Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1470510 > unrolled thread

Re: [PATCH v2] locking/percpu-rwsem: Optimize readers and reduce global impact

Started byJohn Stultz <john.stultz@linaro.org>
First post2016-08-26 04:30 +0200
Last post2016-08-26 22:20 +0200
Articles 4 — 4 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH v2] locking/percpu-rwsem: Optimize readers and reduce  global impact John Stultz <john.stultz@linaro.org> - 2016-08-26 04:30 +0200
    Re: [PATCH v2] locking/percpu-rwsem: Optimize readers and reduce  global impact Tejun Heo <tj@kernel.org> - 2016-08-26 15:00 +0200
      Re: [PATCH v2] locking/percpu-rwsem: Optimize readers and reduce  global impact Dmitry Shmidt <dimitrysh@google.com> - 2016-08-26 18:50 +0200
        Re: [PATCH v2] locking/percpu-rwsem: Optimize readers and reduce  global impact Om Dhyade <odhyade@codeaurora.org> - 2016-08-26 22:20 +0200

#1470510 — Re: [PATCH v2] locking/percpu-rwsem: Optimize readers and reduce global impact

FromJohn Stultz <john.stultz@linaro.org>
Date2016-08-26 04:30 +0200
SubjectRe: [PATCH v2] locking/percpu-rwsem: Optimize readers and reduce global impact
Message-ID<saf45-Zl-5@gated-at.bofh.it>
On Wed, Aug 24, 2016 at 2:30 PM, Tejun Heo <tj@kernel.org> wrote:
> Hello, John.
>
> On Wed, Aug 24, 2016 at 02:16:52PM -0700, John Stultz wrote:
>> Hey Peter, Tejun, Oleg,
>>   So while you're tweaks for the percpu-rwsem have greatly helped the
>> regression folks were seeing (many thanks, by the way), as noted
>> above, the performance regression with the global lock compared to
>> earlier kernels is still ~3x slower (though again, much better then
>> the 80x slower that was seen earlier).
>>
>> So I was wondering if patches to go back to the per signal_struct
>> locking would still be considered? Or is the global lock approach the
>> only way forward?
>
> We can't simply revert but we can make the lock per signal_struct
> again.  It's just that it'd be quite a bit more complex (but, again,
> if we need it...) and for cases where migrations aren't as frequent
> percpu-rwsem would be at least a bit lower overhead.  Can you please
> test with the following patch applied just in case?
>
>  https://git.kernel.org/cgit/linux/kernel/git/tj/cgroup.git/commit/?h=for-4.8-fixes&id=568ac888215c7fb2fabe8ea739b00ec3c1f5d440


Hey! Good news. This patch along with Peter's locking changes pushes
the latencies down to an apparently acceptable level!

Many thanks for the pointer!

thanks
-john

[toc] | [next] | [standalone]


#1470697

FromTejun Heo <tj@kernel.org>
Date2016-08-26 15:00 +0200
Message-ID<saoTL-7aM-25@gated-at.bofh.it>
In reply to#1470510
Hello, John.

On Thu, Aug 25, 2016 at 07:14:07PM -0700, John Stultz wrote:
> Hey! Good news. This patch along with Peter's locking changes pushes
> the latencies down to an apparently acceptable level!

Ah, that's good to hear.  Please feel free to ping me if you guys
wanna talk about cgroup usage in android.

Thanks!

-- 
tejun

[toc] | [prev] | [next] | [standalone]


#1470891

FromDmitry Shmidt <dimitrysh@google.com>
Date2016-08-26 18:50 +0200
Message-ID<sasum-18E-29@gated-at.bofh.it>
In reply to#1470697
On Fri, Aug 26, 2016 at 5:51 AM, Tejun Heo <tj@kernel.org> wrote:
> Hello, John.
>
> On Thu, Aug 25, 2016 at 07:14:07PM -0700, John Stultz wrote:
>> Hey! Good news. This patch along with Peter's locking changes pushes
>> the latencies down to an apparently acceptable level!
>
> Ah, that's good to hear.  Please feel free to ping me if you guys
> wanna talk about cgroup usage in android.

Thanks a lot for all your help resolving this issue!

> Thanks!
>
> --
> tejun

[toc] | [prev] | [next] | [standalone]


#1470980

FromOm Dhyade <odhyade@codeaurora.org>
Date2016-08-26 22:20 +0200
Message-ID<savLA-3ml-13@gated-at.bofh.it>
In reply to#1470891

On 8/26/2016 9:47 AM, Dmitry Shmidt wrote:
> On Fri, Aug 26, 2016 at 5:51 AM, Tejun Heo <tj@kernel.org> wrote:
>> Hello, John.
>>
>> On Thu, Aug 25, 2016 at 07:14:07PM -0700, John Stultz wrote:
>>> Hey! Good news. This patch along with Peter's locking changes pushes
>>> the latencies down to an apparently acceptable level!
>>
>> Ah, that's good to hear.  Please feel free to ping me if you guys
>> wanna talk about cgroup usage in android.
>
> Thanks a lot for all your help resolving this issue!
>
Thank you for the help.
>> Thanks!
>>
>> --
>> tejun

-- 
The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web