From: Vlastimil Babka <vbabka@suse.cz>
To: Mel Gorman <mgorman@suse.com>, Linux-MM <linux-mm@kvack.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>,
Rik van Riel <riel@redhat.com>, Pintu Kumar <pintu.k@samsung.com>,
Xishi Qiu <qiuxishi@huawei.com>, Gioh Kim <gioh.kim@lge.com>,
LKML <linux-kernel@vger.kernel.org>,
Mel Gorman <mgorman@techsingularity.net>
Subject: Re: [PATCH 08/10] mm, page_alloc: Remove MIGRATE_RESERVE
Date: Wed, 29 Jul 2015 11:59:15 +0200 [thread overview]
Message-ID: <55B8A3F3.6090107@suse.cz> (raw)
In-Reply-To: <1437379219-9160-9-git-send-email-mgorman@suse.com>
On 07/20/2015 10:00 AM, Mel Gorman wrote:
> From: Mel Gorman <mgorman@suse.de>
>
> MIGRATE_RESERVE preserves an old property of the buddy allocator that existed
> prior to fragmentation avoidance -- min_free_kbytes worth of pages tended to
> remain free until the only alternative was to fail the allocation. At the
^ I think you meant contiguous instead of free? Is it because
splitting chooses lowest possible order, and grouping by mobility means you
might be splitting e.g. order-5 movable page instead of using order-0 unmovable
page? And that the fallback heuristics specifically select highest available
order? I think it's not that obvious, so worth mentioning.
> time it was discovered that high-order atomic allocations relied on this
> property so MIGRATE_RESERVE was introduced. A later patch will introduce
> an alternative MIGRATE_HIGHATOMIC so this patch deletes MIGRATE_RESERVE
> and supporting code so it'll be easier to review. Note that this patch
> in isolation may look like a false regression if someone was bisecting
> high-order atomic allocation failures.
>
> Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
next prev parent reply other threads:[~2015-07-29 9:59 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-20 8:00 [RFC PATCH 00/10] Remove zonelist cache and high-order watermark checking Mel Gorman
2015-07-20 8:00 ` [PATCH 01/10] mm, page_alloc: Delete the zonelist_cache Mel Gorman
2015-07-21 23:47 ` David Rientjes
2015-07-23 10:58 ` Mel Gorman
2015-07-20 8:00 ` [PATCH 02/10] mm, page_alloc: Remove unnecessary parameter from zone_watermark_ok_safe Mel Gorman
2015-07-21 23:49 ` David Rientjes
2015-07-28 12:20 ` Vlastimil Babka
2015-07-20 8:00 ` [PATCH 03/10] mm, page_alloc: Remove unnecessary recalculations for dirty zone balancing Mel Gorman
2015-07-22 0:08 ` David Rientjes
2015-07-23 12:28 ` Mel Gorman
2015-07-28 12:25 ` Vlastimil Babka
2015-07-20 8:00 ` [PATCH 04/10] mm, page_alloc: Remove unnecessary taking of a seqlock when cpusets are disabled Mel Gorman
2015-07-22 0:11 ` David Rientjes
2015-07-28 12:32 ` Vlastimil Babka
2015-07-20 8:00 ` [PATCH 05/10] mm, page_alloc: Remove unnecessary updating of GFP flags during normal operation Mel Gorman
2015-07-28 13:36 ` Vlastimil Babka
2015-07-28 13:47 ` Peter Zijlstra
2015-07-28 15:48 ` Mel Gorman
2015-07-20 8:00 ` [PATCH 06/10] mm, page_alloc: Use jump label to check if page grouping by mobility is enabled Mel Gorman
2015-07-28 13:42 ` Vlastimil Babka
2015-07-20 8:00 ` [PATCH 07/10] mm, page_alloc: Use masks and shifts when converting GFP flags to migrate types Mel Gorman
2015-07-20 8:00 ` [PATCH 08/10] mm, page_alloc: Remove MIGRATE_RESERVE Mel Gorman
2015-07-29 9:59 ` Vlastimil Babka [this message]
2015-07-29 12:25 ` Mel Gorman
2015-07-20 8:00 ` [PATCH 09/10] mm, page_alloc: Reserve pageblocks for high-order atomic allocations on demand Mel Gorman
2015-07-29 11:35 ` Vlastimil Babka
2015-07-29 12:53 ` Mel Gorman
2015-07-31 8:28 ` Vlastimil Babka
2015-07-31 8:43 ` Mel Gorman
2015-07-31 5:54 ` Joonsoo Kim
2015-07-31 7:11 ` Mel Gorman
2015-07-31 7:25 ` Vlastimil Babka
2015-07-31 8:22 ` Mel Gorman
2015-07-31 8:30 ` Joonsoo Kim
2015-07-31 8:26 ` Joonsoo Kim
2015-07-31 8:41 ` Mel Gorman
2015-07-20 8:00 ` [PATCH 10/10] mm, page_alloc: Only enforce watermarks for order-0 allocations Mel Gorman
2015-07-29 12:25 ` Vlastimil Babka
2015-07-29 13:04 ` Mel Gorman
2015-07-31 6:08 ` Joonsoo Kim
2015-07-31 7:19 ` Mel Gorman
2015-07-31 8:40 ` Joonsoo Kim
2015-07-31 6:14 ` [RFC PATCH 00/10] Remove zonelist cache and high-order watermark checking Joonsoo Kim
2015-07-31 7:20 ` Mel Gorman
2015-08-12 10:45 [PATCH 00/10] Remove zonelist cache and high-order watermark checking v2 Mel Gorman
2015-08-12 10:45 ` [PATCH 08/10] mm, page_alloc: Remove MIGRATE_RESERVE Mel Gorman
2015-09-21 10:52 [PATCH 00/10] Remove zonelist cache and high-order watermark checking v4 Mel Gorman
2015-09-21 10:52 ` [PATCH 08/10] mm, page_alloc: Remove MIGRATE_RESERVE Mel Gorman
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=55B8A3F3.6090107@suse.cz \
--to=vbabka@suse.cz \
--cc=gioh.kim@lge.com \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.com \
--cc=mgorman@techsingularity.net \
--cc=pintu.k@samsung.com \
--cc=qiuxishi@huawei.com \
--cc=riel@redhat.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).