From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1JnVix-0003RT-Mb
	for qemu-devel@nongnu.org; Sun, 20 Apr 2008 05:20:40 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1JnViu-0003Qe-LB
	for qemu-devel@nongnu.org; Sun, 20 Apr 2008 05:20:37 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1JnVit-0003Q6-UO
	for qemu-devel@nongnu.org; Sun, 20 Apr 2008 05:20:36 -0400
Received: from [77.75.163.100] (helo=smtp.hotelhot.dk)
	by monty-python.gnu.org with esmtp (Exim 4.60)
	(envelope-from <mail@flac.kalibalik.dk>) id 1JnVit-0000Q9-H6
	for qemu-devel@nongnu.org; Sun, 20 Apr 2008 05:20:35 -0400
Received: from smtp.hotelhot.dk (localhost [127.0.0.1])
	by smtp.hotelhot.dk (Postfix) with ESMTP id 8478814068
	for <qemu-devel@nongnu.org>; Sun, 20 Apr 2008 11:20:39 +0200 (CEST)
Received: from kalibalik.dk (router.kalibalik.dk [192.168.0.1])
	by smtp.hotelhot.dk (Postfix) with SMTP id 693E614062
	for <qemu-devel@nongnu.org>; Sun, 20 Apr 2008 11:20:38 +0200 (CEST)
From: Anders Melchiorsen <mail@flac.kalibalik.dk>
Date: Sun, 20 Apr 2008 11:20:27 +0200
Message-Id: <12086832292508-git-send-email-mail@flac.kalibalik.dk>
Subject: [Qemu-devel] Linux SIGIO handling changes
Reply-To: qemu-devel@nongnu.org
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.gnu.org/pipermail/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: qemu-devel@nongnu.org

I am resending this patch, split into two parts now.

This cleans up SIGIO handling to improve latency:
- SIGALRM for alarm timers
- enable SIGIO on qemu_set_fd_handler2()

The issue was found in KVM, where it is much more visible,
because there is no periodic timer. However, it has been
confirmed (by Aurelien Jarno) that even for qemu, this
approach "improves network transfers in a huge way".

Please apply, or give a firm rejection so I can stop resending.

Cheers,
Anders.