-Hartman

net: NETIF_F_HW_CSUM does not imply FCoE CRC offload

2010-12-09T21:32:37+00:00

commit 66c68bcc489fadd4f5e8839e966e3a366e50d1d5 upstream.

NETIF_F_HW_CSUM indicates the ability to update an TCP/IP-style 16-bit
checksum with the checksum of an arbitrary part of the packet data,
whereas the FCoE CRC is something entirely different.

Signed-off-by: Ben Hutchings 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

net: use rcu_barrier() in rollback_registered_many

2010-09-14T21:27:29+00:00

netdev_wait_allrefs() waits that all references to a device vanishes.

It currently uses a _very_ pessimistic 250 ms delay between each probe.
Some users reported that no more than 4 devices can be dismantled per
second, this is a pretty serious problem for some setups.

Most of the time, a refcount is about to be released by an RCU callback,
that is still in flight because rollback_registered_many() uses a
synchronize_rcu() call instead of rcu_barrier(). Problem is visible if
number of online cpus is one, because synchronize_rcu() is then a no op.

time to remove 50 ipip tunnels on a UP machine :

before patch : real 11.910s
after patch : real 1.250s

Reported-by: Nicolas Dichtel 
Reported-by: Octavian Purdila 
Reported-by: Benjamin LaHaise 
Signed-off-by: Eric Dumazet 
Signed-off-by: David S. Miller

net: fix tx queue selection for bridged devices implementing select_queue

2010-09-07T20:57:20+00:00

When a net device is implementing the select_queue callback and is part of
a bridge, frames coming from the bridge already have a tx queue associated
to the socket (introduced in commit a4ee3ce3293dc931fab19beb472a8bde1295aebe,
"net: Use sk_tx_queue_mapping for connected sockets"). The call to
sk_tx_queue_get will then return the tx queue used by the bridge instead
of calling the select_queue callback.

In case of mac80211 this broke QoS which is implemented by using the
select_queue callback. Furthermore it introduced problems with rt2x00
because frames with the same TID and RA sometimes appeared on different
tx queues which the hw cannot handle correctly.

Fix this by always calling select_queue first if it is available and only
afterwards use the socket tx queue mapping.

Signed-off-by: Helmut Schaa 
Signed-off-by: David S. Miller

net: Fix a memmove bug in dev_gro_receive()

2010-08-18T00:37:28+00:00

>Xin Xiaohui wrote:
> I looked into the code dev_gro_receive(), found the code here:
> if the frags[0] is pulled to 0, then the page will be released,
> and memmove() frags left.
> Is that right? I'm not sure if memmove do right or not, but
> frags[0].size is never set after memove at least. what I think
> a simple way is not to do anything if we found frags[0].size == 0.
> The patch is as followed.
...

This version of the patch fixes the bug directly in memmove.

Reported-by: "Xin, Xiaohui" 
Signed-off-by: Jarek Poplawski 
Signed-off-by: David S. Miller

net: disable preemption before call smp_processor_id()

2010-08-08T03:35:43+00:00

Although netif_rx() isn't expected to be called in process context with
preemption enabled, it'd better handle this case. And this is why get_cpu()
is used in the non-RPS #ifdef branch. If tree RCU is selected,
rcu_read_lock() won't disable preemption, so preempt_disable() should be
called explictly.

Signed-off-by: Changli Gao 
Signed-off-by: David S. Miller

net: Fix napi_gro_frags vs netpoll path

2010-08-05T20:21:25+00:00

The netpoll_rx_on() check in __napi_gro_receive() skips part of the
"common" GRO_NORMAL path, especially "pull:" in dev_gro_receive(),
where at least eth header should be copied for entirely paged skbs.

Signed-off-by: Jarek Poplawski 
Acked-by: Herbert Xu 
Signed-off-by: David S. Miller

Revert "net: remove zap_completion_queue"

2010-08-03T07:24:04+00:00

This reverts commit 15e83ed78864d0625e87a85f09b297c0919a4797.

As explained by Johannes Berg, the optimization made here is
invalid.  Or, at best, incomplete.

Not only destructor invocation, but conntract entry releasing
must be executed outside of hw IRQ context.

So just checking "skb->destructor" is insufficient.

Signed-off-by: David S. Miller

net: cleanup inclusion

2010-08-03T05:45:49+00:00

Commit ab95bfe01f9872459c8678572ccadbf646badad0 replaces bridge and macvlan
hooks in __netif_receive_skb(), so dev.c doesn't need to include their headers.

Signed-off-by: Changli Gao 
Signed-off-by: David S. Miller

net: ingress filter message limit

2010-08-01T07:33:23+00:00

If user misconfigures ingress and causes a redirection loop, don't
overwhelm the log.  This is also a error case so make it unlikely.
Found by inspection, luckily not in real system.

Signed-off-by: Stephen Hemminger 
Signed-off-by: David S. Miller