From: Heming Zhao <heming.zhao@suse.com>
To: joseph.qi@linux.alibaba.com, mark@fasheh.com, jlbec@evilplan.org,
hch@lst.de
Cc: Heming Zhao <heming.zhao@suse.com>,
ocfs2-devel@lists.linux.dev, linux-kernel@vger.kernel.org
Subject: [RFC PATCH v9 0/3] migrate ocfs2 DIO from buffer_head to iomap
Date: Fri, 20 Nov 2026 14:00:10 +0800 [thread overview]
Message-ID: <20261120060014.12087-1-heming.zhao@suse.com> (raw)
This patch set migrates ocfs2 Direct I/O (DIO) from buffer_head mode
to iomap mode. It is based on commit aebb72744764 (v7.4-rc3), and
also incorporates Joseph's code and ideas from [1]:
V9 changelog:
Address sashiko's review comments [3] (all changes are in patch [3/3]):
ocfs2_file_write_to_iter():
- switch pr lock to ex when DIO falls back to buffered IO
- Fall back to buffered I/O for a direct write when block size is smaller
than the page size and the write does not cover whole pages
- clear orphan when rw_lock is still held
ocfs2_dio_wr_map_blocks may use a value beyand u32 to call
ocfs2_clusters_to_blocks()
initialize oi->ip_dio_orphan_pid and oi->ip_dio_orphan_end in
ocfs2_inode_init_once()
--------------
# test report
This patch set has been tested on 7.4-rc3 kernel
## ltp
runltp -f dio -d /mnt/ocfs2/
results: passed
## xfstests
./check -g quick -T -b -s ocfs2 -e generic/081 \
-e generic/266 -e generic/272 -e generic/281 \
-e generic/338 -e generic/347 -e generic/361 \
-e generic/628 -e generic/629 -e generic/648 -e generic/650
OCFS2_MKFS_LOCAL_OPT="-b 2K -C 4K --fs-features=local --fs-feature-level=max-features"
Failures: generic/003 generic/056 generic/091 generic/094 generic/177
generic/211 generic/225 generic/263 generic/321 generic/322 generic/329
generic/331 generic/370 generic/376 generic/424 generic/448 generic/467
generic/477 generic/481 generic/498 generic/510 generic/512 generic/513
generic/534 generic/537 generic/552 generic/557 generic/563 generic/578
generic/588 generic/607 generic/620 generic/630 generic/755 generic/760
generic/771 generic/777 generic/779 generic/784
Failed 39 of 632 tests
OCFS2_MKFS_LOCAL_OPT="-b 4K -C 4K --fs-features=local --fs-feature-level=max-features"
Failures: generic/003 generic/056 generic/070 generic/091 generic/094
generic/177 generic/211 generic/225 generic/263 generic/321 generic/322
generic/329 generic/331 generic/370 generic/376 generic/424 generic/448
generic/467 generic/477 generic/481 generic/498 generic/510 generic/512
generic/513 generic/534 generic/537 generic/552 generic/557 generic/563
generic/578 generic/588 generic/607 generic/620 generic/630 generic/755
generic/760 generic/771 generic/777 generic/779 generic/784
Failed 40 of 632 tests
OCFS2_MKFS_LOCAL_OPT="-b 2K -C 8K --fs-features=local --fs-feature-level=max-features"
Failures: generic/003 generic/056 generic/075 generic/091 generic/094
generic/112 generic/177 generic/211 generic/225 generic/263 generic/321
generic/322 generic/329 generic/331 generic/370 generic/376 generic/424
generic/448 generic/467 generic/477 generic/481 generic/498 generic/510
generic/512 generic/513 generic/534 generic/537 generic/552 generic/557
generic/563 generic/564 generic/578 generic/588 generic/607 generic/620
generic/630 generic/755 generic/759 generic/760 generic/771 generic/777
generic/779 generic/784 ocfs2/001
Failed 44 of 632 tests
## fio
(4k blocksize 4k clustersize, local mount)
fio --direct=1 --iodepth=32 --rw=write --ioengine=libaio --bs=128k --size=1g \
--numjobs=4 --runtime=100 --group_reporting --filename=/mnt/ocfs2/write.1g --name=tst
before patch: write: IOPS=32.2k, BW=4028MiB/s
after patch: write: IOPS=38.6k, BW=4824MiB/s
fio --direct=1 --iodepth=32 --rw=read --ioengine=libaio --bs=128k --size=1g \
--numjobs=4 --runtime=100 --group_reporting --filename=/mnt/ocfs2/write.1g --name=tst
before patch: read: IOPS=49.1k, BW=6132MiB/s
after patch: read: IOPS=50.9k, BW=6360MiB/s
fio --direct=1 --iodepth=32 --rw=randwrite --ioengine=libaio --bs=4k \
--size=1g --numjobs=4 --runtime=100 --group_reporting --filename=/mnt/ocfs2/write.1g --name=tst
before patch: write: IOPS=14.4k, BW=56.4MiB/s
after patch: write: IOPS=14.4k, BW=56.1MiB/s
fio --direct=1 --iodepth=32 --rw=randwrite --ioengine=libaio --bs=4k \
--size=1g --numjobs=4 --runtime=100 --group_reporting --filename=/mnt/ocfs2/write.1g --name=tst
before patch: read: IOPS=382k, BW=1492MiB/s
after patch: read: IOPS=419k, BW=1637MiB/s
------
[1]:
https://gitee.com/anolis/storage-cloud-kernel/commit/ea3fe75e2060118ba093e90b1cd34fa382914b29
[2]:
https://sashiko.dev/#/patchset/20261103135605.22943-1-heming.zhao%40suse.com
[3]:
v1 patch set:
https://lore.kernel.org/ocfs2-devel/20260724052559.15409-1-heming.zhao@suse.com/
v2 patch set:
https://lore.kernel.org/ocfs2-devel/20260727061802.18485-1-heming.zhao@suse.com/
v3 patch set:
https://lore.kernel.org/ocfs2-devel/20261018143319.22329-1-heming.zhao@suse.com/
v3 sashiko review:
https://sashiko.dev/#/patchset/20261018143319.22329-1-heming.zhao%40suse.com
v4 patch set:
https://lore.kernel.org/ocfs2-devel/20261103135605.22943-1-heming.zhao@suse.com/
v4 sashiko review:
https://sashiko.dev/#/patchset/20261103135605.22943-1-heming.zhao%40suse.com
v5 patch set:
https://lore.kernel.org/ocfs2-devel/20261105071930.16399-1-heming.zhao@suse.com/
v5 sashiko review:
https://sashiko.dev/#/patchset/20261105071930.16399-1-heming.zhao%40suse.com
v6 patch set:
https://lore.kernel.org/ocfs2-devel/20261115154013.9379-1-heming.zhao@suse.com/
v6 sashiko review:
https://sashiko.dev/#/patchset/20261115154013.9379-1-heming.zhao%40suse.com
v7 patch set:
https://lore.kernel.org/ocfs2-devel/20261118002050.9019-1-heming.zhao@suse.com/
v7 sashiko review:
- https://sashiko.dev/#/patchset/20261117150053.18417-1-heming.zhao%40suse.com
- https://sashiko.dev/#/patchset/20261118002050.9019-1-heming.zhao%40suse.com
v8 patch set:
https://lore.kernel.org/ocfs2-devel/20261119064029.11910-1-heming.zhao@suse.com/
v8 sashiko review:
https://sashiko.dev/#/patchset/20261119064029.11910-1-heming.zhao%40suse.com
-----------------------
Heming Zhao (3):
ocfs2: Add new ocfs2_map_blocks() to introduce iomap feature
ocfs2: switch DIO read path from buffer_head to iomap
ocfs2: switch dio write path from buffer_head to iomap
fs/ocfs2/Kconfig | 2 +-
fs/ocfs2/aops.c | 1685 ++++++++++++++++++++++++++++++++---------
fs/ocfs2/aops.h | 43 +-
fs/ocfs2/extent_map.c | 34 +-
fs/ocfs2/extent_map.h | 3 +
fs/ocfs2/file.c | 314 +++++++-
fs/ocfs2/inode.h | 9 +
fs/ocfs2/mmap.c | 2 +-
fs/ocfs2/ocfs2.h | 25 +-
fs/ocfs2/super.c | 3 +
10 files changed, 1690 insertions(+), 430 deletions(-)
--
2.55.0
next reply other threads:[~2026-11-20 6:00 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-20 6:00 Heming Zhao [this message]
2026-11-20 6:00 ` [RFC PATCH v9 1/3] ocfs2: Add new ocfs2_map_blocks() to introduce iomap feature Heming Zhao
2026-11-20 6:00 ` [RFC PATCH v9 2/3] ocfs2: switch DIO read path from buffer_head to iomap Heming Zhao
2026-11-20 6:00 ` [RFC PATCH v9 3/3] ocfs2: switch dio write " Heming Zhao
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=20261120060014.12087-1-heming.zhao@suse.com \
--to=heming.zhao@suse.com \
--cc=hch@lst.de \
--cc=jlbec@evilplan.org \
--cc=joseph.qi@linux.alibaba.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mark@fasheh.com \
--cc=ocfs2-devel@lists.linux.dev \
/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
all inboxes | Powered by JetHome®