From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([140.186.70.92]:55101)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <paolo.bonzini@gmail.com>) id 1Rgepn-0007XX-FU
	for qemu-devel@nongnu.org; Fri, 30 Dec 2011 10:57:32 -0500
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <paolo.bonzini@gmail.com>) id 1Rgepm-00035P-Ee
	for qemu-devel@nongnu.org; Fri, 30 Dec 2011 10:57:31 -0500
Received: from mail-ee0-f45.google.com ([74.125.83.45]:39087)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <paolo.bonzini@gmail.com>) id 1Rgepm-00035J-AA
	for qemu-devel@nongnu.org; Fri, 30 Dec 2011 10:57:30 -0500
Received: by eekb45 with SMTP id b45so16165597eek.4
	for <qemu-devel@nongnu.org>; Fri, 30 Dec 2011 07:57:29 -0800 (PST)
Sender: Paolo Bonzini <paolo.bonzini@gmail.com>
Message-ID: <4EFDDF66.3090909@redhat.com>
Date: Fri, 30 Dec 2011 16:57:26 +0100
From: Paolo Bonzini <pbonzini@redhat.com>
MIME-Version: 1.0
References: <CAF4+tmopXFry77uZitO5=KjPkgcjJE_gwpY8V051Vk=TWsx2QA@mail.gmail.com>
In-Reply-To: <CAF4+tmopXFry77uZitO5=KjPkgcjJE_gwpY8V051Vk=TWsx2QA@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH] Init win32 CRITICAL_SECTION before
 starting thread; crash when attaching disks
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: Bogdan Harjoc <harjoc@gmail.com>
Cc: qemu-devel@nongnu.org

On 12/29/2011 06:29 PM, Bogdan Harjoc wrote:
> Git commit 8d3bc51 crashes on win32 on startup because
> qemu_tcg_init_vcpu calls:
>
> qemu_thread_create(th, qemu_tcg_cpu_thread_fn, ...
> ...
> qemu_thread_get_handle(th)
>
> which locks th->data->cs, a CRITICAL_SECTION which is initialized only
> in the thread_fn, so it finds garbage.
>
> Attached patch initializes it before calling _beginthreadex. GDB/windbg
> probably start newly created threads sooner, because this doesn't happen
> under a debugger.
>
> With the patch below it boots until it crashes somewhere while attaching
> disks (-hda raw_img).
>
> "bt" in gdb only returns "#0  0x00000000 in ??" and generate-core-file
> didn't work.
>
> Cheers,
>
> diff -du qemu-8d3bc51\qemu-thread-win32.c
> qemu-8d3bc51-new\qemu-thread-win32.c
> --- qemu-8d3bc51\qemu-thread-win32.c    Tue Dec 27 17:28:58 2011
> +++ qemu-8d3bc51-new\qemu-thread-win32.c    Thu Dec 29 18:59:50 2011
> @@ -215,8 +215,6 @@
>       if (data->mode == QEMU_THREAD_DETACHED) {
>           g_free(data);
>           data = NULL;
> -    } else {
> -        InitializeCriticalSection(&data->cs);
>       }
>       TlsSetValue(qemu_thread_tls_index, data);
>       qemu_thread_exit(start_routine(thread_arg));
> @@ -287,6 +285,10 @@
>       data->arg = arg;
>       data->mode = mode;
>       data->exited = false;
> +
> +    if (data->mode != QEMU_THREAD_DETACHED) {
> +        InitializeCriticalSection(&data->cs);
> +    }
>
>       hThread = (HANDLE) _beginthreadex(NULL, 0, win32_start_routine,
>                                         data, 0, &thread->tid);
>

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>