From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=57580 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1Q50YY-000242-43 for qemu-devel@nongnu.org; Wed, 30 Mar 2011 14:55:50 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Q50YW-0004qp-Pw for qemu-devel@nongnu.org; Wed, 30 Mar 2011 14:55:49 -0400 Received: from verein.lst.de ([213.95.11.211]:59486 helo=newverein.lst.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Q50YW-0004qc-LE for qemu-devel@nongnu.org; Wed, 30 Mar 2011 14:55:48 -0400 Date: Wed, 30 Mar 2011 20:55:46 +0200 From: Christoph Hellwig Subject: Re: [Qemu-devel] virtio-blk.c handling of i/o which is not a 512 multiple Message-ID: <20110330185546.GA26589@lst.de> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Conor Murphy Cc: qemu-devel@nongnu.org On Wed, Mar 30, 2011 at 08:48:18AM +0000, Conor Murphy wrote: > Is there a requirement for virtio-blk guest drivers that all i/o requests are > sized in multiples of 512 bytes? Yes, like for any other block driver. Of course this should not actually crash qemu, but rather fail the request. Does the patch below give you a correct error report? Index: qemu/hw/virtio-blk.c =================================================================== --- qemu.orig/hw/virtio-blk.c 2011-03-30 20:46:10.268665534 +0200 +++ qemu/hw/virtio-blk.c 2011-03-30 20:49:45.655247322 +0200 @@ -290,6 +290,10 @@ static void virtio_blk_handle_write(Virt virtio_blk_rw_complete(req, -EIO); return; } + if (req->qiov.size % req->dev->conf->logical_block_size) { + virtio_blk_rw_complete(req, -EIO); + return; + } if (mrb->num_writes == 32) { virtio_submit_multiwrite(req->dev->bs, mrb); @@ -317,6 +321,10 @@ static void virtio_blk_handle_read(VirtI virtio_blk_rw_complete(req, -EIO); return; } + if (req->qiov.size % req->dev->conf->logical_block_size) { + virtio_blk_rw_complete(req, -EIO); + return; + } acb = bdrv_aio_readv(req->dev->bs, sector, &req->qiov, req->qiov.size / BDRV_SECTOR_SIZE,