From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:40593) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1X1dpM-0003r4-8z for qemu-devel@nongnu.org; Mon, 30 Jun 2014 11:49:16 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1X1dpD-0007G1-0o for qemu-devel@nongnu.org; Mon, 30 Jun 2014 11:49:08 -0400 Received: from mail-pa0-f41.google.com ([209.85.220.41]:49744) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1X1dpC-0007Fl-Rc for qemu-devel@nongnu.org; Mon, 30 Jun 2014 11:48:58 -0400 Received: by mail-pa0-f41.google.com with SMTP id fb1so8805817pad.14 for ; Mon, 30 Jun 2014 08:48:57 -0700 (PDT) From: Ming Lei Date: Mon, 30 Jun 2014 23:47:37 +0800 Message-Id: <1404143260-9368-1-git-send-email-ming.lei@canonical.com> Subject: [Qemu-devel] [PATCH v1 0/3] linux-aio: introduce submit I/O at batch List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Maydell , qemu-devel@nongnu.org, Paolo Bonzini , Stefan Hajnoczi Cc: Kevin Wolf , Fam Zheng , "Michael S. Tsirkin" Hi, The commit 580b6b2aa2(dataplane: use the QEMU block layer for I/O) introduces ~40% throughput regression on virtio-blk dataplane, and one of causes is that submitting I/O at batch is removed. This patchset trys to introduce this mechanism on block, at least, linux-aio can benefit from that. With these patches, it is observed that thoughout on virtio-blk dataplane can be improved a lot, see data in commit log of patch 3/3. It should be possible to apply the batch mechanism to other devices (such as virtio-scsi) too. V1: - move queuing io stuff into linux-aio.c as suggested by Paolo Thanks, -- Ming Lei