From mboxrd@z Thu Jan  1 00:00:00 1970
From: =?UTF-8?B?QW5kcmVhcyBGw6RyYmVy?= <afaerber@suse.de>
Subject: Re: [Qemu-devel] [PATCH qom-cpu for-1.4?] kvm: Pass CPUState to kvm_on_sigbus_vcpu()
Date: Mon, 28 Jan 2013 16:29:35 +0100
Message-ID: <5106995F.4010702@suse.de>
References: <1359129329-21274-1-git-send-email-afaerber@suse.de> <20130128135048.GD22871@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: QUOTED-PRINTABLE
Cc: Marcelo Tosatti <mtosatti@redhat.com>, qemu-devel@nongnu.org,
	Overall <kvm@vger.kernel.org>
To: Gleb Natapov <gleb@redhat.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from cantor2.suse.de ([195.135.220.15]:41113 "EHLO mx2.suse.de"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1754166Ab3A1P3p (ORCPT <rfc822;kvm@vger.kernel.org>);
	Mon, 28 Jan 2013 10:29:45 -0500
In-Reply-To: <20130128135048.GD22871@redhat.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

Am 28.01.2013 14:50, schrieb Gleb Natapov:
> On Fri, Jan 25, 2013 at 04:55:29PM +0100, Andreas F=C3=A4rber wrote:
>> Since commit 20d695a9254c1b086a456d3b79a3c311236643ba (kvm: Pass
>> CPUState to kvm_arch_*) CPUArchState is no longer needed.
>>
>> Allows to change qemu_kvm_eat_signals() argument as well.
>>
>> Signed-off-by: Andreas F=C3=A4rber <afaerber@suse.de>
> Reviewed-by: Gleb Natapov <gleb@redhat.com>

Thanks, applied to qom-cpu:
https://github.com/afaerber/qemu-cpu/commits/qom-cpu

Background was:
https://lists.nongnu.org/archive/html/qemu-devel/2013-01/msg03087.html

<<<
[...] qemu_init_vcpu() still operates on CPUArchState and thus cannot b=
e
moved into CPUClass yet. The reason is that
cpus.c:qemu_kvm_cpu_thread_fn sets cpu_single_env, and I do not see a
solution for that - suggestions or patches welcome.

However, I see that kvm-all.c:kvm_on_sigbus_vcpu() can be switched to
CPUState now, so that cpus.c:qemu_kvm_eat_signals() can be changed to
CPUState, used from cpus.c:qemu_kvm_wait_io_event().
But cpus.c:cpu_thread_is_idle() still uses env->halted, which is blocke=
d
by the search for an acceptable solution to flush the TLB at CPUState
level (exec.c:cpu_common_post_load()).
>>>

A less elegant but working solution is on my qom-cpu-8 branch (based of=
f
qom-cpu-next): I introduced a void *env_ptr CPUState field. While
potentially opening a can of worms I wanted to avoid, it allows us to
defer finding a solution to the target_ulong-dependent TLB some more.

Regards,
Andreas

--=20
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 N=C3=BCrnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imend=C3=B6rffer; HRB 16746 AG N=C3=
=BCrnberg