Re: [netfilter bug] BUG: using smp_processor_id() in preemptible [00000000] code: ssh/9115, caller is ipt_do_table+0xc8/0x559

[Ingo Molnar <mingo@elte.hu>]

* Eric Dumazet <dada1@cosmosbay.com> wrote:

> Ingo Molnar a écrit :
> > Not sure whether this bug has been reported before (a quick subject 
> > search on netdev and lkml suggests that it's not):
> > 
> > i switched over -tip to latest -git an hour ago (it was on a v2.6.29 
> > base before that), and one of my testboxes started emitting lots of 
> > these warnings:
> 
> 
> Hi Ingo
> 
> David put into its tree fix for that a few hours ago
> 
> commit fa9a86ddc8ecd2830a5e773facc250f110300ae7
> 
> (netfilter: iptables: lock free counters) forgot to disable BH
> in arpt_do_table(), ipt_do_table() and  ip6t_do_table()
> 
> Use rcu_read_lock_bh() instead of rcu_read_lock() cures the problem.
> 
> Reported-and-bisected-by: Roman Mindalev <r000n@r000n.net>
> Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
> Acked-by: Patrick McHardy <kaber@trash.net>
> Acked-by: Stephen Hemminger <shemminger@vyatta.com>
> Signed-off-by: David S. Miller <davem@davemloft.net>
> 
> 
> Should be pushed to Linus in next round
> 
> Thank you
> 
> http://git2.kernel.org/?p=linux/kernel/git/davem/net-2.6.git;a=commitdiff;h=fa9a86ddc8ecd2830a5e773facc250f110300ae7

Small request: it's useful to plain-text include such commits in 
such email replies - if the patch is not too large. http is not very 
helpful to reply to specifics and it's not very useful when i want 
toapply it either. I now have to go hunting for the latest tree :)

	Ingo