All of lore.kernel.org
 help / color / mirror / Atom feed
From: Minchan Kim <minchan@kernel.org>
To: Vlastimil Babka <vbabka@suse.cz>
Cc: linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>,
	David Rientjes <rientjes@google.com>,
	Mel Gorman <mgorman@suse.de>,
	Joonsoo Kim <iamjoonsoo.kim@lge.com>,
	Michal Nazarewicz <mina86@mina86.com>,
	Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>,
	Christoph Lameter <cl@linux.com>, Rik van Riel <riel@redhat.com>,
	Zhang Yanfei <zhangyanfei@cn.fujitsu.com>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3 09/13] mm, compaction: skip buddy pages by their order in the migrate scanner
Date: Mon, 23 Jun 2014 12:05:41 +0900	[thread overview]
Message-ID: <20140623030541.GE12413@bbox> (raw)
In-Reply-To: <1403279383-5862-10-git-send-email-vbabka@suse.cz>

On Fri, Jun 20, 2014 at 05:49:39PM +0200, Vlastimil Babka wrote:
> The migration scanner skips PageBuddy pages, but does not consider their order
> as checking page_order() is generally unsafe without holding the zone->lock,
> and acquiring the lock just for the check wouldn't be a good tradeoff.
> 
> Still, this could avoid some iterations over the rest of the buddy page, and
> if we are careful, the race window between PageBuddy() check and page_order()
> is small, and the worst thing that can happen is that we skip too much and miss
> some isolation candidates. This is not that bad, as compaction can already fail
> for many other reasons like parallel allocations, and those have much larger
> race window.
> 
> This patch therefore makes the migration scanner obtain the buddy page order
> and use it to skip the whole buddy page, if the order appears to be in the
> valid range.
> 
> It's important that the page_order() is read only once, so that the value used
> in the checks and in the pfn calculation is the same. But in theory the
> compiler can replace the local variable by multiple inlines of page_order().
> Therefore, the patch introduces page_order_unsafe() that uses ACCESS_ONCE to
> prevent this.
> 
> Testing with stress-highalloc from mmtests shows a 15% reduction in number of
> pages scanned by migration scanner. This change is also a prerequisite for a
> later patch which is detecting when a cc->order block of pages contains
> non-buddy pages that cannot be isolated, and the scanner should thus skip to
> the next block immediately.
> 
> Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Minchan Kim <minchan@kernel.org>

-- 
Kind regards,
Minchan Kim

--
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: Minchan Kim <minchan@kernel.org>
To: Vlastimil Babka <vbabka@suse.cz>
Cc: linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>,
	David Rientjes <rientjes@google.com>,
	Mel Gorman <mgorman@suse.de>,
	Joonsoo Kim <iamjoonsoo.kim@lge.com>,
	Michal Nazarewicz <mina86@mina86.com>,
	Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>,
	Christoph Lameter <cl@linux.com>, Rik van Riel <riel@redhat.com>,
	Zhang Yanfei <zhangyanfei@cn.fujitsu.com>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3 09/13] mm, compaction: skip buddy pages by their order in the migrate scanner
Date: Mon, 23 Jun 2014 12:05:41 +0900	[thread overview]
Message-ID: <20140623030541.GE12413@bbox> (raw)
In-Reply-To: <1403279383-5862-10-git-send-email-vbabka@suse.cz>

On Fri, Jun 20, 2014 at 05:49:39PM +0200, Vlastimil Babka wrote:
> The migration scanner skips PageBuddy pages, but does not consider their order
> as checking page_order() is generally unsafe without holding the zone->lock,
> and acquiring the lock just for the check wouldn't be a good tradeoff.
> 
> Still, this could avoid some iterations over the rest of the buddy page, and
> if we are careful, the race window between PageBuddy() check and page_order()
> is small, and the worst thing that can happen is that we skip too much and miss
> some isolation candidates. This is not that bad, as compaction can already fail
> for many other reasons like parallel allocations, and those have much larger
> race window.
> 
> This patch therefore makes the migration scanner obtain the buddy page order
> and use it to skip the whole buddy page, if the order appears to be in the
> valid range.
> 
> It's important that the page_order() is read only once, so that the value used
> in the checks and in the pfn calculation is the same. But in theory the
> compiler can replace the local variable by multiple inlines of page_order().
> Therefore, the patch introduces page_order_unsafe() that uses ACCESS_ONCE to
> prevent this.
> 
> Testing with stress-highalloc from mmtests shows a 15% reduction in number of
> pages scanned by migration scanner. This change is also a prerequisite for a
> later patch which is detecting when a cc->order block of pages contains
> non-buddy pages that cannot be isolated, and the scanner should thus skip to
> the next block immediately.
> 
> Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Minchan Kim <minchan@kernel.org>

-- 
Kind regards,
Minchan Kim

  reply	other threads:[~2014-06-23  3:04 UTC|newest]

Thread overview: 114+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-06-20 15:49 [PATCH v3 00/13] compaction: balancing overhead and success rates Vlastimil Babka
2014-06-20 15:49 ` Vlastimil Babka
2014-06-20 15:49 ` [PATCH v3 01/13] mm, THP: don't hold mmap_sem in khugepaged when allocating THP Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-20 17:45   ` Kirill A. Shutemov
2014-06-20 17:45     ` Kirill A. Shutemov
2014-06-23  5:39     ` Zhang Yanfei
2014-06-23  5:39       ` Zhang Yanfei
2014-06-23  9:52       ` Vlastimil Babka
2014-06-23  9:52         ` Vlastimil Babka
2014-06-23 10:40         ` Zhang Yanfei
2014-06-23 10:40           ` Zhang Yanfei
2014-06-20 15:49 ` [PATCH v3 02/13] mm, compaction: defer each zone individually instead of preferred zone Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-23  2:24   ` Minchan Kim
2014-06-23  2:24     ` Minchan Kim
2014-06-23  6:26   ` Zhang Yanfei
2014-06-23  6:26     ` Zhang Yanfei
2014-06-24  8:23   ` Joonsoo Kim
2014-06-24  8:23     ` Joonsoo Kim
2014-06-24 15:29     ` Vlastimil Babka
2014-06-24 15:29       ` Vlastimil Babka
2014-06-25  1:02       ` Joonsoo Kim
2014-06-25  1:02         ` Joonsoo Kim
2014-06-20 15:49 ` [PATCH v3 03/13] mm, compaction: do not recheck suitable_migration_target under lock Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-20 15:49 ` [PATCH v3 04/13] mm, compaction: move pageblock checks up from isolate_migratepages_range() Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-23  6:57   ` Zhang Yanfei
2014-06-23  6:57     ` Zhang Yanfei
2014-06-24  4:52   ` Naoya Horiguchi
2014-06-24  4:52     ` Naoya Horiguchi
2014-06-24 15:34     ` Vlastimil Babka
2014-06-24 15:34       ` Vlastimil Babka
2014-06-24 16:58       ` Naoya Horiguchi
2014-06-24 16:58         ` Naoya Horiguchi
2014-06-25  8:50         ` Vlastimil Babka
2014-06-25  8:50           ` Vlastimil Babka
2014-06-25 15:46           ` Naoya Horiguchi
2014-06-25 15:46             ` Naoya Horiguchi
2014-06-24  8:33   ` Joonsoo Kim
2014-06-24  8:33     ` Joonsoo Kim
2014-06-24 15:42     ` Vlastimil Babka
2014-06-24 15:42       ` Vlastimil Babka
2014-06-25  0:53       ` Joonsoo Kim
2014-06-25  0:53         ` Joonsoo Kim
2014-06-25  8:59         ` Vlastimil Babka
2014-06-25  8:59           ` Vlastimil Babka
2014-06-27  5:57           ` Joonsoo Kim
2014-06-27  5:57             ` Joonsoo Kim
2014-06-20 15:49 ` [PATCH v3 05/13] mm, compaction: report compaction as contended only due to lock contention Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-23  1:39   ` Minchan Kim
2014-06-23  1:39     ` Minchan Kim
2014-06-23  8:55     ` Zhang Yanfei
2014-06-23  8:55       ` Zhang Yanfei
2014-06-23 23:35       ` Minchan Kim
2014-06-23 23:35         ` Minchan Kim
2014-06-24  1:07         ` Zhang Yanfei
2014-06-24  1:07           ` Zhang Yanfei
2014-07-11  8:28     ` Vlastimil Babka
2014-07-11  8:28       ` Vlastimil Babka
2014-07-11  9:38       ` Vlastimil Babka
2014-07-11  9:38         ` Vlastimil Babka
2014-06-20 15:49 ` [PATCH v3 06/13] mm, compaction: periodically drop lock and restore IRQs in scanners Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-23  2:53   ` Minchan Kim
2014-06-23  2:53     ` Minchan Kim
2014-07-11 12:03     ` Vlastimil Babka
2014-07-11 12:03       ` Vlastimil Babka
2014-06-23  9:13   ` Zhang Yanfei
2014-06-23  9:13     ` Zhang Yanfei
2014-06-24 15:39   ` Naoya Horiguchi
2014-06-24 15:39     ` Naoya Horiguchi
2014-06-24 15:44     ` Vlastimil Babka
2014-06-24 15:44       ` Vlastimil Babka
2014-06-20 15:49 ` [PATCH v3 07/13] mm, compaction: skip rechecks when lock was already held Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-23  9:16   ` Zhang Yanfei
2014-06-23  9:16     ` Zhang Yanfei
2014-06-24 18:55   ` Naoya Horiguchi
2014-06-24 18:55     ` Naoya Horiguchi
2014-06-20 15:49 ` [PATCH v3 08/13] mm, compaction: remember position within pageblock in free pages scanner Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-23  3:04   ` Minchan Kim
2014-06-23  3:04     ` Minchan Kim
2014-06-23  9:17   ` Zhang Yanfei
2014-06-23  9:17     ` Zhang Yanfei
2014-06-24 19:09   ` Naoya Horiguchi
2014-06-24 19:09     ` Naoya Horiguchi
2014-06-20 15:49 ` [PATCH v3 09/13] mm, compaction: skip buddy pages by their order in the migrate scanner Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-23  3:05   ` Minchan Kim [this message]
2014-06-23  3:05     ` Minchan Kim
2014-06-23  9:29   ` Zhang Yanfei
2014-06-23  9:29     ` Zhang Yanfei
2014-06-20 15:49 ` [PATCH v3 10/13] mm: rename allocflags_to_migratetype for clarity Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-24 20:34   ` Naoya Horiguchi
2014-06-24 20:34     ` Naoya Horiguchi
2014-06-20 15:49 ` [PATCH v3 11/13] mm, compaction: pass gfp mask to compact_control Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-23  3:06   ` Minchan Kim
2014-06-23  3:06     ` Minchan Kim
2014-06-23  9:31   ` Zhang Yanfei
2014-06-23  9:31     ` Zhang Yanfei
2014-06-20 15:49 ` [PATCH v3 12/13] mm, compaction: try to capture the just-created high-order freepage Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka
2014-06-25  1:57   ` Naoya Horiguchi
2014-06-25  1:57     ` Naoya Horiguchi
2014-06-25  8:57     ` Vlastimil Babka
2014-06-25  8:57       ` Vlastimil Babka
2014-06-20 15:49 ` [RFC PATCH v3 13/13] mm, compaction: do not migrate pages when that cannot satisfy page fault allocation Vlastimil Babka
2014-06-20 15:49   ` Vlastimil Babka

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=20140623030541.GE12413@bbox \
    --to=minchan@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=cl@linux.com \
    --cc=iamjoonsoo.kim@lge.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    --cc=mina86@mina86.com \
    --cc=n-horiguchi@ah.jp.nec.com \
    --cc=riel@redhat.com \
    --cc=rientjes@google.com \
    --cc=vbabka@suse.cz \
    --cc=zhangyanfei@cn.fujitsu.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.