From: Christoph Hellwig <hch@lst.de>
To: "Darrick J. Wong" <djwong@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>, Jens Axboe <axboe@kernel.dk>,
Carlos Maiolino <cem@kernel.org>,
Damien Le Moal <dlemoal@kernel.org>,
Hans Holmberg <hans.holmberg@wdc.com>,
Keith Busch <kbusch@kernel.org>,
linux-block@vger.kernel.org, linux-xfs@vger.kernel.org,
Carlos Maiolino <cmaiolino@redhat.com>
Subject: Re: [PATCH 3/3] xfs: rework zone GC buffer management
Date: Thu, 15 Jan 2026 07:12:52 +0100 [thread overview]
Message-ID: <20260115061252.GB9205@lst.de> (raw)
In-Reply-To: <20260114195947.GI15551@frogsfrogsfrogs>
On Wed, Jan 14, 2026 at 11:59:47AM -0800, Darrick J. Wong wrote:
> > Signed-off-by: Christoph Hellwig <hch@lst.de>
> > Reviewed-by: Hans Holmberg <hans.holmberg@wdc.com>
> > Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
> > Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
>
> Huh. I guess I never touched this patch at all???
Not yet :)
> Ok so it looks like we're switching out garbage collecting with two
> independent (but not) scratchpads for what amounts to a ring buffer
> consisting of a bunch of discontig folios.
Yeah.
> Now when gc wants to move some data, we compute the number of folios we
> need, and bump scratch_head by that amount. Then we attach those folios
> to a read bio and submit it. When that finishes, we call bio_reuse with
> REQ_OP_WRITE and submit that. When the write finishes we do the "remap
> if mapping hasn't changed" dance to update the metadata, and advance the
> tail by however many folios we've stopped using.
>
> If I got that right, then
You did, thanks!
next prev parent reply other threads:[~2026-01-15 6:13 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-14 13:06 improve zoned XFS GC buffer management v4 Christoph Hellwig
2026-01-14 13:06 ` [PATCH 1/3] block: add a bio_reuse helper Christoph Hellwig
2026-01-14 13:06 ` [PATCH 2/3] xfs: use bio_reuse in the zone GC code Christoph Hellwig
2026-01-14 19:50 ` Darrick J. Wong
2026-01-14 13:06 ` [PATCH 3/3] xfs: rework zone GC buffer management Christoph Hellwig
2026-01-14 19:59 ` Darrick J. Wong
2026-01-15 6:12 ` Christoph Hellwig [this message]
2026-01-25 0:03 ` Chris Mason
2026-01-26 20:30 ` Keith Busch
2026-01-27 6:53 ` Christoph Hellwig
2026-01-21 12:21 ` improve zoned XFS GC buffer management v4 Carlos Maiolino
-- strict thread matches above, loose matches on Subject: below --
2026-01-13 7:19 improve zoned XFS GC buffer management v3 Christoph Hellwig
2026-01-13 7:19 ` [PATCH 3/3] xfs: rework zone GC buffer management Christoph Hellwig
2026-01-06 7:58 improve zoned XFS GC buffer management v2 Christoph Hellwig
2026-01-06 7:58 ` [PATCH 3/3] xfs: rework zone GC buffer management Christoph Hellwig
2026-01-09 12:24 ` Carlos Maiolino
2025-12-18 6:31 improve zoned XFS " Christoph Hellwig
2025-12-18 6:31 ` [PATCH 3/3] xfs: rework zone " Christoph Hellwig
2025-12-18 7:12 ` Damien Le Moal
2025-12-19 8:06 ` Hans Holmberg
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=20260115061252.GB9205@lst.de \
--to=hch@lst.de \
--cc=axboe@kernel.dk \
--cc=cem@kernel.org \
--cc=cmaiolino@redhat.com \
--cc=djwong@kernel.org \
--cc=dlemoal@kernel.org \
--cc=hans.holmberg@wdc.com \
--cc=kbusch@kernel.org \
--cc=linux-block@vger.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.