From: Christoph Hellwig <hch@infradead.org>
To: "Ritesh Harjani (IBM)" <ritesh.list@gmail.com>
Cc: linux-ext4@vger.kernel.org, linux-xfs@vger.kernel.org,
linux-fsdevel@vger.kernel.org,
Matthew Wilcox <willy@infradead.org>,
"Darrick J . Wong" <djwong@kernel.org>,
Ojaswin Mujoo <ojaswin@linux.ibm.com>, Jan Kara <jack@suse.cz>
Subject: Re: [RFCv3 7/7] iomap: Optimize data access patterns for filesystems with indirect mappings
Date: Fri, 26 Apr 2024 21:47:03 -0700 [thread overview]
Message-ID: <ZiyDRw5zuF-jcBhZ@infradead.org> (raw)
In-Reply-To: <4e2752e99f55469c4eb5f2fe83e816d529110192.1714046808.git.ritesh.list@gmail.com>
On Thu, Apr 25, 2024 at 06:58:51PM +0530, Ritesh Harjani (IBM) wrote:
> Currently the bios for reads within iomap are only submitted at
> 2 places -
> 1. If we cannot merge the new req. with previous bio, only then we
> submit the previous bio.
> 2. Submit the bio at the end of the entire read processing.
>
> This means for filesystems with indirect block mapping, we call into
> ->iomap_begin() again w/o submitting the previous bios. That causes
> unoptimized data access patterns for blocks which are of BH_Boundary type.
The same is true for extent mappings. And it's not ideal there either,
although it only inreases the bio submission latency.
prev parent reply other threads:[~2024-04-27 4:47 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-04-25 13:28 [RFCv3 0/7] ext2 iomap changes and iomap improvements Ritesh Harjani (IBM)
2024-04-25 13:28 ` [RFCv3 1/7] ext2: Remove comment related to journal handle Ritesh Harjani (IBM)
2024-04-26 15:21 ` Darrick J. Wong
2024-04-25 13:28 ` [RFCv3 2/7] ext2: Convert ext2 regular file buffered I/O to use iomap Ritesh Harjani (IBM)
2024-04-26 15:29 ` Darrick J. Wong
2024-04-25 13:28 ` [RFCv3 3/7] ext2: Enable large folio support Ritesh Harjani (IBM)
2024-04-26 15:30 ` Darrick J. Wong
2024-04-25 13:28 ` [RFCv3 4/7] ext2: Implement seq counter for validating cached iomap Ritesh Harjani (IBM)
2024-04-26 15:41 ` Darrick J. Wong
2024-05-12 13:20 ` Jan Kara
2024-04-25 13:28 ` [RFCv3 5/7] iomap: Fix iomap_adjust_read_range for plen calculation Ritesh Harjani (IBM)
2024-04-26 6:49 ` Christoph Hellwig
2024-04-26 8:52 ` Ritesh Harjani
2024-04-26 15:43 ` Darrick J. Wong
2024-04-25 13:28 ` [RFCv3 6/7] iomap: Optimize iomap_read_folio Ritesh Harjani (IBM)
2024-04-25 13:53 ` Matthew Wilcox
2024-04-26 6:53 ` Christoph Hellwig
2024-04-26 8:50 ` Ritesh Harjani
2024-04-27 4:44 ` Christoph Hellwig
2024-04-25 13:28 ` [RFCv3 7/7] iomap: Optimize data access patterns for filesystems with indirect mappings Ritesh Harjani (IBM)
2024-04-26 16:24 ` Darrick J. Wong
2024-04-26 17:17 ` Ritesh Harjani
2024-04-26 17:25 ` Ritesh Harjani
2024-04-26 17:37 ` Matthew Wilcox
2024-04-26 17:55 ` Ritesh Harjani
2024-04-26 18:13 ` Matthew Wilcox
2024-04-26 18:57 ` Ritesh Harjani
2024-04-26 19:19 ` Matthew Wilcox
2024-04-27 4:54 ` Christoph Hellwig
2024-04-27 6:03 ` Ritesh Harjani
2024-04-27 4:47 ` Christoph Hellwig [this message]
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=ZiyDRw5zuF-jcBhZ@infradead.org \
--to=hch@infradead.org \
--cc=djwong@kernel.org \
--cc=jack@suse.cz \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-xfs@vger.kernel.org \
--cc=ojaswin@linux.ibm.com \
--cc=ritesh.list@gmail.com \
--cc=willy@infradead.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;
as well as URLs for NNTP newsgroup(s).