From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:43071) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Tb8QO-0004OX-0M for qemu-devel@nongnu.org; Wed, 21 Nov 2012 06:25:05 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Tb8QM-0007Jf-TK for qemu-devel@nongnu.org; Wed, 21 Nov 2012 06:24:59 -0500 Received: from e35.co.us.ibm.com ([32.97.110.153]:59573) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Tb8QM-0007JW-MQ for qemu-devel@nongnu.org; Wed, 21 Nov 2012 06:24:58 -0500 Received: from /spool/local by e35.co.us.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Wed, 21 Nov 2012 04:24:56 -0700 Received: from d03relay02.boulder.ibm.com (d03relay02.boulder.ibm.com [9.17.195.227]) by d03dlp02.boulder.ibm.com (Postfix) with ESMTP id AA9D03E40045 for ; Wed, 21 Nov 2012 04:24:51 -0700 (MST) Received: from d03av05.boulder.ibm.com (d03av05.boulder.ibm.com [9.17.195.85]) by d03relay02.boulder.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id qALBOrD9300674 for ; Wed, 21 Nov 2012 04:24:53 -0700 Received: from d03av05.boulder.ibm.com (loopback [127.0.0.1]) by d03av05.boulder.ibm.com (8.14.4/8.13.1/NCO v10.0 AVout) with ESMTP id qALBOqST002364 for ; Wed, 21 Nov 2012 04:24:53 -0700 Message-ID: <1353497089.10808.8.camel@br8hfpp0.de.ibm.com> From: Heinz Graalfs Date: Wed, 21 Nov 2012 12:24:49 +0100 In-Reply-To: <50ACA634.8040002@de.ibm.com> References: <1353488287-47077-1-git-send-email-borntraeger@de.ibm.com> <50AC9B96.9070908@redhat.com> <50ACA634.8040002@de.ibm.com> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Mime-Version: 1.0 Subject: Re: [Qemu-devel] [PATCH/RFC] block: Ensure that block size constraints are considered List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Christian Borntraeger Cc: Kevin Wolf , jfrei@linux.vnet.ibm.com, qemu-devel@nongnu.org, Stefan Hajnoczi , agraf@suse.de On Wed, 2012-11-21 at 11:00 +0100, Christian Borntraeger wrote: > On 21/11/12 10:15, Kevin Wolf wrote: > > Am 21.11.2012 09:58, schrieb Christian Borntraeger: > >> From: Heinz Graalfs > >> > >> While testing IPL code (booting) for s390x we faced some problems > >> with cache=none on dasds (4k block size) on bdrv_preads with length > >> values != block size. > >> > >> This patch makes sure that bdrv_pread and friends work fine with > >> unaligned access even with cache=none > >> - propagate alignment value also into bs->file struct > >> - modify the size in case of no cache to avoid EINVAL on > >> pread() etc. (file was opened with O_DIRECT). > >> > >> This patch seems to cure the problems. > >> > >> CC: Kevin Wolf > >> CC: Stefan Hajnoczi > >> Signed-off-by: Heinz Graalfs > >> Signed-off-by: Christian Borntraeger > >> --- > >> block.c | 3 +++ > >> block/raw-posix.c | 6 ++++++ > >> 2 files changed, 9 insertions(+) > >> > >> diff --git a/block.c b/block.c > >> index 854ebd6..f23c562 100644 > >> --- a/block.c > >> +++ b/block.c > >> @@ -4242,6 +4242,9 @@ BlockDriverAIOCB *bdrv_aio_ioctl(BlockDriverState *bs, > >> void bdrv_set_buffer_alignment(BlockDriverState *bs, int align) > >> { > >> bs->buffer_alignment = align; > >> + if ((bs->open_flags & BDRV_O_NOCACHE)) { > >> + bs->file->buffer_alignment = align; > >> + } > > > > Any reason to restrict this to BDRV_O_NOCACHE? > > > > There have been patches to change the BDRV_O_NOCACHE flag from the > > monitor, in which case bdrv_set_buffer_alignment() wouldn't be called > > anew and O_DIRECT requests start to fail again. > > > Right, should be ok to remove the check. > > > > > >> } > >> > >> void *qemu_blockalign(BlockDriverState *bs, size_t size) > >> diff --git a/block/raw-posix.c b/block/raw-posix.c > >> index f2f0404..baebf1d 100644 > >> --- a/block/raw-posix.c > >> +++ b/block/raw-posix.c > >> @@ -700,6 +700,12 @@ static BlockDriverAIOCB *paio_submit(BlockDriverState *bs, int fd, > >> acb->aio_nbytes = nb_sectors * 512; > >> acb->aio_offset = sector_num * 512; > >> > >> + /* O_DIRECT also requires an aligned length */ > >> + if (bs->open_flags & BDRV_O_NOCACHE) { > >> + acb->aio_nbytes += acb->bs->buffer_alignment - 1; > >> + acb->aio_nbytes &= ~(acb->bs->buffer_alignment - 1); > >> + } > > > > Modifying aio_nbytes, but not the iov looks wrong to me. This may work > > in the handle_aiocb_rw_linear() code path, but not with actual vectored I/O. > > I think it seemed to work because the vectored I/O cases that we were testing were properly > aligned or were in the QEMU_AIO_MISALIGNED case which does bounce buffering anyway. > But I am not sure... > > Heinz can you have a look at this and identify the exact place were it was failing > and why this patch helps? (I just know it does). That might help to understand > if we also need to touch the iovs. The pread() call in handle_aiocb_rw_linear() is failing with errno 22. At least for this path the patch ensures that the length is correctly set. I need to look into the vectored I/O part in more detail. > Christian > >