From: Stefan Hajnoczi <stefanha@redhat.com>
To: qemu-devel@nongnu.org
Cc: Kevin Wolf <kwolf@redhat.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Anthony Liguori <aliguori@us.ibm.com>,
pingfank@linux.vnet.ibm.com,
Stefan Hajnoczi <stefanha@redhat.com>
Subject: [Qemu-devel] [PATCH v2 00/17] aio: drop io_flush()
Date: Wed, 17 Apr 2013 15:07:36 +0200 [thread overview]
Message-ID: <1366204073-8468-1-git-send-email-stefanha@redhat.com> (raw)
This series gets rid of aio's .io_flush() callback. It's based on Paolo's
insight that bdrv_drain_all() can be implemented using the bs->tracked_requests
list. io_flush() is redundant since the block layer already knows if requests
are pending.
The point of this effort is to simplify our event loop(s). If we can reduce
custom features like io_flush() it becomes easier to unify event loops and
reuse glib or other options.
This is also important to me for dataplane, since bdrv_drain_all() is one of
the synchronization points between threads. QEMU monitor commands invoke
bdrv_drain_all() while the block device is accessed from a dataplane thread.
Background on io_flush() semantics:
The io_flush() handler must return 1 if this aio fd handler is active. That
is, requests are pending and we'd like to make progress by monitoring the fd.
If io_flush() returns 0, the aio event loop skips monitoring this fd. This is
critical for block drivers like iscsi where we have an idle TCP socket which we
want to block on *only* when there are pending requests.
The series works as follows:
Part 1 - stop relying on .io_flush()
The first patches change aio_poll() callers to check their termination
condition themselves instead of relying on .io_flush():
1 block: stop relying on io_flush() in bdrv_drain_all()
2 dataplane/virtio-blk: check exit conditions before aio_poll()
3 tests: adjust test-aio to new aio_poll() semantics
4 tests: adjust test-thread-pool to new aio_poll() semantics
Part 2 - stop calling .io_flush() from aio_poll()
The semantic change to aio_poll() is made here:
5 aio: stop using .io_flush()
Part 3 - drop io_flush() handlers, just pass NULL for the io_flush argument
Remove the now dead code from all .io_flush() users:
6 block/curl: drop curl_aio_flush()
7 block/gluster: drop qemu_gluster_aio_flush_cb()
8 block/iscsi: drop iscsi_process_flush()
9 block/linux-aio: drop qemu_laio_completion_cb()
10 block/nbd: drop nbd_have_request()
11 block/rbd: drop qemu_rbd_aio_flush_cb()
12 block/sheepdog: drop have_co_req() and aio_flush_request()
13 block/ssh: drop return_true()
14 dataplane/virtio-blk: drop flush_true() and flush_io()
15 thread-pool: drop thread_pool_active()
Part 4 - remove io_flush arguments from aio functions
The big, mechanical patch that drops the io_flush argument:
16 tests: drop event_active_cb()
Note that I split Part 3 from Part 4 so it's easy to review individual block
drivers.
Stefan Hajnoczi (17):
block: stop relying on io_flush() in bdrv_drain_all()
dataplane/virtio-blk: check exit conditions before aio_poll()
tests: adjust test-aio to new aio_poll() semantics
tests: adjust test-thread-pool to new aio_poll() semantics
aio: stop using .io_flush()
block/curl: drop curl_aio_flush()
block/gluster: drop qemu_gluster_aio_flush_cb()
block/iscsi: drop iscsi_process_flush()
block/linux-aio: drop qemu_laio_completion_cb()
block/nbd: drop nbd_have_request()
block/rbd: drop qemu_rbd_aio_flush_cb()
block/sheepdog: drop have_co_req() and aio_flush_request()
block/ssh: drop return_true()
dataplane/virtio-blk: drop flush_true() and flush_io()
thread-pool: drop thread_pool_active()
tests: drop event_active_cb()
aio: drop io_flush argument
aio-posix.c | 36 ++++++------------
aio-win32.c | 37 ++++++++-----------
async.c | 4 +-
block.c | 47 ++++++++++++++++-------
block/curl.c | 25 ++-----------
block/gluster.c | 21 ++---------
block/iscsi.c | 10 +----
block/linux-aio.c | 18 +--------
block/nbd.c | 18 ++-------
block/rbd.c | 16 +-------
block/sheepdog.c | 33 ++++-------------
block/ssh.c | 12 +-----
hw/block/dataplane/virtio-blk.c | 25 +++----------
include/block/aio.h | 11 +-----
main-loop.c | 9 ++---
tests/test-aio.c | 82 +++++++++++++++++++++--------------------
tests/test-thread-pool.c | 24 ++++++------
thread-pool.c | 11 +-----
18 files changed, 155 insertions(+), 284 deletions(-)
--
1.8.1.4
next reply other threads:[~2013-04-17 13:08 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-04-17 13:07 Stefan Hajnoczi [this message]
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 01/17] block: stop relying on io_flush() in bdrv_drain_all() Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 02/17] dataplane/virtio-blk: check exit conditions before aio_poll() Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 03/17] tests: adjust test-aio to new aio_poll() semantics Stefan Hajnoczi
2013-04-17 13:12 ` Paolo Bonzini
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 04/17] tests: adjust test-thread-pool " Stefan Hajnoczi
2013-04-17 13:12 ` Paolo Bonzini
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 05/17] aio: stop using .io_flush() Stefan Hajnoczi
2013-04-17 13:20 ` Paolo Bonzini
2013-04-17 14:53 ` Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 06/17] block/curl: drop curl_aio_flush() Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 07/17] block/gluster: drop qemu_gluster_aio_flush_cb() Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 08/17] block/iscsi: drop iscsi_process_flush() Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 09/17] block/linux-aio: drop qemu_laio_completion_cb() Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 10/17] block/nbd: drop nbd_have_request() Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 11/17] block/rbd: drop qemu_rbd_aio_flush_cb() Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 12/17] block/sheepdog: drop have_co_req() and aio_flush_request() Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 13/17] block/ssh: drop return_true() Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 14/17] dataplane/virtio-blk: drop flush_true() and flush_io() Stefan Hajnoczi
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 15/17] thread-pool: drop thread_pool_active() Stefan Hajnoczi
2013-04-17 13:13 ` Paolo Bonzini
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 16/17] tests: drop event_active_cb() Stefan Hajnoczi
2013-04-17 13:13 ` Paolo Bonzini
2013-04-17 13:07 ` [Qemu-devel] [PATCH v2 17/17] aio: drop io_flush argument Stefan Hajnoczi
2013-04-17 13:15 ` Paolo Bonzini
2013-04-17 14:56 ` Stefan Hajnoczi
2013-04-17 13:20 ` [Qemu-devel] [PATCH v2 00/17] aio: drop io_flush() Paolo Bonzini
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=1366204073-8468-1-git-send-email-stefanha@redhat.com \
--to=stefanha@redhat.com \
--cc=aliguori@us.ibm.com \
--cc=kwolf@redhat.com \
--cc=pbonzini@redhat.com \
--cc=pingfank@linux.vnet.ibm.com \
--cc=qemu-devel@nongnu.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).