From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59286) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1acunm-0001QI-GO for qemu-devel@nongnu.org; Mon, 07 Mar 2016 08:02:27 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1acunh-0003B6-EY for qemu-devel@nongnu.org; Mon, 07 Mar 2016 08:02:22 -0500 Received: from mx1.redhat.com ([209.132.183.28]:37655) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1acunh-0003B0-9O for qemu-devel@nongnu.org; Mon, 07 Mar 2016 08:02:17 -0500 References: <56D6033E.2070402@redhat.com> From: Paolo Bonzini Message-ID: <56DD7BD5.3030209@redhat.com> Date: Mon, 7 Mar 2016 14:02:13 +0100 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] qemu AIO worker threads change causes Guest OS hangup List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Huaicheng Li (coperd)" Cc: qemu-devel@nongnu.org, stefanha@redhat.com On 06/03/2016 03:42, Huaicheng Li (coperd) wrote: > >> On Mar 1, 2016, at 3:01 PM, Paolo Bonzini > > wrote: >> >> This is done >> because the worker threads only care about the queued request list, not >> about active or completed requests. > > Do you think it would be useful to add an API for inserting one request > back > to the queued list? For example, In case of request failure, we can > insert it > back to the list for re-handling according to some rule before returning > it directly > to guest os. Hi, this is usually handled at a higher level than the thread pool. See for example hw/block/virtio-blk.c's restart mechanism, which is enabled with the rerror=stop and werror=stop options of -drive. Paolo