From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from [140.186.70.92] (port=52670 helo=eggs.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1OOvws-00087u-Rz
	for qemu-devel@nongnu.org; Wed, 16 Jun 2010 12:58:49 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69)
	(envelope-from <pbonzini@redhat.com>) id 1OOvwK-0001Mj-Cr
	for qemu-devel@nongnu.org; Wed, 16 Jun 2010 12:58:13 -0400
Received: from mx1.redhat.com ([209.132.183.28]:14792)
	by eggs.gnu.org with esmtp (Exim 4.69)
	(envelope-from <pbonzini@redhat.com>) id 1OOvwK-0001Mc-5t
	for qemu-devel@nongnu.org; Wed, 16 Jun 2010 12:58:12 -0400
Message-ID: <4C18FFB5.3050001@redhat.com>
Date: Wed, 16 Jun 2010 18:45:41 +0200
From: Paolo Bonzini <pbonzini@redhat.com>
MIME-Version: 1.0
Subject: Re: [Qemu-devel] Re: [PATCH V4 2/3] qemu: Generic task offloading
	framework: threadlets
References: <20100616115404.10988.62371.stgit@localhost.localdomain>
	<20100616115656.10988.96529.stgit@localhost.localdomain>
	<4C18C4C8.8090901@redhat.com>
	<20100616142236.GA20052@shareable.org>
	<4C18DFD7.1090102@redhat.com> <4C18E1E8.3030606@linux.vnet.ibm.com>
	<20100616145858.GA19642@shareable.org>
	<20100616150714.GA24172@shareable.org>
In-Reply-To: <20100616150714.GA24172@shareable.org>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
List-Id: qemu-devel.nongnu.org
List-Unsubscribe: <http://lists.nongnu.org/mailman/listinfo/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: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Jamie Lokier <jamie@shareable.org>
Cc: Gautham R Shenoy <ego@in.ibm.com>, Qemu-development List <qemu-devel@nongnu.org>, Anthony Liguori <aliguori@linux.vnet.ibm.com>, "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>, Corentin Chary <corentin.chary@gmail.com>, Avi Kivity <avi@redhat.com>

On 06/16/2010 05:07 PM, Jamie Lokier wrote:
> Putting the whole condition inside the mutex->cond_wait region, not
> just empty queue test, will remove the need for timed wait.

Yes, the condition must include all the cases when you broadcast.

> Broadcast
> is still needed, or alternatively a cond_signal from each exiting
> thread will allow them to wake and close without a thundering herd.

If the pthreads cause a thundering herd, that's the pthreads problem.  A 
properly implemented condvar will just requeue the threads to the mutex. 
  NPTL does, the Win32 condvar most likely won't. :)

Anyway if Corentin doesn't need the flush_threadlets_queue all this can 
be dropped.

Paolo