lkml.org 
[lkml]   [2021]   [Sep]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [memcg] 45208c9105: aim7.jobs-per-min -14.0% regression
On Mon, Sep 6, 2021 at 8:30 PM Feng Tang <feng.tang@intel.com> wrote:
>
> Hi Shakeel,
>
> On Sun, Sep 05, 2021 at 03:15:46PM -0700, Shakeel Butt wrote:
> > On Sun, Sep 5, 2021 at 5:27 AM kernel test robot <oliver.sang@intel.com> wrote:
> [...]
> > > =========================================================================================
> > > compiler/cpufreq_governor/disk/fs/kconfig/load/rootfs/tbox_group/test/testcase/ucode:
> > > gcc-9/performance/1BRD_48G/xfs/x86_64-rhel-8.3/3000/debian-10.4-x86_64-20200603.cgz/lkp-icl-2sp2/disk_rr/aim7/0xd000280
> > >
> > > commit:
> > > 3c28c7680e ("memcg: switch lruvec stats to rstat")
> > > 45208c9105 ("memcg: infrastructure to flush memcg stats")
> >
> > I am looking into this. I was hoping we have resolution for [1] as
> > these patches touch similar data structures.
> >
> > [1] https://lore.kernel.org/all/20210811031734.GA5193@xsang-OptiPlex-9020/T/#u
>
> I tried 2 debug methods for that 36.4% vm-scalability regression:
>
> 1. Disable the HW cache prefetcher, no effect on this case
> 2. relayout and add padding to 'struct cgroup_subsys_state', reduce
> the regression to 3.1%
>

Thanks Feng but it seems like the issue for this commit is different.
Rearranging the layout didn't help. Actually the cause of slowdown is
the call to queue_work() inside __mod_memcg_lruvec_state().

At the moment, queue_work() is called after 32 updates. I changed it
to 128 and the slowdown of will-it-scale:page_fault[1|2|3] halved
(from around 10% to 5%). I am unable to run reaim or
will-it-scale:fallocate2 as I was getting weird errors.

Feng, is it possible for you to run these benchmarks with the change
(basically changing MEMCG_CHARGE_BATCH to 128 in the if condition
before queue_work() inside __mod_memcg_lruvec_state())?

For the formal patch/fix, I will write down a better explanation on
what should be the batch size.

thanks,
Shakeel

\
 
 \ /
  Last update: 2021-09-10 03:07    [W:2.297 / U:0.316 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site