From: Peter Xu <peterx@redhat.com>
To: qemu-devel@nongnu.org
Cc: "Avihai Horon" <avihaih@nvidia.com>,
"Cédric Le Goater" <clg@redhat.com>,
peterx@redhat.com, "Fabiano Rosas" <farosas@suse.de>,
"Alex Williamson" <alex.williamson@redhat.com>
Subject: [PATCH v3 0/8] Migration: Make misc.h helpers available for whole VM lifecycle
Date: Thu, 24 Oct 2024 17:30:48 -0400 [thread overview]
Message-ID: <20241024213056.1395400-1-peterx@redhat.com> (raw)
Based-on: <20241024165627.1372621-1-peterx@redhat.com>
CI: https://gitlab.com/peterx/qemu/-/pipelines/1511349805
This is a follow up of below patch from Avihai as a replacement:
https://lore.kernel.org/qemu-devel/20241020130108.27148-3-avihaih@nvidia.com/
This is v3 of the series, and it happened again: changelog doesn't make
sense to compare to v2 because it's rewritten. Meanwhile, now this series
is based on the other TYPE_SINGLETON series I posted just now:
https://lore.kernel.org/r/20241024165627.1372621-1-peterx@redhat.com
It turns out I found more things to cleanup, as the versions spin up.
Patch 1: I found that object_ref() in migration thread is actually also
racy.. this should fix it.
Patch 2-7: It turns out I decided to clean things up first, then it'll
make my last patch easier on adding the mutex protection for
the current_migration reference
Patch 8: The fix for NULL-dereference / race for the exported
functions. VFIO can hit it only because VFIO's specialty
in using migration helpers in vmstate handlers, I guess. I
found most functions always safe because even if they're used
outside migration/ they're most likely still invoked with
migration thread context. So I found only two functions that
really need protections, exactly what VFIO is using.
Comments welcomed, thanks.
Peter Xu (8):
migration: Take migration object refcount earlier for threads
migration: Unexport dirty_bitmap_mig_init()
migration: Unexport ram_mig_init()
migration: Drop migration_is_setup_or_active()
migration: Drop migration_is_idle()
migration: Drop migration_is_device()
migration: Unexport migration_is_active()
migration: Protect updates to current_migration with a mutex
include/migration/misc.h | 11 +---
migration/migration.h | 8 +++
migration/ram.h | 1 +
hw/vfio/common.c | 6 +-
hw/virtio/virtio-mem.c | 2 +-
migration/migration.c | 118 +++++++++++++++++++--------------------
migration/ram.c | 7 +--
net/vhost-vdpa.c | 3 +-
system/dirtylimit.c | 3 +-
system/qdev-monitor.c | 4 +-
10 files changed, 81 insertions(+), 82 deletions(-)
--
2.45.0
next reply other threads:[~2024-10-24 21:32 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-24 21:30 Peter Xu [this message]
2024-10-24 21:30 ` [PATCH v3 1/8] migration: Take migration object refcount earlier for threads Peter Xu
2024-10-25 8:36 ` Cédric Le Goater
2024-10-29 18:48 ` Fabiano Rosas
2024-10-24 21:30 ` [PATCH v3 2/8] migration: Unexport dirty_bitmap_mig_init() Peter Xu
2024-10-29 18:48 ` Fabiano Rosas
2024-10-24 21:30 ` [PATCH v3 3/8] migration: Unexport ram_mig_init() Peter Xu
2024-10-25 8:36 ` Cédric Le Goater
2024-10-29 18:49 ` Fabiano Rosas
2024-10-24 21:30 ` [PATCH v3 4/8] migration: Drop migration_is_setup_or_active() Peter Xu
2024-10-25 8:59 ` Cédric Le Goater
2024-10-29 19:04 ` Fabiano Rosas
2024-10-24 21:30 ` [PATCH v3 5/8] migration: Drop migration_is_idle() Peter Xu
2024-10-25 12:39 ` Cédric Le Goater
2024-10-29 19:07 ` Fabiano Rosas
2024-10-24 21:30 ` [PATCH v3 6/8] migration: Drop migration_is_device() Peter Xu
2024-10-24 21:30 ` [PATCH v3 7/8] migration: Unexport migration_is_active() Peter Xu
2024-10-28 7:43 ` Avihai Horon
2024-10-28 15:45 ` Peter Xu
2024-10-28 16:41 ` Avihai Horon
2024-10-28 16:58 ` Peter Xu
2024-10-28 17:20 ` Avihai Horon
2024-10-28 19:06 ` Peter Xu
2024-10-30 14:39 ` Avihai Horon
2024-10-30 14:43 ` Peter Xu
2024-10-24 21:30 ` [PATCH v3 8/8] migration: Protect updates to current_migration with a mutex Peter Xu
2024-10-25 12:50 ` Cédric Le Goater
2024-10-28 15:50 ` Peter Xu
2024-11-04 15:26 ` Cédric Le Goater
2024-10-29 20:22 ` [PATCH v3 0/8] Migration: Make misc.h helpers available for whole VM lifecycle Peter Xu
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=20241024213056.1395400-1-peterx@redhat.com \
--to=peterx@redhat.com \
--cc=alex.williamson@redhat.com \
--cc=avihaih@nvidia.com \
--cc=clg@redhat.com \
--cc=farosas@suse.de \
--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).