From: Dave Chinner <david@fromorbit.com>
To: "Darrick J. Wong" <djwong@kernel.org>
Cc: linux-xfs@vger.kernel.org
Subject: Re: [PATCH 10/11] xfs: parallelize inode inactivation
Date: Wed, 24 Mar 2021 09:21:52 +1100 [thread overview]
Message-ID: <20210323222152.GH63242@dread.disaster.area> (raw)
In-Reply-To: <161543199635.1947934.2885924822578773349.stgit@magnolia>
On Wed, Mar 10, 2021 at 07:06:36PM -0800, Darrick J. Wong wrote:
> From: Darrick J. Wong <djwong@kernel.org>
>
> Split the inode inactivation work into per-AG work items so that we can
> take advantage of parallelization.
How does this scale out when we have thousands of AGs?
I'm guessing that the gc_workqueue has the default "unbound"
parallelism that means it will run up to 4 kworkers per CPU at a
time? Which means we could have hundreds of ags trying to hammer on
inactivations at the same time? And so bash hard on the log and
completely starve the syscall front end of log space?
It seems to me that this needs to bound the amount of concurrent
work to quite low numbers - even though it is per-ag, we do not want
this to swamp the system in kworkers blocked on log reservations
when such concurrency it not necessary.
Cheers,
Dave.
--
Dave Chinner
david@fromorbit.com
next prev parent reply other threads:[~2021-03-23 22:24 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-11 3:05 [PATCHSET v3 00/11] xfs: deferred inode inactivation Darrick J. Wong
2021-03-11 3:05 ` [PATCH 01/11] xfs: prevent metadata files from being inactivated Darrick J. Wong
2021-03-11 13:05 ` Christoph Hellwig
2021-03-22 23:13 ` Dave Chinner
2021-03-11 3:05 ` [PATCH 02/11] xfs: refactor the predicate part of xfs_free_eofblocks Darrick J. Wong
2021-03-11 13:09 ` Christoph Hellwig
2021-03-15 18:46 ` Christoph Hellwig
2021-03-18 4:33 ` Darrick J. Wong
2021-03-19 1:48 ` Darrick J. Wong
2021-03-11 3:05 ` [PATCH 03/11] xfs: don't reclaim dquots with incore reservations Darrick J. Wong
2021-03-15 18:29 ` Christoph Hellwig
2021-03-22 23:31 ` Dave Chinner
2021-03-23 0:01 ` Darrick J. Wong
2021-03-23 1:48 ` Dave Chinner
2021-03-11 3:06 ` [PATCH 04/11] xfs: decide if inode needs inactivation Darrick J. Wong
2021-03-15 18:47 ` Christoph Hellwig
2021-03-15 19:06 ` Darrick J. Wong
2021-03-11 3:06 ` [PATCH 05/11] xfs: rename the blockgc workqueue Darrick J. Wong
2021-03-15 18:49 ` Christoph Hellwig
2021-03-11 3:06 ` [PATCH 06/11] xfs: deferred inode inactivation Darrick J. Wong
2021-03-16 7:27 ` Christoph Hellwig
2021-03-16 15:47 ` Darrick J. Wong
2021-03-17 15:21 ` Christoph Hellwig
2021-03-17 15:49 ` Darrick J. Wong
2021-03-22 23:46 ` Dave Chinner
2021-03-22 23:37 ` Dave Chinner
2021-03-23 0:24 ` Darrick J. Wong
2021-03-23 1:44 ` Dave Chinner
2021-03-23 4:00 ` Darrick J. Wong
2021-03-23 5:19 ` Dave Chinner
2021-03-24 2:04 ` Darrick J. Wong
2021-03-24 4:57 ` Dave Chinner
2021-03-25 4:20 ` Darrick J. Wong
2021-03-24 17:53 ` Christoph Hellwig
2021-03-25 4:26 ` Darrick J. Wong
2021-03-11 3:06 ` [PATCH 07/11] xfs: expose sysfs knob to control inode inactivation delay Darrick J. Wong
2021-03-11 3:06 ` [PATCH 08/11] xfs: force inode inactivation and retry fs writes when there isn't space Darrick J. Wong
2021-03-15 18:54 ` Christoph Hellwig
2021-03-15 19:06 ` Darrick J. Wong
2021-03-11 3:06 ` [PATCH 09/11] xfs: force inode garbage collection before fallocate when space is low Darrick J. Wong
2021-03-11 3:06 ` [PATCH 10/11] xfs: parallelize inode inactivation Darrick J. Wong
2021-03-15 18:55 ` Christoph Hellwig
2021-03-15 19:03 ` Darrick J. Wong
2021-03-23 22:21 ` Dave Chinner [this message]
2021-03-24 3:52 ` Darrick J. Wong
2021-03-11 3:06 ` [PATCH 11/11] xfs: create a polled function to force " Darrick J. Wong
2021-03-23 22:31 ` Dave Chinner
2021-03-24 3:34 ` Darrick J. Wong
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=20210323222152.GH63242@dread.disaster.area \
--to=david@fromorbit.com \
--cc=djwong@kernel.org \
--cc=linux-xfs@vger.kernel.org \
/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