Re: [Bug #11308] tbench regression on each kernel release from 2.6.22-> 2.6.28

[Eric Dumazet]

Mike Galbraith a écrit :
> On Wed, 2008-09-17 at 14:49 +0200, Ingo Molnar wrote:
> > * Mike Galbraith <efault@xxxxxx> wrote:
> >
> > > > It looks like a potentially bogus bisection result, but _maybe_ it 
> > > > has relevance: changes the size of "struct security_operations", 
> > > > which could have alignment and layout effects on all sorts of kernel 
> > > > variables, kmalloc sizes, etc.
> > > This may well be a mythical creature infestation for all I know ;-), 
> > > but it's address is somewhere in the 2069f45..847106f block, 316 
> > > commits, none of which look like they should be the least bit 
> > > interesting to netperf.  I reverted this particular commit in 27.git, 
> > > got the expected result.  Looks like I'll keep poking at it, can't 
> > > seem to resist.  Grr.
> > are you sure it's 2069f45..847106f? Filtering out the 
> > likely-uninteresting commits:
>
> Yeah, as sure as I can be.  I've built both (et al) kernels several
> times, and it has repeated every time.  Would be nice if someone would
> try to confirm/deny though.  For my little quad, I do..
>
> #!/bin/sh
>
> echo 0 > /proc/sys/kernel/sched_wakeup_granularity_ns 
>
> netserver -p 12865
> netserver -p 12866
> netserver -p 12867
> netserver -p 12868
>
> netperf -p 12865 -t TCP_RR -l 60 -H 127.0.0.1 -T 0,0 -- -r 1,1&
> netperf -p 12866 -t TCP_RR -l 60 -H 127.0.0.1 -T 1,1 -- -r 1,1&
> netperf -p 12867 -t TCP_RR -l 60 -H 127.0.0.1 -T 2,2 -- -r 1,1&
> netperf -p 12868 -t TCP_RR -l 60 -H 127.0.0.1 -T 3,3 -- -r 1,1&
>
> wait
> killall netserver
>
>
> > git log --pretty=format:"%h: %s" 2069f45..847106f | grep -viE \
> > 'block|alsa|pcmcia|sound|Merge|iosched|blk|DAC960|scsi|s390|paride|pktcdvd|filter|cdrom|drm'
> >
> > gives us:
> >
> >  7daf705: Start using the new '%pS' infrastructure to print symbols
> >  6f0f0fd: security: remove register_security hook
> >  93cbace: security: remove dummy module fix
> >  5915eb5: security: remove dummy module
> >  b478a9f: security: remove unused sb_get_mnt_opts hook
> >  32502b8: splice: fix generic_file_splice_read() race with page invalidation
> >  8b3d356: ramfs: enable splice write
> >  a144ff0: xen: Avoid allocations causing swap activity on the resume path
> >
> > which really only leaves that security commit your bisection fingered. 
> > Which _slightly_ raises its likelyhood of being implicated. Structure 
> > size changes can move two formerly far-apart netperf-relevant symbols on 
> > the same cacheline, which can start cache ping-pong-ing badly.
>
> I sure hope it's something like ping-pong, it's driving me NUTS.

Could you please try following patch ?

[PATCH] security_ops moved to read_mostly section

"struct security_operations *security_ops" should be moved to read_mostly 
section in order to NOT let it share a cache line with higly modified variables.

Signed-off-by: Eric Dumazet <dada1@xxxxxxxxxxxxx>

diff --git a/security/security.c b/security/security.c
index 3a4b4f5..0b13d65 100644
--- a/security/security.c
+++ b/security/security.c
@@ -24,7 +24,7 @@ static __initdata char chosen_lsm[SECURITY_NAME_MAX + 1];
extern struct security_operations default_security_ops;
extern void security_fixup_ops(struct security_operations *ops);

-struct security_operations *security_ops;	/* Initialized to NULL */
+struct security_operations *security_ops __read_mostly;e

/* amount of vm to protect from userspace access */
unsigned long mmap_min_addr = CONFIG_SECURITY_DEFAULT_MMAP_MIN_ADDR;



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/