From: Zhang Yi <yi.zhang@huaweicloud.com>
To: Christoph Hellwig <hch@lst.de>, Christian Brauner <brauner@kernel.org>
Cc: "Darrick J. Wong" <djwong@kernel.org>,
Chandan Babu R <chandan.babu@oracle.com>,
Ritesh Harjani <ritesh.list@gmail.com>,
linux-xfs@vger.kernel.org, linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH 09/13] iomap: don't chain bios
Date: Mon, 27 Nov 2023 20:53:22 +0800 [thread overview]
Message-ID: <0f136350-3242-3e20-3b8a-56a39c66b001@huaweicloud.com> (raw)
In-Reply-To: <20231126124720.1249310-10-hch@lst.de>
Hi, Christoph.
On 2023/11/26 20:47, Christoph Hellwig wrote:
> Back in the days when a single bio could only be filled to the hardware
> limits, and we scheduled a work item for each bio completion, chaining
> multiple bios for a single ioend made a lot of sense to reduce the number
> of completions. But these days bios can be filled until we reach the
> number of vectors or total size limit, which means we can always fit at
> least 1 megabyte worth of data in the worst case, but usually a lot more
> due to large folios. The only thing bio chaining is buying us now is
> to reduce the size of the allocation from an ioend with an embedded bio
> into a plain bio, which is a 52 bytes differences on 64-bit systems.
>
> This is not worth the added complexity, so remove the bio chaining and
> only use the bio embedded into the ioend. This will help to simplify
> further changes to the iomap writeback code.
>
> Signed-off-by: Christoph Hellwig <hch@lst.de>
A nice cleanup! I'm just a little curious about the writeback performance
impact of this patch. Do you have any actual test data on xfs?
Thanks,
Yi.
next prev parent reply other threads:[~2023-11-27 12:53 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-11-26 12:47 RFC: map multiple blocks per ->map_blocks in iomap writeback Christoph Hellwig
2023-11-26 12:47 ` [PATCH 01/13] iomap: clear the per-folio dirty bits on all writeback failures Christoph Hellwig
2023-11-27 3:47 ` Ritesh Harjani
2023-11-27 6:29 ` Christoph Hellwig
2023-11-26 12:47 ` [PATCH 02/13] iomap: treat inline data in iomap_writepage_map as an I/O error Christoph Hellwig
2023-11-27 5:01 ` Ritesh Harjani
2023-11-27 6:33 ` Christoph Hellwig
2023-11-29 4:40 ` Darrick J. Wong
2023-11-29 5:39 ` Christoph Hellwig
2023-11-26 12:47 ` [PATCH 03/13] iomap: move the io_folios field out of struct iomap_ioend Christoph Hellwig
2023-11-27 5:33 ` Ritesh Harjani
2023-11-29 4:41 ` Darrick J. Wong
2023-11-26 12:47 ` [PATCH 04/13] iomap: drop the obsolete PF_MEMALLOC check in iomap_do_writepage Christoph Hellwig
2023-11-27 6:39 ` Ritesh Harjani
2023-11-27 6:41 ` Christoph Hellwig
2023-11-29 4:45 ` Darrick J. Wong
2023-11-26 12:47 ` [PATCH 05/13] iomap: factor out a iomap_writepage_handle_eof helper Christoph Hellwig
2023-11-27 6:57 ` Ritesh Harjani
2023-11-27 7:02 ` Ritesh Harjani
2023-11-27 7:12 ` Christoph Hellwig
2023-11-29 4:48 ` Darrick J. Wong
2023-11-26 12:47 ` [PATCH 06/13] iomap: move all remaining per-folio logic into xfs_writepage_map Christoph Hellwig
2023-11-27 7:36 ` Ritesh Harjani
2023-11-27 19:20 ` Josef Bacik
2023-11-29 4:50 ` Darrick J. Wong
2023-11-26 12:47 ` [PATCH 07/13] iomap: clean up the iomap_new_ioend calling convention Christoph Hellwig
2023-11-27 7:43 ` Ritesh Harjani
2023-11-27 8:13 ` Christoph Hellwig
2023-11-29 4:51 ` Darrick J. Wong
2023-11-26 12:47 ` [PATCH 08/13] iomap: move the iomap_sector sector calculation out of iomap_add_to_ioend Christoph Hellwig
2023-11-27 9:54 ` Ritesh Harjani
2023-11-27 13:54 ` Christoph Hellwig
2023-11-29 4:53 ` Darrick J. Wong
2023-11-29 5:42 ` Christoph Hellwig
2023-11-26 12:47 ` [PATCH 09/13] iomap: don't chain bios Christoph Hellwig
2023-11-27 12:53 ` Zhang Yi [this message]
2023-11-27 13:51 ` Christoph Hellwig
2023-11-29 4:59 ` Darrick J. Wong
2023-11-29 5:40 ` Christoph Hellwig
2023-11-26 12:47 ` [PATCH 10/13] iomap: only call mapping_set_error once for each failed bio Christoph Hellwig
2023-11-29 5:03 ` Darrick J. Wong
2023-11-29 5:41 ` Christoph Hellwig
2023-11-26 12:47 ` [PATCH 11/13] iomap: factor out a iomap_writepage_map_block helper Christoph Hellwig
2023-11-29 5:06 ` Darrick J. Wong
2023-11-26 12:47 ` [PATCH 12/13] iomap: submit ioends immediately Christoph Hellwig
2023-11-29 5:14 ` Darrick J. Wong
2023-11-26 12:47 ` [PATCH 13/13] iomap: map multiple blocks at a time Christoph Hellwig
2023-11-29 5:25 ` Darrick J. Wong
2023-11-29 5:44 ` 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=0f136350-3242-3e20-3b8a-56a39c66b001@huaweicloud.com \
--to=yi.zhang@huaweicloud.com \
--cc=brauner@kernel.org \
--cc=chandan.babu@oracle.com \
--cc=djwong@kernel.org \
--cc=hch@lst.de \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-xfs@vger.kernel.org \
--cc=ritesh.list@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).