From: Anton Nefedov <anton.nefedov@virtuozzo.com>
To: qemu-devel@nongnu.org
Cc: den@virtuozzo.com, kwolf@redhat.com, mreitz@redhat.com,
eblake@redhat.com, "Denis V. Lunev" <den@openvz.org>,
Anton Nefedov <anton.nefedov@virtuozzo.com>
Subject: [Qemu-devel] [PATCH v2 04/15] qcow2: alloc space for COW in one chunk
Date: Thu, 1 Jun 2017 18:14:22 +0300 [thread overview]
Message-ID: <1496330073-51338-5-git-send-email-anton.nefedov@virtuozzo.com> (raw)
In-Reply-To: <1496330073-51338-1-git-send-email-anton.nefedov@virtuozzo.com>
From: "Denis V. Lunev" <den@openvz.org>
Currently each single write operation can result in 3 write operations
if guest offsets are not cluster aligned. One write is performed for the
real payload and two for COW-ed areas. Thus the data possibly lays
non-contiguously on the host filesystem. This will reduce further
sequential read performance significantly.
The patch allocates the space in the file with cluster granularity,
ensuring
1. better host offset locality
2. less space allocation operations
(which can be expensive on distributed storage)
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: Anton Nefedov <anton.nefedov@virtuozzo.com>
---
block/qcow2.c | 24 +++++++++++++++++++++++-
1 file changed, 23 insertions(+), 1 deletion(-)
diff --git a/block/qcow2.c b/block/qcow2.c
index b3ba5da..cd5efba 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -1575,6 +1575,24 @@ fail:
return ret;
}
+static void handle_alloc_space(BlockDriverState *bs, QCowL2Meta *l2meta)
+{
+ BDRVQcow2State *s = bs->opaque;
+ QCowL2Meta *m;
+
+ for (m = l2meta; m != NULL; m = m->next) {
+ uint64_t bytes = m->nb_clusters << s->cluster_bits;
+
+ if (m->cow_start.nb_bytes == 0 && m->cow_end.nb_bytes == 0) {
+ continue;
+ }
+
+ /* try to alloc host space in one chunk for better locality */
+ bdrv_co_pwrite_zeroes(bs->file, m->alloc_offset, bytes,
+ BDRV_REQ_ALLOCATE);
+ }
+}
+
static coroutine_fn int qcow2_co_pwritev(BlockDriverState *bs, uint64_t offset,
uint64_t bytes, QEMUIOVector *qiov,
int flags)
@@ -1656,8 +1674,12 @@ static coroutine_fn int qcow2_co_pwritev(BlockDriverState *bs, uint64_t offset,
if (ret < 0) {
goto fail;
}
-
qemu_co_mutex_unlock(&s->lock);
+
+ if (bs->file->bs->supported_zero_flags & BDRV_REQ_ALLOCATE) {
+ handle_alloc_space(bs, l2meta);
+ }
+
BLKDBG_EVENT(bs->file, BLKDBG_WRITE_AIO);
trace_qcow2_writev_data(qemu_coroutine_self(),
cluster_offset + offset_in_cluster);
--
2.7.4
next prev parent reply other threads:[~2017-06-01 15:15 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-01 15:14 [Qemu-devel] [PATCH v2 00/15] qcow2: space preallocation and COW improvements Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 01/15] block: introduce BDRV_REQ_ALLOCATE flag Anton Nefedov
2017-06-01 19:07 ` Eric Blake
2017-06-02 13:08 ` Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 02/15] file-posix: support BDRV_REQ_ALLOCATE Anton Nefedov
2017-06-01 19:49 ` Eric Blake
2017-06-01 19:54 ` Eric Blake
2017-06-02 14:34 ` Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 03/15] blkdebug: " Anton Nefedov
2017-06-01 19:50 ` Eric Blake
2017-06-02 13:13 ` Anton Nefedov
2017-06-01 15:14 ` Anton Nefedov [this message]
2017-06-05 14:28 ` [Qemu-devel] [PATCH v2 04/15] qcow2: alloc space for COW in one chunk Eric Blake
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 05/15] qcow2: do not COW the empty areas Anton Nefedov
2017-06-05 14:40 ` Eric Blake
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 06/15] qcow2: preallocation at image expand Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 07/15] qcow2: set inactive flag Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 08/15] qcow2: truncate preallocated space Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 09/15] qcow2: check space leak at the end of the image Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 10/15] qcow2: handle_prealloc(): find out if area zeroed by earlier preallocation Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 11/15] qcow2: fix misleading comment about L2 linking Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 12/15] qcow2-cluster: slightly refactor handle_dependencies() Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 13/15] qcow2-cluster: make handle_dependencies() logic easier to follow Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 14/15] qcow2: allow concurrent unaligned writes to the same clusters Anton Nefedov
2017-06-01 15:14 ` [Qemu-devel] [PATCH v2 15/15] iotest 046: test simultaneous cluster write error case Anton Nefedov
2017-06-01 18:41 ` [Qemu-devel] [PATCH v2 00/15] qcow2: space preallocation and COW improvements John Snow
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=1496330073-51338-5-git-send-email-anton.nefedov@virtuozzo.com \
--to=anton.nefedov@virtuozzo.com \
--cc=den@openvz.org \
--cc=den@virtuozzo.com \
--cc=eblake@redhat.com \
--cc=kwolf@redhat.com \
--cc=mreitz@redhat.com \
--cc=qemu-devel@nongnu.org \
/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).