From: Stefan Hajnoczi <stefanha@redhat.com>
To: qemu-devel@nongnu.org
Cc: Aarushi Mehta <mehta.aaru20@gmail.com>,
Stefan Hajnoczi <stefanha@redhat.com>,
Stefano Garzarella <sgarzare@redhat.com>,
surajshirvankar@gmail.com, Hanna Reitz <hreitz@redhat.com>,
qemu-block@nongnu.org, Kevin Wolf <kwolf@redhat.com>,
Paolo Bonzini <pbonzini@redhat.com>, Fam Zheng <fam@euphon.net>
Subject: [PATCH 0/3] [RESEND] block: unify block and fdmon io_uring
Date: Tue, 1 Apr 2025 10:27:18 -0400 [thread overview]
Message-ID: <20250401142721.280287-1-stefanha@redhat.com> (raw)
For some reason the patch series didn't reach the mailing list. Resending.
Every AioContext has an io_uring context for file descriptor monitoring. In
addition, block/io_uring.c also has an io_uring context for disk I/O. This
patch series eliminates the extra io_uring context so that AioContext has a
single io_uring context.
Patch 2 introduces the aio_add_sqe() API for submitting custom io_uring SQEs.
Patch 3 then replaces its io_uring context with the aio_add_sqe() API so that
disk I/O requests share the AioContext's io_uring context.
I will post performance results when they become available.
Based-on: 20250326183340.1010531-1-stefanha@redhat.com
Stefan Hajnoczi (3):
aio-posix: treat io_uring setup failure as fatal
aio-posix: add aio_add_sqe() API for user-defined io_uring requests
block/io_uring: use aio_add_sqe()
include/block/aio.h | 74 +++++-
include/block/raw-aio.h | 5 -
util/aio-posix.h | 1 +
block/file-posix.c | 25 +-
block/io_uring.c | 489 ++++++++++------------------------------
stubs/io_uring.c | 32 ---
util/aio-posix.c | 9 +
util/async.c | 35 ---
util/fdmon-io_uring.c | 155 +++++++++----
block/trace-events | 12 +-
stubs/meson.build | 3 -
util/trace-events | 4 +
12 files changed, 316 insertions(+), 528 deletions(-)
delete mode 100644 stubs/io_uring.c
--
2.49.0
next reply other threads:[~2025-04-01 14:32 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-01 14:27 Stefan Hajnoczi [this message]
2025-04-01 14:27 ` [PATCH 1/3] aio-posix: treat io_uring setup failure as fatal Stefan Hajnoczi
2025-04-25 15:51 ` Kevin Wolf
2025-04-28 18:35 ` Stefan Hajnoczi
2025-04-01 14:27 ` [PATCH 2/3] aio-posix: add aio_add_sqe() API for user-defined io_uring requests Stefan Hajnoczi
2025-04-01 14:27 ` [PATCH 3/3] block/io_uring: use aio_add_sqe() Stefan Hajnoczi
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=20250401142721.280287-1-stefanha@redhat.com \
--to=stefanha@redhat.com \
--cc=fam@euphon.net \
--cc=hreitz@redhat.com \
--cc=kwolf@redhat.com \
--cc=mehta.aaru20@gmail.com \
--cc=pbonzini@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=sgarzare@redhat.com \
--cc=surajshirvankar@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).