From: Marek Szyprowski <m.szyprowski@samsung.com>
To: "'Michal Nazarewicz'" <mina86@mina86.com>,
"'Mel Gorman'" <mel@csn.ul.ie>
Cc: linux-kernel@vger.kernel.org,
linux-arm-kernel@lists.infradead.org,
linux-media@vger.kernel.org, linux-mm@kvack.org,
linaro-mm-sig@lists.linaro.org,
"'Kyungmin Park'" <kyungmin.park@samsung.com>,
"'Russell King'" <linux@arm.linux.org.uk>,
"'Andrew Morton'" <akpm@linux-foundation.org>,
"'KAMEZAWA Hiroyuki'" <kamezawa.hiroyu@jp.fujitsu.com>,
"'Daniel Walker'" <dwalker@codeaurora.org>,
"'Arnd Bergmann'" <arnd@arndb.de>,
"'Jesse Barker'" <jesse.barker@linaro.org>,
"'Jonathan Corbet'" <corbet@lwn.net>,
"'Shariq Hasnain'" <shariq.hasnain@linaro.org>,
"'Chunsang Jeong'" <chunsang.jeong@linaro.org>,
"'Dave Hansen'" <dave@linux.vnet.ibm.com>,
"'Benjamin Gaignard'" <benjamin.gaignard@linaro.org>
Subject: RE: [PATCH 02/15] mm: page_alloc: update migrate type of pages on pcp when isolating
Date: Fri, 03 Feb 2012 10:31:50 +0100 [thread overview]
Message-ID: <008101cce256$a85e29f0$f91a7dd0$%szyprowski@samsung.com> (raw)
In-Reply-To: <op.v82hjbd13l0zgt@mpn-glaptop>
Hello,
On Thursday, February 02, 2012 8:53 PM Michał Nazarewicz wrote:
> > On Tue, Jan 31, 2012 at 05:23:59PM +0100, Marek Szyprowski wrote:
> >> Pages, which have incorrect migrate type on free finally
> >> causes pageblock migration type change from MIGRATE_CMA to MIGRATE_MOVABLE.
>
> On Thu, 02 Feb 2012 13:47:29 +0100, Mel Gorman <mel@csn.ul.ie> wrote:
> > I'm not quite seeing this. In free_hot_cold_page(), the pageblock
> > type is checked so the page private should be set to MIGRATE_CMA or
> > MIGRATE_ISOLATE for the CMA area. It's not clear how this can change a
> > pageblock to MIGRATE_MOVABLE in error.
>
> Here's what I think may happen:
>
> When drain_all_pages() is called, __free_one_page() is called for each page on
> pcp list with migrate type deducted from page_private() which is MIGRATE_CMA.
> This result in the page being put on MIGRATE_CMA freelist even though its
> pageblock's migrate type is MIGRATE_ISOLATE.
>
> When allocation happens and pcp list is empty, rmqueue_bulk() will get executed
> with migratetype argument set to MIGRATE_MOVABLE. It calls __rmqueue() to grab
> some pages and because the page described above is on MIGRATE_CMA freelist it
> may be returned back to rmqueue_bulk().
>
> But, pageblock's migrate type is not MIGRATE_CMA but MIGRATE_ISOLATE, so the
> following code:
>
> #ifdef CONFIG_CMA
> if (is_pageblock_cma(page))
> set_page_private(page, MIGRATE_CMA);
> else
> #endif
> set_page_private(page, migratetype);
>
> will set it's private to MIGRATE_MOVABLE and in the end the page lands back
> on MIGRATE_MOVABLE pcp list but this time with page_private == MIGRATE_MOVABLE
> and not MIGRATE_CMA.
>
> One more drain_all_pages() (which may happen since alloc_contig_range() calls
> set_migratetype_isolate() for each block) and next __rmqueue_fallback() may
> convert the whole pageblock to MIGRATE_MOVABLE.
>
> I know, this sounds crazy and improbable, but I couldn't find an easier path
> to destruction. As you pointed, once the page is allocated, free_hot_cold_page()
> will do the right thing by reading pageblock's migrate type.
>
> Marek is currently experimenting with various patches including the following
> change:
>
> #ifdef CONFIG_CMA
> int mt = get_pageblock_migratetype(page);
> if (is_migrate_cma(mt) || mt == MIGRATE_ISOLATE)
> set_page_private(page, mt);
> else
> #endif
> set_page_private(page, migratetype);
>
> As a matter of fact, if __rmqueue() was changed to return migrate type of the
> freelist it took page from, we could avoid this get_pageblock_migratetype() all
> together. For now, however, I'd rather not go that way just yet -- I'll be happy
> to dig into it once CMA gets merged.
After this and some other changes I'm unable to reproduce that issue. I did a whole
night tests and it still works fine, so it looks that it has been finally solved.
I will post v20 patchset soon :)
Best regards
--
Marek Szyprowski
Samsung Poland R&D Center
next prev parent reply other threads:[~2012-02-03 9:32 UTC|newest]
Thread overview: 61+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-01-26 9:00 [PATCHv19 00/15] Contiguous Memory Allocator Marek Szyprowski
2012-01-26 9:00 ` [PATCH 01/15] mm: page_alloc: remove trailing whitespace Marek Szyprowski
2012-01-30 10:59 ` Mel Gorman
2012-01-26 9:00 ` [PATCH 02/15] mm: page_alloc: update migrate type of pages on pcp when isolating Marek Szyprowski
2012-01-30 11:15 ` Mel Gorman
2012-01-30 15:41 ` Michal Nazarewicz
2012-01-30 16:14 ` Mel Gorman
2012-01-31 16:23 ` Marek Szyprowski
2012-02-02 12:47 ` Mel Gorman
2012-02-02 19:53 ` Michal Nazarewicz
2012-02-03 9:31 ` Marek Szyprowski [this message]
2012-02-03 11:27 ` Mel Gorman
2012-01-26 9:00 ` [PATCH 03/15] mm: compaction: introduce isolate_migratepages_range() Marek Szyprowski
2012-01-30 11:24 ` Mel Gorman
2012-01-30 12:42 ` Michal Nazarewicz
2012-01-30 13:25 ` Mel Gorman
2012-01-26 9:00 ` [PATCH 04/15] mm: compaction: introduce isolate_freepages_range() Marek Szyprowski
2012-01-30 11:48 ` Mel Gorman
2012-01-30 11:55 ` Mel Gorman
2012-01-26 9:00 ` [PATCH 05/15] mm: compaction: export some of the functions Marek Szyprowski
2012-01-30 11:57 ` Mel Gorman
2012-01-30 12:33 ` Michal Nazarewicz
2012-01-26 9:00 ` [PATCH 06/15] mm: page_alloc: introduce alloc_contig_range() Marek Szyprowski
2012-01-30 12:11 ` Mel Gorman
2012-01-26 9:00 ` [PATCH 07/15] mm: page_alloc: change fallbacks array handling Marek Szyprowski
2012-01-30 12:12 ` Mel Gorman
2012-01-26 9:00 ` [PATCH 08/15] mm: mmzone: MIGRATE_CMA migration type added Marek Szyprowski
2012-01-30 12:35 ` Mel Gorman
2012-01-30 13:06 ` Michal Nazarewicz
2012-01-30 14:52 ` Mel Gorman
2012-01-26 9:00 ` [PATCH 09/15] mm: page_isolation: MIGRATE_CMA isolation functions added Marek Szyprowski
2012-01-26 9:00 ` [PATCH 10/15] mm: extract reclaim code from __alloc_pages_direct_reclaim() Marek Szyprowski
2012-01-30 12:42 ` Mel Gorman
2012-01-26 9:00 ` [PATCH 11/15] mm: trigger page reclaim in alloc_contig_range() to stabilize watermarks Marek Szyprowski
2012-01-30 13:05 ` Mel Gorman
2012-01-31 17:15 ` Marek Szyprowski
2012-01-26 9:00 ` [PATCH 12/15] drivers: add Contiguous Memory Allocator Marek Szyprowski
2012-01-27 9:44 ` [Linaro-mm-sig] " Ohad Ben-Cohen
2012-01-27 10:53 ` Marek Szyprowski
2012-01-27 14:27 ` Clark, Rob
2012-01-27 14:51 ` Marek Szyprowski
2012-01-27 14:59 ` Ohad Ben-Cohen
2012-01-27 15:17 ` Marek Szyprowski
2012-01-28 18:57 ` Ohad Ben-Cohen
2012-01-30 7:43 ` Marek Szyprowski
2012-01-30 9:16 ` Ohad Ben-Cohen
2012-01-27 14:56 ` Ohad Ben-Cohen
2012-01-26 9:00 ` [PATCH 13/15] X86: integrate CMA with DMA-mapping subsystem Marek Szyprowski
2012-01-26 9:00 ` [PATCH 14/15] ARM: " Marek Szyprowski
2012-01-26 9:00 ` [PATCH 15/15] ARM: Samsung: use CMA for 2 memory banks for s5p-mfc device Marek Szyprowski
2012-01-26 15:31 ` [PATCHv19 00/15] Contiguous Memory Allocator Arnd Bergmann
2012-01-26 15:38 ` Michal Nazarewicz
2012-01-26 15:48 ` Marek Szyprowski
2012-01-28 0:26 ` Andrew Morton
2012-01-29 18:09 ` Rob Clark
2012-01-29 20:32 ` Anca Emanuel
2012-01-29 20:51 ` Arnd Bergmann
2012-01-30 13:25 ` Mel Gorman
2012-01-30 15:43 ` Michal Nazarewicz
[not found] ` <CA+M3ks7h1t6DbPSAhPN6LJ5Dw84hSukfWG16avh2eZL+o4caJg@mail.gmail.com>
2012-02-01 8:47 ` Marek Szyprowski
2012-02-10 18:10 ` Marek Szyprowski
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='008101cce256$a85e29f0$f91a7dd0$%szyprowski@samsung.com' \
--to=m.szyprowski@samsung.com \
--cc=akpm@linux-foundation.org \
--cc=arnd@arndb.de \
--cc=benjamin.gaignard@linaro.org \
--cc=chunsang.jeong@linaro.org \
--cc=corbet@lwn.net \
--cc=dave@linux.vnet.ibm.com \
--cc=dwalker@codeaurora.org \
--cc=jesse.barker@linaro.org \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=kyungmin.park@samsung.com \
--cc=linaro-mm-sig@lists.linaro.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux@arm.linux.org.uk \
--cc=mel@csn.ul.ie \
--cc=mina86@mina86.com \
--cc=shariq.hasnain@linaro.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).