From: Balbir Singh <bsingharora@gmail.com>
To: Mel Gorman <mgorman@techsingularity.net>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Linux-MM <linux-mm@kvack.org>, Rik van Riel <riel@surriel.com>,
Vlastimil Babka <vbabka@suse.cz>,
Johannes Weiner <hannes@cmpxchg.org>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 00/27] Move LRU page reclaim from zones to nodes v7
Date: Mon, 27 Jun 2016 22:48:03 +1000 [thread overview]
Message-ID: <57712083.8060500@gmail.com> (raw)
In-Reply-To: <20160624075059.GC1868@techsingularity.net>
On 24/06/16 17:50, Mel Gorman wrote:
> On Fri, Jun 24, 2016 at 04:35:45PM +1000, Balbir Singh wrote:
>>> 1. The residency of a page partially depends on what zone the page was
>>> allocated from. This is partially combatted by the fair zone allocation
>>> policy but that is a partial solution that introduces overhead in the
>>> page allocator paths.
>>>
>>> 2. Currently, reclaim on node 0 behaves slightly different to node 1. For
>>> example, direct reclaim scans in zonelist order and reclaims even if
>>> the zone is over the high watermark regardless of the age of pages
>>> in that LRU. Kswapd on the other hand starts reclaim on the highest
>>> unbalanced zone. A difference in distribution of file/anon pages due
>>> to when they were allocated results can result in a difference in
>>> again. While the fair zone allocation policy mitigates some of the
>>> problems here, the page reclaim results on a multi-zone node will
>>> always be different to a single-zone node.
>>> it was scheduled on as a result.
>>>
>>> 3. kswapd and the page allocator scan zones in the opposite order to
>>> avoid interfering with each other but it's sensitive to timing. This
>>> mitigates the page allocator using pages that were allocated very recently
>>> in the ideal case but it's sensitive to timing. When kswapd is allocating
>>> from lower zones then it's great but during the rebalancing of the highest
>>> zone, the page allocator and kswapd interfere with each other. It's worse
>>> if the highest zone is small and difficult to balance.
>>>
>>> 4. slab shrinkers are node-based which makes it harder to identify the exact
>>> relationship between slab reclaim and LRU reclaim.
>>>
>>
>> Sorry, I am late in reading the thread and the patches, but I am trying to understand
>> the key benefits?
>
> The key benefits were outlined at the beginning of the changelog. The
> one that is missing is the large overhead from the fair zone allocation
> policy which can be removed safely by the feature. The benefit to page
> allocator micro-benchmarks is outlined in the series introduction.
I did look at them, but between 1 to 4, it seemed like the largest benefit
was mm cleanup and better behaviour of reclaim on node 0.
>
>> I know that
>> zones have grown to be overloaded to mean many things now. What is the contention impact
>> of moving the LRU from zone to nodes?
>
> Expected to be minimal. On NUMA machines, most nodes have only one zone.
> On machines with multiple zones, the lock per zone is not that fine-grained
> given the size of the zones on large memory configurations.
>
Makes sense
Thanks,
Balbir Singh.
next prev parent reply other threads:[~2016-06-27 12:48 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-21 14:15 [PATCH 00/27] Move LRU page reclaim from zones to nodes v7 Mel Gorman
2016-06-21 14:15 ` [PATCH 01/27] mm, vmstat: Add infrastructure for per-node vmstats Mel Gorman
2016-06-21 14:15 ` [PATCH 02/27] mm, vmscan: Move lru_lock to the node Mel Gorman
2016-06-21 14:15 ` [PATCH 03/27] mm, vmscan: Move LRU lists to node Mel Gorman
2016-06-22 12:50 ` Vlastimil Babka
2016-06-21 14:15 ` [PATCH 04/27] mm, vmscan: Begin reclaiming pages on a per-node basis Mel Gorman
2016-06-22 14:04 ` Vlastimil Babka
2016-06-22 16:00 ` Vlastimil Babka
2016-06-23 11:07 ` Mel Gorman
2016-06-23 11:13 ` Michal Hocko
2016-06-23 10:58 ` Mel Gorman
2016-06-21 14:15 ` [PATCH 05/27] mm, vmscan: Have kswapd only scan based on the highest requested zone Mel Gorman
2016-06-21 14:15 ` [PATCH 06/27] mm, vmscan: Make kswapd reclaim in terms of nodes Mel Gorman
2016-06-21 14:15 ` [PATCH 07/27] mm, vmscan: Remove balance gap Mel Gorman
2016-06-21 14:15 ` [PATCH 08/27] mm, vmscan: Simplify the logic deciding whether kswapd sleeps Mel Gorman
2016-06-22 15:30 ` Vlastimil Babka
2016-06-21 14:15 ` [PATCH 09/27] mm, vmscan: By default have direct reclaim only shrink once per node Mel Gorman
2016-06-21 14:15 ` [PATCH 10/27] mm, vmscan: Remove duplicate logic clearing node congestion and dirty state Mel Gorman
2016-06-21 14:15 ` [PATCH 11/27] mm: vmscan: Do not reclaim from kswapd if there is any eligible zone Mel Gorman
2016-06-21 14:15 ` [PATCH 12/27] mm, vmscan: Make shrink_node decisions more node-centric Mel Gorman
2016-06-22 13:20 ` Michal Hocko
2016-06-22 15:42 ` Vlastimil Babka
2016-06-21 14:15 ` [PATCH 13/27] mm, memcg: Move memcg limit enforcement from zones to nodes Mel Gorman
2016-06-22 13:15 ` Michal Hocko
2016-06-21 14:15 ` [PATCH 14/27] mm, workingset: Make working set detection node-aware Mel Gorman
2016-06-21 14:15 ` [PATCH 15/27] mm, page_alloc: Consider dirtyable memory in terms of nodes Mel Gorman
2016-06-22 14:15 ` Michal Hocko
2016-06-22 14:27 ` Michal Hocko
2016-06-23 12:53 ` Mel Gorman
2016-06-23 13:13 ` Michal Hocko
2016-06-21 14:15 ` [PATCH 16/27] mm: Move page mapped accounting to the node Mel Gorman
2016-06-21 22:32 ` Andrew Morton
2016-06-23 8:35 ` Mel Gorman
2016-06-22 14:35 ` Michal Hocko
2016-06-21 14:15 ` [PATCH 17/27] mm: Rename NR_ANON_PAGES to NR_ANON_MAPPED Mel Gorman
2016-06-22 14:28 ` Michal Hocko
2016-06-21 14:15 ` [PATCH 18/27] mm: Move most file-based accounting to the node Mel Gorman
2016-06-22 14:38 ` Michal Hocko
2016-06-21 14:15 ` [PATCH 19/27] mm: Move vmscan writes and file write " Mel Gorman
2016-06-22 14:40 ` Michal Hocko
2016-06-23 13:57 ` Mel Gorman
2016-06-23 14:06 ` Vlastimil Babka
2016-06-23 16:03 ` Mel Gorman
2016-06-21 14:15 ` [PATCH 20/27] mm, vmscan: Update classzone_idx if buffer_heads_over_limit Mel Gorman
2016-06-22 14:49 ` Michal Hocko
2016-06-21 14:16 ` [PATCH 21/27] mm, vmscan: Only wakeup kswapd once per node for the requested classzone Mel Gorman
2016-06-22 16:08 ` Vlastimil Babka
2016-06-21 14:16 ` [PATCH 22/27] mm: Convert zone_reclaim to node_reclaim Mel Gorman
2016-06-21 14:16 ` [PATCH 23/27] mm, vmscan: Add classzone information to tracepoints Mel Gorman
2016-06-21 14:16 ` [PATCH 24/27] mm, page_alloc: Remove fair zone allocation policy Mel Gorman
2016-06-21 14:16 ` [PATCH 25/27] mm: page_alloc: Cache the last node whose dirty limit is reached Mel Gorman
2016-06-21 14:16 ` [PATCH 26/27] mm: vmstat: Replace __count_zone_vm_events with a zone id equivalent Mel Gorman
2016-06-21 14:16 ` [PATCH 27/27] mm: vmstat: Account per-zone stalls and pages skipped during reclaim Mel Gorman
2016-06-23 10:26 ` [PATCH 00/27] Move LRU page reclaim from zones to nodes v7 Mel Gorman
2016-06-23 11:27 ` Michal Hocko
2016-06-23 12:33 ` Mel Gorman
2016-06-23 12:44 ` Michal Hocko
2016-06-23 21:45 ` Andrew Morton
2016-06-24 6:35 ` Balbir Singh
2016-06-24 7:50 ` Mel Gorman
2016-06-27 12:48 ` Balbir Singh [this message]
-- strict thread matches above, loose matches on Subject: below --
2016-06-21 11:43 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=57712083.8060500@gmail.com \
--to=bsingharora@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@techsingularity.net \
--cc=riel@surriel.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;
as well as URLs for NNTP newsgroup(s).