Received: from eggs.gnu.org ([209.51.188.92]:41338) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1goW7x-00030V-Az for qemu-devel@nongnu.org; Tue, 29 Jan 2019 11:20:46 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1goW7v-0002Ge-Sb for qemu-devel@nongnu.org; Tue, 29 Jan 2019 11:20:45 -0500 Received: from mx1.redhat.com ([209.132.183.28]:38762) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1goW7v-0002Fk-LF for qemu-devel@nongnu.org; Tue, 29 Jan 2019 11:20:43 -0500 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.phx2.redhat.com [10.5.11.16]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 1981EC059B7A for ; Tue, 29 Jan 2019 16:20:42 +0000 (UTC) From: Markus Armbruster References: <20190129051432.22023-1-peterx@redhat.com> Date: Tue, 29 Jan 2019 17:20:40 +0100 In-Reply-To: <20190129051432.22023-1-peterx@redhat.com> (Peter Xu's message of "Tue, 29 Jan 2019 13:14:32 +0800") Message-ID: <87lg33306f.fsf@dusky.pond.sub.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH] iothread: fix iothread hang when stop too soon List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Xu Cc: qemu-devel@nongnu.org, =?utf-8?B?THVrw6HFoQ==?= Doktor , Thomas Huth , "Dr . David Alan Gilbert" , Stefan Hajnoczi , Paolo Bonzini , =?utf-8?Q?Marc-Andr=C3=A9_Lureau?= Marc-Andr=C3=A9 (cc'ed) recently fixed a deadlock (commit 34f1f3e06d8 "monitor: avoid potential dead-lock when cleaning up"). Looks like we got more. Peter Xu writes: > Lukas reported an hard to reproduce QMP iothread hang on s390 that > QEMU might hang at pthread_join() of the QMP monitor iothread before > quitting: > > Thread 1 > #0 0x000003ffad10932c in pthread_join > #1 0x0000000109e95750 in qemu_thread_join > at /home/thuth/devel/qemu/util/qemu-thread-posix.c:570 > #2 0x0000000109c95a1c in iothread_stop > #3 0x0000000109bb0874 in monitor_cleanup > #4 0x0000000109b55042 in main > > While the iothread is still in the main loop: > > Thread 4 > #0 0x000003ffad0010e4 in ?? > #1 0x000003ffad553958 in g_main_context_iterate.isra.19 > #2 0x000003ffad553d90 in g_main_loop_run > #3 0x0000000109c9585a in iothread_run > at /home/thuth/devel/qemu/iothread.c:74 > #4 0x0000000109e94752 in qemu_thread_start > at /home/thuth/devel/qemu/util/qemu-thread-posix.c:502 > #5 0x000003ffad10825a in start_thread > #6 0x000003ffad00dcf2 in thread_start > > IMHO it's because there's a race between the main thread and iothread > when stopping the thread in following sequence: > > main thread iothread > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D =3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > aio_poll() > iothread_get_g_main_context > set iothread->worker_context > iothread_stop > schedule iothread_stop_bh > execute iothread_stop_bh [1] > set iothread->running=3Dfalse > (since main_loop=3D=3DNULL so > skip to quit main loop. > Note: although main_loop is > NULL but worker_context is > not!) > atomic_read(&iothread->worker_conte= xt) [2] > create main_loop object > g_main_loop_run() [3] > pthread_join() [4] > > We can see that when execute iothread_stop_bh() at [1] it's possible > that main_loop is still NULL because it's only created until the first > check of the worker_context later at [2]. Then the iothread will hang > in the main loop [3] and it'll starve the main thread too [4]. > > Here the simple solution should be that we check again the "running" > variable before check against worker_context. > > CC: Thomas Huth > CC: Dr. David Alan Gilbert > CC: Stefan Hajnoczi > CC: Luk=C3=A1=C5=A1 Doktor > CC: Markus Armbruster > CC: Eric Blake > CC: Paolo Bonzini > Reported-by: Luk=C3=A1=C5=A1 Doktor > Signed-off-by: Peter Xu > --- > > This hasn't yet been verified on the initial s390 systems, but since I > can reproduce it locally with this code clip: > > IOThread *iothread =3D iothread_create("test", NULL); > iothread_get_g_main_context(iothread); > iothread_stop(iothread); > > so I'm still posting this out for review first in case it was hit by > other users. > --- > iothread.c | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > diff --git a/iothread.c b/iothread.c > index 2fb1cdf55d..e615b7ae52 100644 > --- a/iothread.c > +++ b/iothread.c > @@ -63,7 +63,11 @@ static void *iothread_run(void *opaque) > while (iothread->running) { > aio_poll(iothread->ctx, true); >=20=20 > - if (atomic_read(&iothread->worker_context)) { > + /* > + * We must check the running state again in case it was > + * changed in previous aio_poll() > + */ > + if (iothread->running && atomic_read(&iothread->worker_context))= { > GMainLoop *loop; >=20=20 > g_main_context_push_thread_default(iothread->worker_context);