From: Eric Blake <eblake@redhat.com>
To: qemu-devel@nongnu.org
Cc: qemu-block@nongnu.org, berrange@redhat.com, kwolf@redhat.com
Subject: [PATCH v2 00/12] Fix deadlock with bdrv_open of self-served NBD
Date: Sat, 8 Nov 2025 16:59:21 -0600 [thread overview]
Message-ID: <20251108230525.3169174-14-eblake@redhat.com> (raw)
v1 was here:
https://lore.kernel.org/qemu-devel/20251103202849.3687643-10-eblake@redhat.com/T/#t
In v2:
- split or otherwise add a few patches
- reorder latent bug fixes earlier in series
- improved commit messages
- improved consistency in trace naming, and make traces unconditional
- rearrange contents of QIONetListener rather than adding field to
QIOChannelSocket
- add new opt-in API rather than retro-fitting existing
qio_net_listener_set_client_func
- use existing QIO aio functions rather than lower-level aio.h functions
- further enhance iotest
See individual patches for more details on changes since v1. The series
is different enough that I did not preserve any Reviewed-by tags from v1.
A deadlock fix counts as a bug fix, so even though this missed soft
freeze, I'm hoping we can get it in -rc1 or at the latest -rc2. But
it's also not a new regression, but a long-standing issue; so if it
slides into 11.0 out of concerns of being too risky this close to
release, I understand.
Key:
[----] : patches are identical
[####] : number of functional differences between upstream/downstream patch
[down] : patch is downstream-only
The flags [FC] indicate (F)unctional and (C)ontextual differences, respectively
001/12:[down] 'iotests: Drop execute permissions on vvfat.out'
002/12:[0027] [FC] 'qio: Add trace points to net_listener'
003/12:[down] 'qio: Unwatch before notify in QIONetListener'
004/12:[0026] [FC] 'qio: Remember context of qio_net_listener_set_client_func_full'
005/12:[0003] [FC] 'qio: Minor optimization when callback function is unchanged'
006/12:[0039] [FC] 'qio: Factor out helpers qio_net_listener_[un]watch'
007/12:[0002] [FC] 'qio: Hoist ref of listener outside loop'
008/12:[down] 'qio: Provide accessor around QIONetListener->sioc'
009/12:[down] 'qio: Prepare NetListener to use AioContext'
010/12:[down] 'qio: Add QIONetListener API for using AioContext'
011/12:[down] 'nbd: Avoid deadlock in client connecting to same-process server'
012/12:[0035] [FC] 'iotests: Add coverage of recent NBD qio deadlock fix'
Eric Blake (12):
iotests: Drop execute permissions on vvfat.out
qio: Add trace points to net_listener
qio: Unwatch before notify in QIONetListener
qio: Remember context of qio_net_listener_set_client_func_full
qio: Minor optimization when callback function is unchanged
qio: Factor out helpers qio_net_listener_[un]watch
qio: Hoist ref of listener outside loop
qio: Provide accessor around QIONetListener->sioc
qio: Prepare NetListener to use AioContext
qio: Add QIONetListener API for using AioContext
nbd: Avoid deadlock in client connecting to same-process server
iotests: Add coverage of recent NBD qio deadlock fix
include/io/net-listener.h | 47 +++-
blockdev-nbd.c | 4 +-
chardev/char-socket.c | 3 +-
io/net-listener.c | 233 ++++++++++++------
migration/socket.c | 5 +-
ui/vnc.c | 36 ++-
io/trace-events | 5 +
tests/qemu-iotests/tests/nbd-in-qcow2-chain | 94 +++++++
.../qemu-iotests/tests/nbd-in-qcow2-chain.out | 75 ++++++
tests/qemu-iotests/tests/vvfat.out | 0
10 files changed, 411 insertions(+), 91 deletions(-)
create mode 100755 tests/qemu-iotests/tests/nbd-in-qcow2-chain
create mode 100644 tests/qemu-iotests/tests/nbd-in-qcow2-chain.out
mode change 100755 => 100644 tests/qemu-iotests/tests/vvfat.out
--
2.51.1
next reply other threads:[~2025-11-08 23:19 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-08 22:59 Eric Blake [this message]
2025-11-08 22:59 ` [PATCH v2 01/12] iotests: Drop execute permissions on vvfat.out Eric Blake
2025-11-10 15:57 ` Daniel P. Berrangé
2025-11-08 22:59 ` [PATCH v2 02/12] qio: Add trace points to net_listener Eric Blake
2025-11-10 15:58 ` Daniel P. Berrangé
2025-11-08 22:59 ` [PATCH v2 03/12] qio: Unwatch before notify in QIONetListener Eric Blake
2025-11-10 16:00 ` Daniel P. Berrangé
2025-11-08 22:59 ` [PATCH v2 04/12] qio: Remember context of qio_net_listener_set_client_func_full Eric Blake
2025-11-10 16:08 ` Daniel P. Berrangé
2025-11-08 22:59 ` [PATCH v2 05/12] qio: Minor optimization when callback function is unchanged Eric Blake
2025-11-10 16:09 ` Daniel P. Berrangé
2025-11-08 22:59 ` [PATCH v2 06/12] qio: Factor out helpers qio_net_listener_[un]watch Eric Blake
2025-11-10 16:14 ` Daniel P. Berrangé
2025-11-08 22:59 ` [PATCH v2 07/12] qio: Hoist ref of listener outside loop Eric Blake
2025-11-11 14:43 ` Daniel P. Berrangé
2025-11-08 22:59 ` [PATCH v2 08/12] qio: Provide accessor around QIONetListener->sioc Eric Blake
2025-11-10 18:31 ` Eric Blake
2025-11-11 14:15 ` Daniel P. Berrangé
2025-11-08 22:59 ` [PATCH v2 09/12] qio: Prepare NetListener to use AioContext Eric Blake
2025-11-11 14:17 ` Daniel P. Berrangé
2025-11-08 22:59 ` [PATCH v2 10/12] qio: Add QIONetListener API for using AioContext Eric Blake
2025-11-11 14:18 ` Daniel P. Berrangé
2025-11-08 22:59 ` [PATCH v2 11/12] nbd: Avoid deadlock in client connecting to same-process server Eric Blake
2025-11-11 14:20 ` Daniel P. Berrangé
2025-11-08 22:59 ` [PATCH v2 12/12] iotests: Add coverage of recent NBD qio deadlock fix Eric Blake
2025-11-10 16:19 ` Daniel P. Berrangé
2025-11-12 6:35 ` Vladimir Sementsov-Ogievskiy
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=20251108230525.3169174-14-eblake@redhat.com \
--to=eblake@redhat.com \
--cc=berrange@redhat.com \
--cc=kwolf@redhat.com \
--cc=qemu-block@nongnu.org \
--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).