On Wed, 31 Oct 2012, Peter LaDow wrote:

On Tue, Oct 30, 2012 at 5:33 PM, Steven Rostedt [off-list ref] wrote:

quoted

From: Thomas Gleixner <redacted>

The netfilter code relies only on the implicit semantics of
local_bh_disable() for serializing wt_write_recseq sections. RT breaks
that and needs explicit serialization here.

Reported-by: Peter LaDow <redacted>
Signed-off-by: Thomas Gleixner <redacted>

quoted

diff --git a/include/linux/locallock.h b/include/linux/locallock.h
index f1804a3..a5eea5d 100644

quoted

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 32cddf7..bed90da2 100644

quoted

diff --git a/net/netfilter/core.c b/net/netfilter/core.c
index 899b71c..5db16ea 100644

I'm trying these out right now.  We've applied these patches to
3.0.36-rt58 (rather than pull in the full 3.0.48-rt72 -- too much risk
for us right now to do a full kernel change).  I am setting up a

Cough. You are missing a boat load of crucial fixes. There is a damned
good reason why 3.0.stable got 12 updates and the -rt version 14.

Your risk assessment is definitley interesting.

Thanks,

	tglx