qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: kwolf@redhat.com, pl@kamp.de, stefanha@redhat.com
Subject: [Qemu-devel] [PATCH v2 06/20] block: make bdrv_co_do_write_zeroes stricter in producing aligned requests
Date: Tue, 19 Nov 2013 18:07:29 +0100	[thread overview]
Message-ID: <1384880863-10434-7-git-send-email-pbonzini@redhat.com> (raw)
In-Reply-To: <1384880863-10434-1-git-send-email-pbonzini@redhat.com>

Right now, bdrv_co_do_write_zeroes will only try to align the
beginning of the request.  However, it is simpler for many
formats to expect the block layer to separate both the head *and*
the tail.  This makes sure that the format's bdrv_co_write_zeroes
function will be called with aligned sector_num and nb_sectors for
the bulk of the request.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 block.c | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/block.c b/block.c
index 4897649..567d669 100644
--- a/block.c
+++ b/block.c
@@ -2761,14 +2761,19 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
     while (nb_sectors > 0 && !ret) {
         int num = nb_sectors;
 
-        /* align request */
-        if (bs->bl.write_zeroes_alignment &&
-            num >= bs->bl.write_zeroes_alignment &&
-            sector_num % bs->bl.write_zeroes_alignment) {
-            if (num > bs->bl.write_zeroes_alignment) {
+        /* Align request.  Block drivers can expect the "bulk" of the request
+         * to be aligned.
+         */
+        if (bs->bl.write_zeroes_alignment
+            && num > bs->bl.write_zeroes_alignment) {
+            if (sector_num % bs->bl.write_zeroes_alignment != 0) {
+                /* Make a small request up to the first aligned sector.  */
                 num = bs->bl.write_zeroes_alignment;
+                num -= sector_num % bs->bl.write_zeroes_alignment;
+            } else if (num >= bs->bl.write_zeroes_alignment) {
+                /* Shorten the request to the last aligned sector.  */
+                num -= (sector_num + num) % bs->bl.write_zeroes_alignment;
             }
-            num -= sector_num % bs->bl.write_zeroes_alignment;
         }
 
         /* limit request size */
@@ -2785,24 +2790,27 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
         if (ret == -ENOTSUP) {
             /* Fall back to bounce buffer if write zeroes is unsupported */
             iov.iov_len = num * BDRV_SECTOR_SIZE;
             if (iov.iov_base == NULL) {
-                /* allocate bounce buffer only once and ensure that it
-                 * is big enough for this and all future requests.
-                 */
-                size_t bufsize = num <= nb_sectors ? num : max_write_zeroes;
-                iov.iov_base = qemu_blockalign(bs, bufsize * BDRV_SECTOR_SIZE);
-                memset(iov.iov_base, 0, bufsize * BDRV_SECTOR_SIZE);
+                iov.iov_base = qemu_blockalign(bs, num * BDRV_SECTOR_SIZE);
+                memset(iov.iov_base, 0, num * BDRV_SECTOR_SIZE);
             }
             qemu_iovec_init_external(&qiov, &iov, 1);
 
             ret = drv->bdrv_co_writev(bs, sector_num, num, &qiov);
+            if (num <= max_write_zeroes) {
+                /* Allocate bounce buffer only once if it is
+                 * big enough for this and all future requests.
+                 */
+                qemu_vfree(iov.iov_base);
+                iov.iov_base = NULL;
+            }
         }
 
         sector_num += num;
         nb_sectors -= num;
     }
 
     qemu_vfree(iov.iov_base);
     return ret;
 }
 
-- 
1.8.4.2

  parent reply	other threads:[~2013-11-19 17:08 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-11-19 17:07 [Qemu-devel] [PATCH v2 00/20] block & scsi: write_zeroes support through the whole stack Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 01/20] block: generalize BlockLimits handling to cover bdrv_aio_discard too Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 02/20] block: add flags to BlockRequest Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 03/20] block: add flags argument to bdrv_co_write_zeroes tracepoint Paolo Bonzini
2013-11-20  9:59   ` Stefan Hajnoczi
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 04/20] block: add bdrv_aio_write_zeroes Paolo Bonzini
2013-11-20 10:02   ` Stefan Hajnoczi
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 05/20] block: handle ENOTSUP from discard in generic code Paolo Bonzini
2013-11-19 17:07 ` Paolo Bonzini [this message]
2013-11-20 10:22   ` [Qemu-devel] [PATCH v2 06/20] block: make bdrv_co_do_write_zeroes stricter in producing aligned requests Stefan Hajnoczi
2013-11-20 11:01     ` Paolo Bonzini
2013-11-20 14:29       ` Stefan Hajnoczi
2013-11-21 11:30   ` Peter Lieven
2013-11-21 11:37     ` Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 07/20] vpc, vhdx: add get_info Paolo Bonzini
2013-11-20 12:39   ` Stefan Hajnoczi
2013-11-20 12:50     ` Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 08/20] block drivers: add discard/write_zeroes properties to bdrv_get_info implementation Paolo Bonzini
2013-11-21 11:33   ` Peter Lieven
2013-11-21 11:39     ` Paolo Bonzini
2013-11-21 11:48       ` Peter Lieven
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 09/20] block drivers: expose requirement for write same alignment from formats Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 10/20] block/iscsi: remove .bdrv_has_zero_init Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 11/20] block/iscsi: updated copyright Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 12/20] block/iscsi: check WRITE SAME support differently depending on MAY_UNMAP Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 13/20] block/iscsi: use UNMAP to write zeroes if LBPRZ=1 Paolo Bonzini
2013-11-21 11:43   ` Peter Lieven
2013-11-21 11:49     ` Paolo Bonzini
2013-11-21 11:54       ` Peter Lieven
2013-11-21 12:05         ` Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 14/20] raw-posix: implement write_zeroes with MAY_UNMAP for files Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 15/20] raw-posix: implement write_zeroes with MAY_UNMAP for block devices Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 16/20] raw-posix: add support for write_zeroes on XFS and " Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 17/20] qemu-iotests: 033 is fast Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 18/20] scsi-disk: catch write protection errors in UNMAP Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 19/20] scsi-disk: reject ANCHOR=1 for UNMAP and WRITE SAME commands Paolo Bonzini
2013-11-19 17:07 ` [Qemu-devel] [PATCH v2 20/20] scsi-disk: correctly implement WRITE SAME Paolo Bonzini
2013-11-19 17:23   ` ronnie sahlberg
2013-11-19 17:27     ` ronnie sahlberg
2013-11-19 17:31     ` Paolo Bonzini
2013-11-20 14:18   ` Stefan Hajnoczi
2013-11-20 14:19     ` Paolo Bonzini

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1384880863-10434-7-git-send-email-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=pl@kamp.de \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).