On 3/1/21 5:49 AM, Michal Privoznik wrote: > Since v6.2.0-rc1~238 (and friends) QMP processing was moved to a > per-domain thread. Therefore, it is now safe to call APIs from > the event loop thread (e.g. just like qemu shim is doing in > qemuShimEventLoop(). However, it is still important to let the > event loop run after each API call (obviously). > > Signed-off-by: Michal Privoznik <mprivozn@xxxxxxxxxx> > --- > docs/drvqemu.html.in | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/docs/drvqemu.html.in b/docs/drvqemu.html.in > index 31d3fee213..3cdd04aa1e 100644 > --- a/docs/drvqemu.html.in > +++ b/docs/drvqemu.html.in > @@ -158,8 +158,10 @@ qemu+ssh://root@xxxxxxxxxxx/system (remote access, SSH tunnelled) > in mind, applications must <strong>NEVER</strong> invoke API > calls from the event loop thread itself, only other threads. > Not following this rule will lead to deadlocks in the API. > - This restriction is intended to be lifted in a future release > - of libvirt, once QMP processing moves to a dedicated thread. > + This restriction was lifted starting from 6.2.0 release, when > + QMP processing moved to a dedicated thread. However, it is > + important to let the event loop run after each API call, even > + the ones made from the vent loop thread itself. event loop > </p> > > <h2><a id="security">Driver security architecture</a></h2> > -- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3226 Virtualization: qemu.org | libvirt.org
