From: Dave Hansen <dave.hansen@linux.intel.com>
To: Lukasz Anaczkowski <lukasz.anaczkowski@intel.com>,
hpa@zytor.com, mingo@redhat.com, tglx@linutronix.de,
ak@linux.intel.com, kirill.shutemov@linux.intel.com,
mhocko@suse.com, akpm@linux-foundation.org,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Cc: harish.srinivasappa@intel.com, lukasz.odzioba@intel.com,
grzegorz.andrejczuk@intel.com, lukasz.daniluk@intel.com
Subject: Re: [PATCH v3] Linux VM workaround for Knights Landing A/D leak
Date: Thu, 16 Jun 2016 13:23:58 -0700 [thread overview]
Message-ID: <57630ADE.3040900@linux.intel.com> (raw)
In-Reply-To: <1466090042-30908-1-git-send-email-lukasz.anaczkowski@intel.com>
On 06/16/2016 08:14 AM, Lukasz Anaczkowski wrote:
> For reclaim this brings the performance back to before Mel's
> flushing changes, but for unmap it disables batching.
This turns out to be pretty catastrophic for unmap. In a workload that
uses, say 200 hardware threads and alloc/frees() a few MB/sec, this ends
up costing hundreds of thousands of extra received IPIs. 10MB=~2500
ptes, and at with 200 threads, that's 250,000 IPIs received just to free
10MB of memory.
The initial testing we did on this was on a *bunch* of threads all doing
alloc/free. But this is bottlenecked on other things, like mmap_sem
being held for write.
The scenario that we really needed to test here was on lots of threads
doing processing and 1 thread doing alloc/free.
--
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@linux.intel.com>
To: Lukasz Anaczkowski <lukasz.anaczkowski@intel.com>,
hpa@zytor.com, mingo@redhat.com, tglx@linutronix.de,
ak@linux.intel.com, kirill.shutemov@linux.intel.com,
mhocko@suse.com, akpm@linux-foundation.org,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Cc: harish.srinivasappa@intel.com, lukasz.odzioba@intel.com,
grzegorz.andrejczuk@intel.com, lukasz.daniluk@intel.com
Subject: Re: [PATCH v3] Linux VM workaround for Knights Landing A/D leak
Date: Thu, 16 Jun 2016 13:23:58 -0700 [thread overview]
Message-ID: <57630ADE.3040900@linux.intel.com> (raw)
In-Reply-To: <1466090042-30908-1-git-send-email-lukasz.anaczkowski@intel.com>
On 06/16/2016 08:14 AM, Lukasz Anaczkowski wrote:
> For reclaim this brings the performance back to before Mel's
> flushing changes, but for unmap it disables batching.
This turns out to be pretty catastrophic for unmap. In a workload that
uses, say 200 hardware threads and alloc/frees() a few MB/sec, this ends
up costing hundreds of thousands of extra received IPIs. 10MB=~2500
ptes, and at with 200 threads, that's 250,000 IPIs received just to free
10MB of memory.
The initial testing we did on this was on a *bunch* of threads all doing
alloc/free. But this is bottlenecked on other things, like mmap_sem
being held for write.
The scenario that we really needed to test here was on lots of threads
doing processing and 1 thread doing alloc/free.
next prev parent reply other threads:[~2016-06-16 20:24 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-14 15:58 [PATCH] Linux VM workaround for Knights Landing A/D leak Lukasz Anaczkowski
2016-06-14 15:58 ` Lukasz Anaczkowski
2016-06-14 16:31 ` kbuild test robot
2016-06-14 16:31 ` kbuild test robot
2016-06-14 16:47 ` Nadav Amit
2016-06-14 16:47 ` Nadav Amit
2016-06-14 16:54 ` Anaczkowski, Lukasz
2016-06-14 16:54 ` Anaczkowski, Lukasz
2016-06-14 17:01 ` [PATCH v2] " Lukasz Anaczkowski
2016-06-14 17:01 ` Lukasz Anaczkowski
2016-06-14 17:24 ` Dave Hansen
2016-06-14 17:24 ` Dave Hansen
2016-06-14 18:34 ` One Thousand Gnomes
2016-06-14 18:34 ` One Thousand Gnomes
2016-06-14 18:54 ` Dave Hansen
2016-06-14 18:54 ` Dave Hansen
2016-06-14 19:19 ` Borislav Petkov
2016-06-14 19:19 ` Borislav Petkov
2016-06-14 20:20 ` H. Peter Anvin
2016-06-14 20:47 ` Borislav Petkov
2016-06-14 20:54 ` H. Peter Anvin
2016-06-14 21:02 ` Borislav Petkov
2016-06-14 21:08 ` H. Peter Anvin
2016-06-14 21:13 ` H. Peter Anvin
2016-06-14 18:10 ` Borislav Petkov
2016-06-14 18:10 ` Borislav Petkov
2016-06-15 13:12 ` Anaczkowski, Lukasz
2016-06-15 13:12 ` Anaczkowski, Lukasz
2016-06-14 18:38 ` Nadav Amit
2016-06-14 18:38 ` Nadav Amit
2016-06-15 13:12 ` Anaczkowski, Lukasz
2016-06-15 13:12 ` Anaczkowski, Lukasz
2016-06-15 20:04 ` Nadav Amit
2016-06-15 20:04 ` Nadav Amit
2016-06-15 20:10 ` Dave Hansen
2016-06-15 20:10 ` Dave Hansen
2016-06-15 20:26 ` Nadav Amit
2016-06-15 20:26 ` Nadav Amit
2016-06-16 15:14 ` [PATCH v3] " Lukasz Anaczkowski
2016-06-16 15:14 ` Lukasz Anaczkowski
2016-06-16 16:43 ` Nadav Amit
2016-06-16 16:43 ` Nadav Amit
2016-06-16 20:23 ` Dave Hansen [this message]
2016-06-16 20:23 ` Dave Hansen
2016-06-14 17:18 ` [PATCH] " Dave Hansen
2016-06-14 17:18 ` Dave Hansen
2016-06-14 20:16 ` Nadav Amit
2016-06-14 20:16 ` Nadav Amit
2016-06-14 21:37 ` Dave Hansen
2016-06-14 21:37 ` Dave Hansen
2016-06-15 2:20 ` Andy Lutomirski
2016-06-15 2:20 ` Andy Lutomirski
2016-06-15 2:35 ` Nadav Amit
2016-06-15 2:35 ` Nadav Amit
2016-06-15 2:36 ` Andy Lutomirski
2016-06-15 2:36 ` Andy Lutomirski
2016-06-15 2:44 ` Nadav Amit
2016-06-15 2:44 ` Nadav Amit
2016-06-15 3:09 ` Andy Lutomirski
2016-06-15 3:09 ` Andy Lutomirski
2016-06-15 3:20 ` Nadav Amit
2016-06-15 3:20 ` Nadav Amit
2016-06-14 16:58 ` kbuild test robot
2016-06-14 16:58 ` kbuild test robot
2016-06-14 17:19 ` Dave Hansen
2016-06-14 17:19 ` Dave Hansen
2016-06-15 13:06 ` Anaczkowski, Lukasz
2016-06-15 13:06 ` Anaczkowski, Lukasz
2016-06-14 17:47 ` kbuild test robot
2016-06-14 17:47 ` kbuild test robot
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=57630ADE.3040900@linux.intel.com \
--to=dave.hansen@linux.intel.com \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=grzegorz.andrejczuk@intel.com \
--cc=harish.srinivasappa@intel.com \
--cc=hpa@zytor.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lukasz.anaczkowski@intel.com \
--cc=lukasz.daniluk@intel.com \
--cc=lukasz.odzioba@intel.com \
--cc=mhocko@suse.com \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
/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.