Re: [PATCH] sched: smart wake-affine

[Michael Wang <wangyun@linux.vnet.ibm.com>]

On 07/04/2013 06:33 PM, Mike Galbraith wrote:
[snip]
>> Well, seems like we still have many follow-up research works after fix
>> the issue ;-)
> 
> Yeah.  Like how to how to exterminate the plus sign, they munch cache
> lines, and have a general tendency to negatively impact benchmarks.
> 
> Q6600 box, hackbench -l 1000
>                                                                        avg
> 3.10.0-regress     2.293     2.297     2.313     2.291     2.295     2.297     1.000
> 3.10.0-regressx    2.560     2.524     2.427     2.599     2.602     2.542     1.106

Wow, I used to think such issue is very hard to be tracked by
benchmarks, is this regression stable?

My test could not get a stable differ, this time a little bit lose but
next time a little bit win, it's always floating, may caused by the
different chip cache-behaviour I suppose...

> 
> pahole said...
> 
> marge:/usr/local/src/kernel/linux-3.x.git # tail virgin
>         long unsigned int          timer_slack_ns;       /*  1512     8 */
>         long unsigned int          default_timer_slack_ns; /*  1520     8 */
>         atomic_t                   ptrace_bp_refcnt;     /*  1528     4 */
> 
>         /* size: 1536, cachelines: 24, members: 125 */
>         /* sum members: 1509, holes: 6, sum holes: 23 */
>         /* bit holes: 1, sum bit holes: 26 bits */
>         /* padding: 4 */
>         /* paddings: 1, sum paddings: 4 */
> };
> 
> marge:/usr/local/src/kernel/linux-3.x.git # tail michael
>         long unsigned int          default_timer_slack_ns; /*  1552     8 */
>         atomic_t                   ptrace_bp_refcnt;     /*  1560     4 */
> 
>         /* size: 1568, cachelines: 25, members: 128 */
>         /* sum members: 1533, holes: 8, sum holes: 31 */
>         /* bit holes: 1, sum bit holes: 26 bits */
>         /* padding: 4 */
>         /* paddings: 1, sum paddings: 4 */
>         /* last cacheline: 32 bytes */
> };
> 
> ..but plugging holes, didn't help, moving this/that around neither, nor
> did letting pahole go wild to get the line back.  It's plus signs I tell
> ya, the evil things must die ;-)

Hmm...so the new members kicked some tail members to a new line...or may
be totally different when compiler take part in...

It's really hard to estimate the influence, especially when the
task_struct is still keep changing...

But the task_struct is really a little big now, may be we could put the
'cold' members into a new structure and just record the pointer, that
may increase the chances of cache-hit the hot members, but it's platform
related and not so easy to be detect...

Regards,
Michael Wang

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