From: Dave Hansen <dave.hansen@intel.com>
To: Dave Hansen <dave@sr71.net>, Michal Hocko <mhocko@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>,
Hugh Dickins <hughd@google.com>, Tejun Heo <tj@kernel.org>,
Linux-MM <linux-mm@kvack.org>,
Andrew Morton <akpm@linux-foundation.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
Vladimir Davydov <vdavydov@parallels.com>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: regression caused by cgroups optimization in 3.17-rc2
Date: Thu, 04 Sep 2014 15:53:46 -0700 [thread overview]
Message-ID: <5408ED7A.5010908@intel.com> (raw)
In-Reply-To: <5408CB2E.3080101@sr71.net>
On 09/04/2014 01:27 PM, Dave Hansen wrote:
> On 09/04/2014 07:27 AM, Michal Hocko wrote:
>> Ouch. free_pages_and_swap_cache completely kills the uncharge batching
>> because it reduces it to PAGEVEC_SIZE batches.
>>
>> I think we really do not need PAGEVEC_SIZE batching anymore. We are
>> already batching on tlb_gather layer. That one is limited so I think
>> the below should be safe but I have to think about this some more. There
>> is a risk of prolonged lru_lock wait times but the number of pages is
>> limited to 10k and the heavy work is done outside of the lock. If this
>> is really a problem then we can tear LRU part and the actual
>> freeing/uncharging into a separate functions in this path.
>>
>> Could you test with this half baked patch, please? I didn't get to test
>> it myself unfortunately.
>
> 3.16 settled out at about 11.5M faults/sec before the regression. This
> patch gets it back up to about 10.5M, which is good. The top spinlock
> contention in the kernel is still from the resource counter code via
> mem_cgroup_commit_charge(), though.
>
> I'm running Johannes' patch now.
This looks pretty good. The area where it plateaus (above 80 threads
where hyperthreading kicks in) might be a bit slower than it was in
3.16, but that could easily be from other things.
> https://www.sr71.net/~dave/intel/bb.html?1=3.16.0-rc4-g67b9d76/&2=3.17.0-rc3-g57b252f
Feel free to add my Tested-by:
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Dave Hansen <dave.hansen@intel.com>
To: Dave Hansen <dave@sr71.net>, Michal Hocko <mhocko@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>,
Hugh Dickins <hughd@google.com>, Tejun Heo <tj@kernel.org>,
Linux-MM <linux-mm@kvack.org>,
Andrew Morton <akpm@linux-foundation.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
Vladimir Davydov <vdavydov@parallels.com>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: regression caused by cgroups optimization in 3.17-rc2
Date: Thu, 04 Sep 2014 15:53:46 -0700 [thread overview]
Message-ID: <5408ED7A.5010908@intel.com> (raw)
In-Reply-To: <5408CB2E.3080101@sr71.net>
On 09/04/2014 01:27 PM, Dave Hansen wrote:
> On 09/04/2014 07:27 AM, Michal Hocko wrote:
>> Ouch. free_pages_and_swap_cache completely kills the uncharge batching
>> because it reduces it to PAGEVEC_SIZE batches.
>>
>> I think we really do not need PAGEVEC_SIZE batching anymore. We are
>> already batching on tlb_gather layer. That one is limited so I think
>> the below should be safe but I have to think about this some more. There
>> is a risk of prolonged lru_lock wait times but the number of pages is
>> limited to 10k and the heavy work is done outside of the lock. If this
>> is really a problem then we can tear LRU part and the actual
>> freeing/uncharging into a separate functions in this path.
>>
>> Could you test with this half baked patch, please? I didn't get to test
>> it myself unfortunately.
>
> 3.16 settled out at about 11.5M faults/sec before the regression. This
> patch gets it back up to about 10.5M, which is good. The top spinlock
> contention in the kernel is still from the resource counter code via
> mem_cgroup_commit_charge(), though.
>
> I'm running Johannes' patch now.
This looks pretty good. The area where it plateaus (above 80 threads
where hyperthreading kicks in) might be a bit slower than it was in
3.16, but that could easily be from other things.
> https://www.sr71.net/~dave/intel/bb.html?1=3.16.0-rc4-g67b9d76/&2=3.17.0-rc3-g57b252f
Feel free to add my Tested-by:
next prev parent reply other threads:[~2014-09-04 22:55 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-09-02 19:05 regression caused by cgroups optimization in 3.17-rc2 Dave Hansen
2014-09-02 19:05 ` Dave Hansen
2014-09-02 20:18 ` Dave Hansen
2014-09-02 20:57 ` Dave Hansen
2014-09-02 20:57 ` Dave Hansen
2014-09-04 14:27 ` Michal Hocko
2014-09-04 14:27 ` Michal Hocko
2014-09-04 20:27 ` Dave Hansen
2014-09-04 20:27 ` Dave Hansen
2014-09-04 22:53 ` Dave Hansen [this message]
2014-09-04 22:53 ` Dave Hansen
2014-09-05 9:28 ` Michal Hocko
2014-09-05 9:28 ` Michal Hocko
2014-09-05 9:25 ` Michal Hocko
2014-09-05 9:25 ` Michal Hocko
2014-09-05 14:47 ` Johannes Weiner
2014-09-05 14:47 ` Johannes Weiner
2014-09-05 15:39 ` Michal Hocko
2014-09-05 15:39 ` Michal Hocko
2014-09-10 16:29 ` Michal Hocko
2014-09-10 16:29 ` Michal Hocko
2014-09-10 16:57 ` Dave Hansen
2014-09-10 16:57 ` Dave Hansen
2014-09-10 17:05 ` Michal Hocko
2014-09-10 17:05 ` Michal Hocko
2014-09-05 12:35 ` Johannes Weiner
2014-09-05 12:35 ` Johannes Weiner
2014-09-08 15:47 ` Dave Hansen
2014-09-08 15:47 ` Dave Hansen
2014-09-09 14:50 ` Johannes Weiner
2014-09-09 14:50 ` Johannes Weiner
2014-09-09 18:23 ` Dave Hansen
2014-09-09 18:23 ` Dave Hansen
2014-09-02 22:18 ` Johannes Weiner
2014-09-02 22:18 ` Johannes Weiner
2014-09-02 22:36 ` Dave Hansen
2014-09-03 0:10 ` Johannes Weiner
2014-09-03 0:10 ` Johannes Weiner
2014-09-03 0:20 ` Linus Torvalds
2014-09-03 0:20 ` Linus Torvalds
2014-09-03 1:33 ` Johannes Weiner
2014-09-03 1:33 ` Johannes Weiner
2014-09-03 3:15 ` Dave Hansen
2014-09-03 3:15 ` Dave Hansen
2014-09-03 0:30 ` Dave Hansen
2014-09-03 0:30 ` Dave Hansen
2014-09-04 15:08 ` Johannes Weiner
2014-09-04 15:08 ` Johannes Weiner
2014-09-04 20:50 ` Dave Hansen
2014-09-04 20:50 ` Dave Hansen
2014-09-05 8:04 ` Michal Hocko
2014-09-05 8:04 ` Michal Hocko
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5408ED7A.5010908@intel.com \
--to=dave.hansen@intel.com \
--cc=akpm@linux-foundation.org \
--cc=dave@sr71.net \
--cc=hannes@cmpxchg.org \
--cc=hughd@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.cz \
--cc=tj@kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=vdavydov@parallels.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.