From: Christoph Hellwig <hch@lst.de>
To: Dave Chinner <david@fromorbit.com>
Cc: Christoph Hellwig <hch@lst.de>, Carlos Maiolino <cem@kernel.org>,
"Darrick J. Wong" <djwong@kernel.org>,
Dave Chinner <dchinner@redhat.com>,
linux-xfs@vger.kernel.org
Subject: Re: [PATCH 07/12] xfs: convert buffer cache to use high order folios
Date: Mon, 10 Mar 2025 14:18:55 +0100 [thread overview]
Message-ID: <20250310131854.GB8546@lst.de> (raw)
In-Reply-To: <Z8i5HSS1ppbtQiJc@dread.disaster.area>
On Thu, Mar 06, 2025 at 07:50:37AM +1100, Dave Chinner wrote:
> The only thing extra that I would do here is take a leaf from the
> kmalloc() call in xlog_kvmalloc() and turn off direct reclaim for
> this allocation because >= 32kB allocations are considered "costly"
> and so will enter the compaction code if direct reclaim is enabled.
>
> Given that we fall back to vmalloc, clearing __GFP_DIRECT_RECLAIM
> and setting __GFP_NORETRY here means that we don't burn lots of CPU
> on memory compaction if there is no high order folios available for
> immediate allocation. And on a busy machine, compaction is likely to
> fail frequently and so this is all wasted CPU time.
>
> This may be one of the reasons why you don't see any change in real
> performance with 64kB directory blocks - we spend more time in
> folio allocation because of compaction overhead than we gain back
> from avoiding the use of vmapped buffers....
FYI, this did not make any difference in my testing.
next prev parent reply other threads:[~2025-03-10 13:19 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-05 14:05 use folios and vmalloc for buffer cache backing memory v2 Christoph Hellwig
2025-03-05 14:05 ` [PATCH 01/12] xfs: unmapped buffer item size straddling mismatch Christoph Hellwig
2025-03-05 14:05 ` [PATCH 02/12] xfs: add a fast path to xfs_buf_zero when b_addr is set Christoph Hellwig
2025-03-05 14:05 ` [PATCH 03/12] xfs: remove xfs_buf.b_offset Christoph Hellwig
2025-03-05 14:05 ` [PATCH 04/12] xfs: remove xfs_buf_is_vmapped Christoph Hellwig
2025-03-05 14:05 ` [PATCH 05/12] xfs: refactor backing memory allocations for buffers Christoph Hellwig
2025-03-05 14:05 ` [PATCH 06/12] xfs: remove the kmalloc to page allocator fallback Christoph Hellwig
2025-03-05 18:18 ` Darrick J. Wong
2025-03-05 23:32 ` Christoph Hellwig
2025-03-05 21:02 ` Dave Chinner
2025-03-05 23:38 ` Christoph Hellwig
2025-03-05 14:05 ` [PATCH 07/12] xfs: convert buffer cache to use high order folios Christoph Hellwig
2025-03-05 18:20 ` Darrick J. Wong
2025-03-05 20:50 ` Dave Chinner
2025-03-05 23:33 ` Christoph Hellwig
2025-03-10 13:18 ` Christoph Hellwig [this message]
2025-03-05 14:05 ` [PATCH 08/12] xfs: kill XBF_UNMAPPED Christoph Hellwig
2025-03-05 14:05 ` [PATCH 09/12] xfs: buffer items don't straddle pages anymore Christoph Hellwig
2025-03-05 14:05 ` [PATCH 10/12] xfs: use vmalloc instead of vm_map_area for buffer backing memory Christoph Hellwig
2025-03-05 18:22 ` Darrick J. Wong
2025-03-05 21:20 ` Dave Chinner
2025-03-05 22:54 ` Darrick J. Wong
2025-03-05 23:28 ` Dave Chinner
2025-03-05 23:45 ` Christoph Hellwig
2025-03-05 23:35 ` Christoph Hellwig
2025-03-06 0:57 ` Dave Chinner
2025-03-06 1:40 ` Christoph Hellwig
2025-03-05 14:05 ` [PATCH 11/12] xfs: cleanup mapping tmpfs folios into the buffer cache Christoph Hellwig
2025-03-05 18:34 ` Darrick J. Wong
2025-03-05 14:05 ` [PATCH 12/12] xfs: trace what memory backs a buffer Christoph Hellwig
-- strict thread matches above, loose matches on Subject: below --
2025-03-10 13:19 use folios and vmalloc for buffer cache backing memory v3 Christoph Hellwig
2025-03-10 13:19 ` [PATCH 07/12] xfs: convert buffer cache to use high order folios Christoph Hellwig
2025-02-26 15:51 use folios and vmalloc for buffer cache backing memory Christoph Hellwig
2025-02-26 15:51 ` [PATCH 07/12] xfs: convert buffer cache to use high order folios Christoph Hellwig
2025-02-26 17:33 ` Darrick J. Wong
2025-03-04 14:06 ` Christoph Hellwig
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=20250310131854.GB8546@lst.de \
--to=hch@lst.de \
--cc=cem@kernel.org \
--cc=david@fromorbit.com \
--cc=dchinner@redhat.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