From: Minchan Kim <minchan@kernel.org>
To: qiuxishi <qiuxishi@gmail.com>
Cc: mgorman@suse.de, akpm@linux-foundation.org,
kamezawa.hiroyu@jp.fujitsu.com, isimatu.yasuaki@jp.fujitsu.com,
qiuxishi@huawei.com, linux-mm@kvack.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 3/3] memory-hotplug: bug fix race between isolation and allocation
Date: Thu, 6 Sep 2012 11:59:12 +0900 [thread overview]
Message-ID: <20120906025912.GE31615@bbox> (raw)
In-Reply-To: <50480BFB.8050501@gmail.com>
Hello Xishi,
On Thu, Sep 06, 2012 at 10:35:39AM +0800, qiuxishi wrote:
> On 2012/9/5 17:40, Mel Gorman wrote:
>
> > On Wed, Sep 05, 2012 at 04:26:02PM +0900, Minchan Kim wrote:
> >> Like below, memory-hotplug makes race between page-isolation
> >> and page-allocation so it can hit BUG_ON in __offline_isolated_pages.
> >>
> >> CPU A CPU B
> >>
> >> start_isolate_page_range
> >> set_migratetype_isolate
> >> spin_lock_irqsave(zone->lock)
> >>
> >> free_hot_cold_page(Page A)
> >> /* without zone->lock */
> >> migratetype = get_pageblock_migratetype(Page A);
> >> /*
> >> * Page could be moved into MIGRATE_MOVABLE
> >> * of per_cpu_pages
> >> */
> >> list_add_tail(&page->lru, &pcp->lists[migratetype]);
> >>
> >> set_pageblock_isolate
> >> move_freepages_block
> >> drain_all_pages
>
> I think here is the problem you want to fix, it is not sure that pcp will be moved
> into MIGRATE_ISOLATE list. They may be moved into MIGRATE_MOVABLE list because
> page_private() maybe 2, it uses page_private() not get_pageblock_migratetype()
>
> So when finish migrating pages, the free pages from pcp may be allocated again, and
> failed in check_pages_isolated().
>
> drain_all_pages()
> drain_local_pages()
> drain_pages()
> free_pcppages_bulk()
> __free_one_page(page, zone, 0, page_private(page))
>
> I reported this problem too. http://marc.info/?l=linux-mm&m=134555113706068&w=2
> How about this change:
> free_pcppages_bulk()
> __free_one_page(page, zone, 0, get_pageblock_migratetype(page))
I already explained why it was not good solution.
Again, here it goes from my previous reply.
"
Anyway, I don't like your approach which I already considered because it hurts hotpath
while the race is really unlikely. Get_pageblock_migratetype is never trivial.
We should avoid the overhead in hotpath and move into memory-hotplug itself.
Do you see my patch in https://patchwork.kernel.org/patch/1225081/ ?
"
--
Kind regards,
Minchan Kim
next prev parent reply other threads:[~2012-09-06 2:57 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-09-06 2:35 [PATCH 3/3] memory-hotplug: bug fix race between isolation and allocation qiuxishi
2012-09-06 2:59 ` Minchan Kim [this message]
-- strict thread matches above, loose matches on Subject: below --
2012-09-05 7:25 [PATCH 0/3] memory-hotplug: handle page race between allocation and isolation Minchan Kim
2012-09-05 7:26 ` [PATCH 3/3] memory-hotplug: bug fix race between isolation and allocation Minchan Kim
2012-09-05 9:40 ` Mel Gorman
2012-09-06 4:49 ` Minchan Kim
2012-09-06 9:24 ` Mel Gorman
2012-09-06 23:32 ` Minchan Kim
2012-09-07 6:26 ` jencce zhou
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=20120906025912.GE31615@bbox \
--to=minchan@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=isimatu.yasuaki@jp.fujitsu.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=qiuxishi@gmail.com \
--cc=qiuxishi@huawei.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 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).