From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1KiVC2-00046l-4K for qemu-devel@nongnu.org; Wed, 24 Sep 2008 10:18:14 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1KiVC0-000463-8N for qemu-devel@nongnu.org; Wed, 24 Sep 2008 10:18:13 -0400 Received: from [199.232.76.173] (port=37088 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1KiVC0-00045y-5N for qemu-devel@nongnu.org; Wed, 24 Sep 2008 10:18:12 -0400 Received: from e2.ny.us.ibm.com ([32.97.182.142]:48806) by monty-python.gnu.org with esmtps (TLS-1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.60) (envelope-from ) id 1KiVC0-0002Jg-1e for qemu-devel@nongnu.org; Wed, 24 Sep 2008 10:18:12 -0400 Received: from d01relay02.pok.ibm.com (d01relay02.pok.ibm.com [9.56.227.234]) by e2.ny.us.ibm.com (8.13.8/8.13.8) with ESMTP id m8OEI8sb015533 for ; Wed, 24 Sep 2008 10:18:08 -0400 Received: from d01av03.pok.ibm.com (d01av03.pok.ibm.com [9.56.224.217]) by d01relay02.pok.ibm.com (8.13.8/8.13.8/NCO v9.1) with ESMTP id m8OEI8ne219008 for ; Wed, 24 Sep 2008 10:18:08 -0400 Received: from d01av03.pok.ibm.com (loopback [127.0.0.1]) by d01av03.pok.ibm.com (8.12.11.20060308/8.13.3) with ESMTP id m8OEI8wB011736 for ; Wed, 24 Sep 2008 10:18:08 -0400 Message-ID: <48DA4BE3.30403@us.ibm.com> Date: Wed, 24 Sep 2008 09:17:07 -0500 From: Anthony Liguori MIME-Version: 1.0 References: <1222202051-26701-1-git-send-email-aliguori@us.ibm.com> <48DA021F.2050504@redhat.com> In-Reply-To: <48DA021F.2050504@redhat.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: [Qemu-devel] Re: [PATCH] Implement an fd pool to get real AIO with posix-aio Reply-To: qemu-devel@nongnu.org List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Avi Kivity Cc: Ryan Harper , qemu-devel@nongnu.org, kvm Avi Kivity wrote: > Anthony Liguori wrote: > > dup()ing the fd on each request is unnecessary work; would be better > to cache the duped fd. Yeah, I was concerned about this too. Ryan reran the fio benchmark and the submission latency and completion latency were identical to the linux-aio patches. That suggests that the overhead of dup() is lost in the noise. Since this is simpler and keeps the number of open file descriptors as low as possible, I was happy about that. Regards, Anthony Liguori > Of course, if this is just a stepping stone, it doesn't matter very much. >