From: Vlastimil Babka <vbabka@suse.cz>
To: Yang Shi <shy828301@gmail.com>,
guro@fb.com, ktkhai@virtuozzo.com, shakeelb@google.com,
david@fromorbit.com, hannes@cmpxchg.org, mhocko@suse.com,
akpm@linux-foundation.org
Cc: linux-mm@kvack.org, linux-fsdevel@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [v7 PATCH 12/12] mm: vmscan: shrink deferred objects proportional to priority
Date: Thu, 11 Feb 2021 14:10:49 +0100 [thread overview]
Message-ID: <acd1915c-306b-08a8-9e0f-b06c1e09fb4c@suse.cz> (raw)
In-Reply-To: <20210209174646.1310591-13-shy828301@gmail.com>
On 2/9/21 6:46 PM, Yang Shi wrote:
> The number of deferred objects might get windup to an absurd number, and it
> results in clamp of slab objects. It is undesirable for sustaining workingset.
>
> So shrink deferred objects proportional to priority and cap nr_deferred to twice
> of cache items.
Makes sense to me, minimally it's simpler than the old code and avoiding absurd
growth of nr_deferred should be a good thing, as well as the "proportional to
priority" part.
I just suspect there's a bit of unnecessary bias in the implementation, as
explained below:
> The idea is borrowed from Dave Chinner's patch:
> https://lore.kernel.org/linux-xfs/20191031234618.15403-13-david@fromorbit.com/
>
> Tested with kernel build and vfs metadata heavy workload in our production
> environment, no regression is spotted so far.
>
> Signed-off-by: Yang Shi <shy828301@gmail.com>
> ---
> mm/vmscan.c | 40 +++++-----------------------------------
> 1 file changed, 5 insertions(+), 35 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 66163082cc6f..d670b119d6bd 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -654,7 +654,6 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
> */
> nr = count_nr_deferred(shrinker, shrinkctl);
>
> - total_scan = nr;
> if (shrinker->seeks) {
> delta = freeable >> priority;
> delta *= 4;
> @@ -668,37 +667,9 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
> delta = freeable / 2;
> }
>
> + total_scan = nr >> priority;
> total_scan += delta;
So, our scan goal consists of the part based on freeable objects (delta), plus a
part of the defferred objects (nr >> priority). Fine.
> - if (total_scan < 0) {
> - pr_err("shrink_slab: %pS negative objects to delete nr=%ld\n",
> - shrinker->scan_objects, total_scan);
> - total_scan = freeable;
> - next_deferred = nr;
> - } else
> - next_deferred = total_scan;
> -
> - /*
> - * We need to avoid excessive windup on filesystem shrinkers
> - * due to large numbers of GFP_NOFS allocations causing the
> - * shrinkers to return -1 all the time. This results in a large
> - * nr being built up so when a shrink that can do some work
> - * comes along it empties the entire cache due to nr >>>
> - * freeable. This is bad for sustaining a working set in
> - * memory.
> - *
> - * Hence only allow the shrinker to scan the entire cache when
> - * a large delta change is calculated directly.
> - */
> - if (delta < freeable / 4)
> - total_scan = min(total_scan, freeable / 2);
> -
> - /*
> - * Avoid risking looping forever due to too large nr value:
> - * never try to free more than twice the estimate number of
> - * freeable entries.
> - */
> - if (total_scan > freeable * 2)
> - total_scan = freeable * 2;
> + total_scan = min(total_scan, (2 * freeable));
Probably unnecessary as we cap next_deferred below anyway? So total_scan cannot
grow without limits anymore. But can't hurt.
> trace_mm_shrink_slab_start(shrinker, shrinkctl, nr,
> freeable, delta, total_scan, priority);
> @@ -737,10 +708,9 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
> cond_resched();
> }
>
> - if (next_deferred >= scanned)
> - next_deferred -= scanned;
> - else
> - next_deferred = 0;
> + next_deferred = max_t(long, (nr - scanned), 0) + total_scan;
And here's the bias I think. Suppose we scanned 0 due to e.g. GFP_NOFS. We count
as newly deferred both the "delta" part of total_scan, which is fine, but also
the "nr >> priority" part, where we failed to our share of the "reduce
nr_deferred" work, but I don't think it means we should also increase
nr_deferred by that amount of failed work.
OTOH if we succeed and scan exactly the whole goal, we are subtracting from
nr_deferred both the "nr >> priority" part, which is correct, but also delta,
which was new work, not deferred one, so that's incorrect IMHO as well.
So the calculation should probably be something like this?
next_deferred = max_t(long, nr + delta - scanned, 0);
Thanks,
Vlastimil
> + next_deferred = min(next_deferred, (2 * freeable));
> +
> /*
> * move the unused scan count back into the shrinker in a
> * manner that handles concurrent updates.
>
next prev parent reply other threads:[~2021-02-11 13:36 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-09 17:46 [v7 PATCH 0/12] Make shrinker's nr_deferred memcg aware Yang Shi
2021-02-09 17:46 ` [v7 PATCH 01/12] mm: vmscan: use nid from shrink_control for tracepoint Yang Shi
2021-02-09 19:14 ` Shakeel Butt
2021-02-10 16:58 ` Yang Shi
2021-02-09 19:21 ` Roman Gushchin
2021-02-09 17:46 ` [v7 PATCH 02/12] mm: vmscan: consolidate shrinker_maps handling code Yang Shi
2021-02-09 20:27 ` Roman Gushchin
2021-02-10 14:19 ` Shakeel Butt
2021-02-09 17:46 ` [v7 PATCH 03/12] mm: vmscan: use shrinker_rwsem to protect shrinker_maps allocation Yang Shi
2021-02-09 20:33 ` Roman Gushchin
2021-02-09 23:28 ` Yang Shi
2021-02-09 17:46 ` [v7 PATCH 04/12] mm: vmscan: remove memcg_shrinker_map_size Yang Shi
2021-02-09 20:43 ` Roman Gushchin
2021-02-09 23:31 ` Yang Shi
2021-02-10 18:14 ` Vlastimil Babka
2021-02-09 17:46 ` [v7 PATCH 05/12] mm: memcontrol: rename shrinker_map to shrinker_info Yang Shi
2021-02-09 20:50 ` Roman Gushchin
2021-02-09 23:33 ` Yang Shi
2021-02-10 0:16 ` Roman Gushchin
2021-02-11 16:47 ` Kirill Tkhai
2021-02-11 17:29 ` Yang Shi
2021-02-09 17:46 ` [v7 PATCH 06/12] mm: vmscan: add shrinker_info_protected() helper Yang Shi
2021-02-10 0:22 ` Roman Gushchin
2021-02-10 1:07 ` Yang Shi
2021-02-10 1:29 ` Roman Gushchin
2021-02-10 12:12 ` Kirill Tkhai
2021-02-10 18:17 ` Vlastimil Babka
2021-02-12 6:54 ` [mm] bd741fb2ad: WARNING:suspicious_RCU_usage kernel test robot
2021-02-09 17:46 ` [v7 PATCH 07/12] mm: vmscan: use a new flag to indicate shrinker is registered Yang Shi
2021-02-10 0:39 ` Roman Gushchin
2021-02-10 1:12 ` Yang Shi
2021-02-10 1:34 ` Roman Gushchin
2021-02-10 1:55 ` Yang Shi
2021-02-10 18:45 ` Yang Shi
2021-02-10 18:23 ` Vlastimil Babka
2021-02-09 17:46 ` [v7 PATCH 08/12] mm: vmscan: add per memcg shrinker nr_deferred Yang Shi
2021-02-10 1:10 ` Roman Gushchin
2021-02-10 1:25 ` Yang Shi
2021-02-10 1:40 ` Roman Gushchin
2021-02-10 1:57 ` Yang Shi
2021-02-09 17:46 ` [v7 PATCH 09/12] mm: vmscan: use per memcg nr_deferred of shrinker Yang Shi
2021-02-10 1:27 ` Roman Gushchin
2021-02-10 1:52 ` Yang Shi
2021-02-10 14:36 ` Kirill Tkhai
2021-02-10 16:41 ` Yang Shi
2021-02-09 17:46 ` [v7 PATCH 10/12] mm: vmscan: don't need allocate shrinker->nr_deferred for memcg aware shrinkers Yang Shi
2021-02-10 1:23 ` Roman Gushchin
2021-02-09 17:46 ` [v7 PATCH 11/12] mm: memcontrol: reparent nr_deferred when memcg offline Yang Shi
2021-02-10 1:18 ` Roman Gushchin
2021-02-09 17:46 ` [v7 PATCH 12/12] mm: vmscan: shrink deferred objects proportional to priority Yang Shi
2021-02-11 13:10 ` Vlastimil Babka [this message]
2021-02-11 17:29 ` Yang Shi
2021-02-11 18:52 ` Vlastimil Babka
2021-02-11 19:15 ` Yang Shi
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=acd1915c-306b-08a8-9e0f-b06c1e09fb4c@suse.cz \
--to=vbabka@suse.cz \
--cc=akpm@linux-foundation.org \
--cc=david@fromorbit.com \
--cc=guro@fb.com \
--cc=hannes@cmpxchg.org \
--cc=ktkhai@virtuozzo.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=shakeelb@google.com \
--cc=shy828301@gmail.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).