[Qemu-devel] [PATCH 1/1] qcow2: avoid extra flushes in qcow2

[Qemu-devel] [PATCH 1/1] qcow2: avoid extra flushes in qcow2

[Denis V. Lunev]

The problem with excessive flushing was found by a couple of performance
tests:
  - parallel directory tree creation (from 2 processes)
  - 32 cached writes + fsync at the end in a loop

For the first one results improved from 2.6 loops/sec to 3.5 loops/sec.
Each loop creates 10^3 directories with 10 files in each.

For the second one results improved from ~600 fsync/sec to ~1100
fsync/sec. Though, it was run on SSD so it probably won't show such
performance gain on rotational media.

qcow2_cache_flush() calls bdrv_flush() unconditionally after writing
cache entries of a particular cache. This can lead to as many as
2 additional fdatasyncs inside bdrv_flush.

We can simply skip all fdatasync calls inside qcow2_co_flush_to_os
as bdrv_flush for sure will do the job. These flushes are necessary to
keep the right order of writes to the different caches. Though this is
not necessary in the current code base as this ordering is ensured through
the flush in qcow2_cache_flush_dependency().

Signed-off-by: Denis V. Lunev <den@openvz.org>
CC: Pavel Borzenkov <pborzenkov@virtuozzo.com>
CC: Kevin Wolf <kwolf@redhat.com>
CC: Max Reitz <mreitz@redhat.com>
---
Changes from v1:
- renamed qcow2_cache_flush_nosync to qcow2_cache_write (looks better to me)
- rewritten commit message entirely

 block/qcow2-cache.c | 11 +++++++++--
 block/qcow2.c       |  4 ++--
 block/qcow2.h       |  1 +
 3 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/block/qcow2-cache.c b/block/qcow2-cache.c
index 0fe8eda..208a060 100644
--- a/block/qcow2-cache.c
+++ b/block/qcow2-cache.c
@@ -226,7 +226,7 @@ static int qcow2_cache_entry_flush(BlockDriverState *bs, Qcow2Cache *c, int i)
     return 0;
 }
 
-int qcow2_cache_flush(BlockDriverState *bs, Qcow2Cache *c)
+int qcow2_cache_write(BlockDriverState *bs, Qcow2Cache *c)
 {
     BDRVQcow2State *s = bs->opaque;
     int result = 0;
@@ -242,8 +242,15 @@ int qcow2_cache_flush(BlockDriverState *bs, Qcow2Cache *c)
         }
     }
 
+    return result;
+}
+
+int qcow2_cache_flush(BlockDriverState *bs, Qcow2Cache *c)
+{
+    int result = qcow2_cache_write(bs, c);
+
     if (result == 0) {
-        ret = bdrv_flush(bs->file->bs);
+        int ret = bdrv_flush(bs->file->bs);
         if (ret < 0) {
             result = ret;
         }
diff --git a/block/qcow2.c b/block/qcow2.c
index 38caa66..a194a8a 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -2837,14 +2837,14 @@ static coroutine_fn int qcow2_co_flush_to_os(BlockDriverState *bs)
     int ret;
 
     qemu_co_mutex_lock(&s->lock);
-    ret = qcow2_cache_flush(bs, s->l2_table_cache);
+    ret = qcow2_cache_write(bs, s->l2_table_cache);
     if (ret < 0) {
         qemu_co_mutex_unlock(&s->lock);
         return ret;
     }
 
     if (qcow2_need_accurate_refcounts(s)) {
-        ret = qcow2_cache_flush(bs, s->refcount_block_cache);
+        ret = qcow2_cache_write(bs, s->refcount_block_cache);
         if (ret < 0) {
             qemu_co_mutex_unlock(&s->lock);
             return ret;
diff --git a/block/qcow2.h b/block/qcow2.h
index a063a3c..7db9795 100644
--- a/block/qcow2.h
+++ b/block/qcow2.h
@@ -583,6 +583,7 @@ int qcow2_cache_destroy(BlockDriverState* bs, Qcow2Cache *c);
 void qcow2_cache_entry_mark_dirty(BlockDriverState *bs, Qcow2Cache *c,
      void *table);
 int qcow2_cache_flush(BlockDriverState *bs, Qcow2Cache *c);
+int qcow2_cache_write(BlockDriverState *bs, Qcow2Cache *c);
 int qcow2_cache_set_dependency(BlockDriverState *bs, Qcow2Cache *c,
     Qcow2Cache *dependency);
 void qcow2_cache_depends_on_flush(Qcow2Cache *c);
-- 
2.1.4

Re: [Qemu-devel] [PATCH v2 1/1] qcow2: avoid extra flushes in qcow2

[Denis V. Lunev]

On 06/02/2016 06:58 PM, Denis V. Lunev wrote:
> The problem with excessive flushing was found by a couple of performance
> tests:
>    - parallel directory tree creation (from 2 processes)
>    - 32 cached writes + fsync at the end in a loop
>
> For the first one results improved from 2.6 loops/sec to 3.5 loops/sec.
> Each loop creates 10^3 directories with 10 files in each.
>
> For the second one results improved from ~600 fsync/sec to ~1100
> fsync/sec. Though, it was run on SSD so it probably won't show such
> performance gain on rotational media.
>
> qcow2_cache_flush() calls bdrv_flush() unconditionally after writing
> cache entries of a particular cache. This can lead to as many as
> 2 additional fdatasyncs inside bdrv_flush.
>
> We can simply skip all fdatasync calls inside qcow2_co_flush_to_os
> as bdrv_flush for sure will do the job. These flushes are necessary to
> keep the right order of writes to the different caches. Though this is
> not necessary in the current code base as this ordering is ensured through
> the flush in qcow2_cache_flush_dependency().
>
> Signed-off-by: Denis V. Lunev <den@openvz.org>
> CC: Pavel Borzenkov <pborzenkov@virtuozzo.com>
> CC: Kevin Wolf <kwolf@redhat.com>
> CC: Max Reitz <mreitz@redhat.com>
actually this is v2 version of the patchset, missed version number in 
the subject

Re: [Qemu-devel] [PATCH 1/1] qcow2: avoid extra flushes in qcow2

[Kevin Wolf]

Am 02.06.2016 um 17:58 hat Denis V. Lunev geschrieben:
> The problem with excessive flushing was found by a couple of performance
> tests:
>   - parallel directory tree creation (from 2 processes)
>   - 32 cached writes + fsync at the end in a loop
> 
> For the first one results improved from 2.6 loops/sec to 3.5 loops/sec.
> Each loop creates 10^3 directories with 10 files in each.
> 
> For the second one results improved from ~600 fsync/sec to ~1100
> fsync/sec. Though, it was run on SSD so it probably won't show such
> performance gain on rotational media.
> 
> qcow2_cache_flush() calls bdrv_flush() unconditionally after writing
> cache entries of a particular cache. This can lead to as many as
> 2 additional fdatasyncs inside bdrv_flush.
> 
> We can simply skip all fdatasync calls inside qcow2_co_flush_to_os
> as bdrv_flush for sure will do the job. These flushes are necessary to
> keep the right order of writes to the different caches. Though this is
> not necessary in the current code base as this ordering is ensured through
> the flush in qcow2_cache_flush_dependency().
> 
> Signed-off-by: Denis V. Lunev <den@openvz.org>
> CC: Pavel Borzenkov <pborzenkov@virtuozzo.com>
> CC: Kevin Wolf <kwolf@redhat.com>
> CC: Max Reitz <mreitz@redhat.com>

Thanks, applied to the block branch.

Kevin