[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Qemu-devel] [PATCH for 2.7 v2 1/1] qcow2: improve qcow2_co_write_zeroes

From: Denis V. Lunev
Subject: [Qemu-devel] [PATCH for 2.7 v2 1/1] qcow2: improve qcow2_co_write_zeroes()
Date: Wed, 27 Apr 2016 10:08:45 +0300

There is a possibility that qcow2_co_write_zeroes() will be called by
with the partial block. This could be synthetically triggered with
    qemu-io -c "write -z 32k 4k"
and can happen in the real life in qemu-nbd. The latter happens under
the following conditions:
    (1) qemu-nbd is started with --detect-zeroes=on and is connected to the
        kernel NBD client
    (2) third party program opens kernel NBD device with O_DIRECT
    (3) third party program performs write operation with memory buffer
        not aligned to the page
In this case qcow2_co_write_zeroes() is unable to perform the operation
and mark entire cluster as zeroed and returns ENOTSUP. Thus the caller
switches to non-optimized version and writes real zeroes to the disk.

The patch creates a shortcut. If the block is full of zeroes at the moment,
the request is extended to cover full block. User-visible situation with
this block is not changed. Before the patch the block is filled in the
image with real zeroes. After that patch the block is marked as zeroed
in metadata. Thus any subsequent changes in backing store chain are not

Kewin, thank you for a cool suggestion.

Signed-off-by: Denis V. Lunev <address@hidden>
CC: Kevin Wolf <address@hidden>
CC: Max Reitz <address@hidden>
Changes from v1:
- description rewritten completely
- new approach suggested by Kevin is implemented

 block/qcow2.c | 24 +++++++++++++++++++-----
 1 file changed, 19 insertions(+), 5 deletions(-)

diff --git a/block/qcow2.c b/block/qcow2.c
index 470734b..405d1da 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -2417,15 +2417,29 @@ static coroutine_fn int 
qcow2_co_write_zeroes(BlockDriverState *bs,
     int ret;
     BDRVQcow2State *s = bs->opaque;
-    /* Emulate misaligned zero writes */
-    if (sector_num % s->cluster_sectors || nb_sectors % s->cluster_sectors) {
-        return -ENOTSUP;
+    int head = sector_num % s->cluster_sectors;
+    int tail = nb_sectors % s->cluster_sectors;
+    if (head != 0 || tail != 0) {
+        int nr;
+        BlockDriverState *file;
+        sector_num -= head;
+        nb_sectors += head + tail;
+        /* check the the request extended to the entire cluster is read
+           as all zeroes at the moment. If so we can mark entire cluster
+           as zeroed in the metadata */
+        ret = bdrv_get_block_status_above(bs, NULL, sector_num, nb_sectors,
+                                          &nr, &file);
+        if (ret < 0 || !(ret & BDRV_BLOCK_ZERO) || sector_num == nr) {
+            /* Emulate misaligned zero writes */
+            return -ENOTSUP;
+        }
     /* Whatever is left can use real zero clusters */
-    ret = qcow2_zero_clusters(bs, sector_num << BDRV_SECTOR_BITS,
-        nb_sectors);
+    ret = qcow2_zero_clusters(bs, sector_num << BDRV_SECTOR_BITS, nb_sectors);
     return ret;

reply via email to

[Prev in Thread] Current Thread [Next in Thread]