From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([208.118.235.92]:46351)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1T2o2w-0001eZ-E8
	for qemu-devel@nongnu.org; Sat, 18 Aug 2012 14:46:55 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1T2o2v-0000ey-Ae
	for qemu-devel@nongnu.org; Sat, 18 Aug 2012 14:46:54 -0400
Received: from mx1.redhat.com ([209.132.183.28]:29233)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1T2o2v-0000eu-2d
	for qemu-devel@nongnu.org; Sat, 18 Aug 2012 14:46:53 -0400
Message-ID: <502FE317.3020607@redhat.com>
Date: Sat, 18 Aug 2012 20:46:47 +0200
From: Paolo Bonzini <pbonzini@redhat.com>
MIME-Version: 1.0
References: <1344426944-7638-1-git-send-email-pbonzini@redhat.com>
	<1344426944-7638-3-git-send-email-pbonzini@redhat.com>
	<502951BC.1090400@linux.vnet.ibm.com>
In-Reply-To: <502951BC.1090400@linux.vnet.ibm.com>
Content-Type: text/plain; charset=ISO-8859-15
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH 2/2] 9p-synth: use mutex on read-side
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: Harsh Bora <harsh@linux.vnet.ibm.com>
Cc: qemu-devel@nongnu.org, aneesh.kumar@linux.vnet.ibm.com

Il 13/08/2012 21:13, Harsh Bora ha scritto:
>>
> 
> Hi Paolo,
> 
> The rcu_read_[un]lock() macros were added as no-ops (based on your
> inputs on #qemu) to replace reader-writer locks with RCU based locking
> as suggested while proposing QemuRWLock API for RW locks (See
> http://lists.gnu.org/archive/html/qemu-devel/2011-10/msg00192.html).

The rwlock was also premature optimization IMHO.

> v9fs_synth_mutex was actually a pthread_rwlock_t earlier.
> I am not sure if reader lock would be better than having a plain mutex
> for readers as well.
> 
> Aneesh, inputs ?
> 
> Also, if we are going forward with this change, we may also want to
> remove definition of QLIST_INSERT_HEAD_RCU, since the code being remove
> below is the only consumer of that macro.

It's in a library file, so I don't mind leaving it in.  I think liu ping
fan's patches were using it.

Paolo