On 09/25/2017 08:57 AM, Stefan Hajnoczi wrote: > Throttling group members are restarted on config change and by > bdrv_drained_begin(). Pending timers should be cancelled before > restarting the queue, otherwise requests see that tg->any_timer_armed[] > is already set and do not schedule a timer. > > For example, a hang occurs at least since QEMU 2.10.0 with -drive > iops=100 because no further timers will be scheduled: > > (guest)$ dd if=/dev/zero of=/dev/vdb oflag=direct count=1000 > (qemu) stop > (qemu) cont > ...I/O is stuck... > > This can be fixed by calling throttle_group_detach_aio_context() from a > bdrv_drained_begin/end() region. This way timers are quiesced properly, > requests are drained, and other throttle group members are scheduled, if > necessary. > > Reported-by: Yongxue Hong <yh...@redhat.com> > Signed-off-by: Stefan Hajnoczi <stefa...@redhat.com> > ---
Reviewed-by: Eric Blake <ebl...@redhat.com> -- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3266 Virtualization: qemu.org | libvirt.org
signature.asc
Description: OpenPGP digital signature
