From: Minchan Kim <minchan@kernel.org>
To: Mel Gorman <mgorman@suse.de>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>,
Xishi Qiu <qiuxishi@huawei.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/3] mm: remain migratetype in freed page
Date: Thu, 6 Sep 2012 11:28:50 +0900 [thread overview]
Message-ID: <20120906022850.GD31615@bbox> (raw)
In-Reply-To: <20120905092534.GE11266@suse.de>
On Wed, Sep 05, 2012 at 10:25:34AM +0100, Mel Gorman wrote:
> On Wed, Sep 05, 2012 at 04:26:01PM +0900, Minchan Kim wrote:
> > Page allocator doesn't keep migratetype information to page
> > when the page is freed. This patch remains the information
> > to freed page's index field which isn't used by free/alloc
> > preparing so it shouldn't change any behavir except below one.
> >
>
> This explanation could have been a *LOT* more helpful.
>
> The page allocator caches the pageblock information in page->private while
> it is in the PCP freelists but this is overwritten with the order of the
> page when freed to the buddy allocator. This patch stores the migratetype
> of the page in the page->index field so that it is available at all times.
I will add your comment in my description.
>
> > This patch adds a new call site in __free_pages_ok so it might be
> > overhead a bit but it's for high order allocation.
> > So I believe damage isn't hurt.
> >
>
> The additional call to set_page_migratetype() is not heavy. If you were
> adding a new call to get_pageblock_migratetype() or something equally
> expensive I would be more concerned.
I'm lucky to avoid your keen eye. ;)
>
> > Signed-off-by: Minchan Kim <minchan@kernel.org>
>
> The information you store in the page->index becomes stale if the page
> gets moved to another free list by move_freepages(). Not sure if that is
> a problem for you or not but it is possible that
> get_page_migratetype(page) != get_pageblock_migratetype(page)
Thanks for the spot. I have to fix it.
>
> > ---
> > include/linux/mm.h | 6 ++++--
> > mm/page_alloc.c | 7 ++++---
> > 2 files changed, 8 insertions(+), 5 deletions(-)
> >
> > diff --git a/include/linux/mm.h b/include/linux/mm.h
> > index 86d61d6..8fd32da 100644
> > --- a/include/linux/mm.h
> > +++ b/include/linux/mm.h
> > @@ -251,12 +251,14 @@ struct inode;
> >
> > static inline void set_page_migratetype(struct page *page, int migratetype)
> > {
> > - set_page_private(page, migratetype);
> > + VM_BUG_ON((unsigned int)migratetype >= MIGRATE_TYPES);
>
> This additional bug check is not mentioned in the changelog. Not clear
> if it's necessary.
I'm not strong so if anyone think it's not necessary, I will drop.
>
> > + page->index = migratetype;
> > }
> >
> > static inline int get_page_migratetype(struct page *page)
> > {
> > - return page_private(page);
> > + VM_BUG_ON((unsigned int)page->index >= MIGRATE_TYPES);
> > + return page->index;
> > }
> >
> > /*
> > diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> > index 103ba66..32985dd 100644
> > --- a/mm/page_alloc.c
> > +++ b/mm/page_alloc.c
> > @@ -723,6 +723,7 @@ static void __free_pages_ok(struct page *page, unsigned int order)
> > {
> > unsigned long flags;
> > int wasMlocked = __TestClearPageMlocked(page);
> > + int migratetype;
> >
> > if (!free_pages_prepare(page, order))
> > return;
> > @@ -731,9 +732,9 @@ static void __free_pages_ok(struct page *page, unsigned int order)
> > if (unlikely(wasMlocked))
> > free_page_mlock(page);
> > __count_vm_events(PGFREE, 1 << order);
> > - free_one_page(page_zone(page), page, order,
> > - get_pageblock_migratetype(page));
> > -
> > + migratetype = get_pageblock_migratetype(page);
> > + set_page_migratetype(page, migratetype);
> > + free_one_page(page_zone(page), page, order, migratetype);
> > local_irq_restore(flags);
> > }
> >
>
> --
> Mel Gorman
> SUSE Labs
>
> --
> 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>
--
Kind regards,
Minchan Kim
next prev parent reply other threads:[~2012-09-06 2:27 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
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 1/3] mm: use get_page_migratetype instead of page_private Minchan Kim
2012-09-05 9:09 ` Mel Gorman
2012-09-06 2:17 ` Minchan Kim
2012-09-06 2:02 ` Kamezawa Hiroyuki
2012-09-06 2:19 ` Minchan Kim
2012-09-05 7:26 ` [PATCH 2/3] mm: remain migratetype in freed page Minchan Kim
2012-09-05 9:25 ` Mel Gorman
2012-09-06 2:28 ` Minchan Kim [this message]
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=20120906022850.GD31615@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@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).