linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Mel Gorman <mgorman@techsingularity.net>
To: Johannes Weiner <hannes@cmpxchg.org>
Cc: linux-mm@kvack.org, Kaiyang Zhao <kaiyang2@cs.cmu.edu>,
	Vlastimil Babka <vbabka@suse.cz>,
	David Rientjes <rientjes@google.com>,
	linux-kernel@vger.kernel.org, kernel-team@fb.com
Subject: Re: [RFC PATCH 25/26] mm: page_alloc: disallow fallbacks when 2M defrag is enabled
Date: Fri, 21 Apr 2023 16:55:23 +0100	[thread overview]
Message-ID: <20230421155523.p4qqx23xpwlm3yt4@techsingularity.net> (raw)
In-Reply-To: <20230421152448.GD320347@cmpxchg.org>

On Fri, Apr 21, 2023 at 11:24:48AM -0400, Johannes Weiner wrote:
> On Fri, Apr 21, 2023 at 03:56:57PM +0100, Mel Gorman wrote:
> > On Tue, Apr 18, 2023 at 03:13:12PM -0400, Johannes Weiner wrote:
> > > Fallbacks are already unlikely due to watermarks being enforced
> > > against MIGRATE_FREE blocks. Eliminate them altogether. This allows
> > > compaction to look exclusively at movable blocks, reducing the number
> > > of pageblocks it needs to scan on an ongoing basis.
> > > 
> > > Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
> > 
> > Conceptually this could be fun if a GFP_NOFS allocation cannot migrate
> > enough memory to free one pageblock and there are no pageblocks
> > available of the correct migratetype. Fallbacks might be unlikely but
> > never being able to fallback is a livelock risk, no?
> 
> The reserves below the watermarks are maintained in neutral
> MIGRATE_FREE blocks. So just like today, critical/limited allocation
> contexts are ensured forward progress as long as there are reserves.
> 
> An argument could be made that because smaller orders type-claim the
> entire neutral block on allocation now, the reserves can deplete
> faster than they do today given comparable watermarks. I haven't run
> into this issue during testing that would have me raise the reserves
> by a factor of NR_MIGRATETYPES. But something to keep an eye out for.

Also bear in mind low memory systems, not even the embedded ones, just
things like a cheap laptop with <= 1G. I haven't consulted the code to
think this through but forcing watermarks to be in multiples of pageblock
could be get trickier the smaller the memory size is and it would have
to be considered what happens when min_free_kbytes for a zone is smaller
than pageblock*MIGRATE_TYPES. It's possible it's protected, or could be
protected, by page_group_by_mobility_disabled but that only applies for
very small systems.

-- 
Mel Gorman
SUSE Labs


  reply	other threads:[~2023-04-21 15:55 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-04-18 19:12 [RFC PATCH 00/26] mm: reliable huge page allocator Johannes Weiner
2023-04-18 19:12 ` [RFC PATCH 01/26] block: bdev: blockdev page cache is movable Johannes Weiner
2023-04-19  4:07   ` Matthew Wilcox
2023-04-21 12:25   ` Mel Gorman
2023-04-18 19:12 ` [RFC PATCH 02/26] mm: compaction: avoid GFP_NOFS deadlocks Johannes Weiner
2023-04-21 12:27   ` Mel Gorman
2023-04-21 14:17     ` Johannes Weiner
2023-04-18 19:12 ` [RFC PATCH 03/26] mm: make pageblock_order 2M per default Johannes Weiner
2023-04-19  0:01   ` Kirill A. Shutemov
2023-04-19  2:55     ` Johannes Weiner
2023-04-19  3:44       ` Johannes Weiner
2023-04-19 11:10     ` David Hildenbrand
2023-04-19 10:36   ` Vlastimil Babka
2023-04-19 11:09     ` David Hildenbrand
2023-04-21 12:37   ` Mel Gorman
2023-04-18 19:12 ` [RFC PATCH 04/26] mm: page_isolation: write proper kerneldoc Johannes Weiner
2023-04-21 12:39   ` Mel Gorman
2023-04-18 19:12 ` [RFC PATCH 05/26] mm: page_alloc: per-migratetype pcplist for THPs Johannes Weiner
2023-04-21 12:47   ` Mel Gorman
2023-04-21 15:06     ` Johannes Weiner
2023-04-28 10:29       ` Mel Gorman
2023-04-18 19:12 ` [RFC PATCH 06/26] mm: page_alloc: consolidate free page accounting Johannes Weiner
2023-04-21 12:54   ` Mel Gorman
2023-04-21 15:08     ` Johannes Weiner
2023-04-18 19:12 ` [RFC PATCH 07/26] mm: page_alloc: move capture_control to the page allocator Johannes Weiner
2023-04-21 12:59   ` Mel Gorman
2023-04-18 19:12 ` [RFC PATCH 08/26] mm: page_alloc: claim blocks during compaction capturing Johannes Weiner
2023-04-21 13:12   ` Mel Gorman
2023-04-25 14:39     ` Johannes Weiner
2023-04-18 19:12 ` [RFC PATCH 09/26] mm: page_alloc: move expand() above compaction_capture() Johannes Weiner
2023-04-18 19:12 ` [RFC PATCH 10/26] mm: page_alloc: allow compaction capturing from larger blocks Johannes Weiner
2023-04-21 14:14   ` Mel Gorman
2023-04-25 15:40     ` Johannes Weiner
2023-04-28 10:41       ` Mel Gorman
2023-04-18 19:12 ` [RFC PATCH 11/26] mm: page_alloc: introduce MIGRATE_FREE Johannes Weiner
2023-04-21 14:25   ` Mel Gorman
2023-04-18 19:12 ` [RFC PATCH 12/26] mm: page_alloc: per-migratetype free counts Johannes Weiner
2023-04-21 14:28   ` Mel Gorman
2023-04-21 15:35     ` Johannes Weiner
2023-04-21 16:03       ` Mel Gorman
2023-04-21 16:32         ` Johannes Weiner
2023-04-18 19:13 ` [RFC PATCH 13/26] mm: compaction: remove compaction result helpers Johannes Weiner
2023-04-21 14:32   ` Mel Gorman
2023-04-18 19:13 ` [RFC PATCH 14/26] mm: compaction: simplify should_compact_retry() Johannes Weiner
2023-04-21 14:36   ` Mel Gorman
2023-04-25  2:15     ` Johannes Weiner
2023-04-25  0:56   ` Huang, Ying
2023-04-25  2:11     ` Johannes Weiner
2023-04-18 19:13 ` [RFC PATCH 15/26] mm: compaction: simplify free block check in suitable_migration_target() Johannes Weiner
2023-04-21 14:39   ` Mel Gorman
2023-04-18 19:13 ` [RFC PATCH 16/26] mm: compaction: improve compaction_suitable() accuracy Johannes Weiner
2023-04-18 19:13 ` [RFC PATCH 17/26] mm: compaction: refactor __compaction_suitable() Johannes Weiner
2023-04-18 19:13 ` [RFC PATCH 18/26] mm: compaction: remove unnecessary is_via_compact_memory() checks Johannes Weiner
2023-04-18 19:13 ` [RFC PATCH 19/26] mm: compaction: drop redundant watermark check in compaction_zonelist_suitable() Johannes Weiner
2023-04-18 19:13 ` [RFC PATCH 20/26] mm: vmscan: use compaction_suitable() check in kswapd Johannes Weiner
2023-04-25  3:12   ` Huang, Ying
2023-04-25 14:26     ` Johannes Weiner
2023-04-26  1:30       ` Huang, Ying
2023-04-26 15:22         ` Johannes Weiner
2023-04-27  5:41           ` Huang, Ying
2023-04-18 19:13 ` [RFC PATCH 21/26] mm: compaction: align compaction goals with reclaim goals Johannes Weiner
2023-04-18 19:13 ` [RFC PATCH 22/26] mm: page_alloc: manage free memory in whole pageblocks Johannes Weiner
2023-04-18 19:13 ` [RFC PATCH 23/26] mm: page_alloc: kill highatomic Johannes Weiner
2023-04-18 19:13 ` [RFC PATCH 24/26] mm: page_alloc: kill watermark boosting Johannes Weiner
2023-04-18 19:13 ` [RFC PATCH 25/26] mm: page_alloc: disallow fallbacks when 2M defrag is enabled Johannes Weiner
2023-04-21 14:56   ` Mel Gorman
2023-04-21 15:24     ` Johannes Weiner
2023-04-21 15:55       ` Mel Gorman [this message]
2023-04-18 19:13 ` [RFC PATCH 26/26] mm: page_alloc: add sanity checks for migratetypes Johannes Weiner
2023-04-18 23:54 ` [RFC PATCH 00/26] mm: reliable huge page allocator Kirill A. Shutemov
2023-04-19  2:08   ` Johannes Weiner
2023-04-19 10:56     ` Vlastimil Babka
2023-04-19  4:11 ` Matthew Wilcox
2023-04-21 16:11   ` Mel Gorman
2023-04-21 17:14     ` Matthew Wilcox
2023-05-02 15:21       ` David Hildenbrand

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=20230421155523.p4qqx23xpwlm3yt4@techsingularity.net \
    --to=mgorman@techsingularity.net \
    --cc=hannes@cmpxchg.org \
    --cc=kaiyang2@cs.cmu.edu \
    --cc=kernel-team@fb.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=rientjes@google.com \
    --cc=vbabka@suse.cz \
    /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