From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:59464)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <amit.shah@redhat.com>) id 1X8403-0005Jw-FW
	for qemu-devel@nongnu.org; Fri, 18 Jul 2014 04:58:50 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <amit.shah@redhat.com>) id 1X83zv-0002J8-KT
	for qemu-devel@nongnu.org; Fri, 18 Jul 2014 04:58:43 -0400
Received: from mx1.redhat.com ([209.132.183.28]:3547)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <amit.shah@redhat.com>) id 1X83zv-0002J2-Cc
	for qemu-devel@nongnu.org; Fri, 18 Jul 2014 04:58:35 -0400
Date: Fri, 18 Jul 2014 14:27:54 +0530
From: Amit Shah <amit.shah@redhat.com>
Message-ID: <20140718085754.GB15020@grmbl.mre>
References: <20140716115229.GA7741@amt.cnet>
	<CABYiri9mNpaY35EVGLTjR6c7SkY2HgttpN_3NihFy2jcOQ9raQ@mail.gmail.com>
	<CABYiri_vDLR=PATk1Lb99nDcJf4Q-yajayyMR1Tuv6bfwVUjgQ@mail.gmail.com>
	<53C6EE7C.60702@beyond.pl>
	<CABYiri8HdQhsoaf2=FHLG+KyrNj05FT015bNcEk9ngH6Dxo7dQ@mail.gmail.com>
	<53C79C41.4000800@beyond.pl> <53C7B989.9000203@beyond.pl>
	<CABYiri-aeP=0NcKaoaa=_au9sruvxnPVNri_Vy2KxKFHRj2fUw@mail.gmail.com>
	<53C7CEE5.4080006@beyond.pl> <53C8DF68.5040705@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <53C8DF68.5040705@redhat.com>
Content-Transfer-Encoding: quoted-printable
Subject: Re: [Qemu-devel] latest rc: virtio-blk hangs forever after migration
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: Fam Zheng <famz@redhat.com>, Andrey Korolyov <andrey@xdel.ru>, Marcin =?utf-8?Q?Gibu=C5=82a?= <m.gibula@beyond.pl>, "qemu-devel@nongnu.org" <qemu-devel@nongnu.org>, Marcelo Tosatti <mtosatti@redhat.com>

On (Fri) 18 Jul 2014 [10:48:40], Paolo Bonzini wrote:
> Il 17/07/2014 15:25, Marcin Gibu=C5=82a ha scritto:
> >+static void do_kvm_cpu_synchronize_state_always(void *arg)
> >+{
> >+    CPUState *cpu =3D arg;
> >+
> >+    kvm_arch_get_registers(cpu);
> >+}
> >+
>=20
> The name of the hack^Wfunction is tricky, because compared to
> do_kvm_cpu_synchronize_state there are three things you change:
>=20
> 1) you always synchronize the state
>=20
> 2) the next call to do_kvm_cpu_synchronize_state will do
> kvm_arch_get_registers
>=20
> 3) the next CPU entry will call kvm_arch_put_registers:
>=20
>         if (cpu->kvm_vcpu_dirty) {
>             kvm_arch_put_registers(cpu, KVM_PUT_RUNTIME_STATE);
>             cpu->kvm_vcpu_dirty =3D false;
>         }
>=20
> It is easy to find out if the "fix" is related to 1 or 2/3: just write
>=20
>      if (cpu->kvm_vcpu_dirty) {
>          printf ("do_kvm_cpu_synchronize_state_always: look at 2/3\n");
>          kvm_arch_get_registers(cpu);
>      } else {
>          printf ("do_kvm_cpu_synchronize_state_always: look at 1\n");
>      }
>=20
> To further refine between 2 and 3, I suppose you can set a breakpoint o=
n
> cpu_synchronize_all_states and kvm_cpu_exec, and see which is called fi=
rst
> after cpu_synchronize_all_states_always.
>=20
> I still lean very much towards reverting the patches now.  We can reapp=
ly
> them, fixed, in 2.1.1.

FWIW I agree with this plan.

		Amit