Received: from [140.186.70.92] (port=55335 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1Ov8N6-0000Fb-Iq for qemu-devel@nongnu.org; Mon, 13 Sep 2010 08:43:00 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69) (envelope-from ) id 1Ov8N5-0004MS-55 for qemu-devel@nongnu.org; Mon, 13 Sep 2010 08:42:56 -0400 Received: from mx1.redhat.com ([209.132.183.28]:10765) by eggs.gnu.org with esmtp (Exim 4.69) (envelope-from ) id 1Ov8N4-0004ML-Ri for qemu-devel@nongnu.org; Mon, 13 Sep 2010 08:42:55 -0400 From: Kevin Wolf Date: Mon, 13 Sep 2010 14:42:48 +0200 Message-Id: <1284381771-7333-11-git-send-email-kwolf@redhat.com> In-Reply-To: <1284381771-7333-1-git-send-email-kwolf@redhat.com> References: <1284381771-7333-1-git-send-email-kwolf@redhat.com> Subject: [Qemu-devel] [STABLE 0.13][PATCH 10/13] qcow2: Remove unnecessary flush after L2 write List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: anthony@codemonkey.ws Cc: kwolf@redhat.com, qemu-devel@nongnu.org When a new cluster was allocated, we only need a flush after the write to the L2 table if it was a COW and we need to decrease the refcounts of the old clusters. Signed-off-by: Kevin Wolf (cherry picked from commit 7ec5e6a4ca43494949465f9f9f3d9e4c7c620503) --- block/qcow2-cluster.c | 16 ++++++++++++---- 1 files changed, 12 insertions(+), 4 deletions(-) diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c index 166922f..f562b16 100644 --- a/block/qcow2-cluster.c +++ b/block/qcow2-cluster.c @@ -655,7 +655,7 @@ static int write_l2_entries(BlockDriverState *bs, uint64_t *l2_table, int ret; BLKDBG_EVENT(bs->file, BLKDBG_L2_UPDATE); - ret = bdrv_pwrite_sync(bs->file, l2_offset + start_offset, + ret = bdrv_pwrite(bs->file, l2_offset + start_offset, &l2_table[l2_start_index], len); if (ret < 0) { return ret; @@ -718,9 +718,17 @@ int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m) goto err; } - for (i = 0; i < j; i++) - qcow2_free_any_clusters(bs, - be64_to_cpu(old_cluster[i]) & ~QCOW_OFLAG_COPIED, 1); + /* + * If this was a COW, we need to decrease the refcount of the old cluster. + * Also flush bs->file to get the right order for L2 and refcount update. + */ + if (j != 0) { + bdrv_flush(bs->file); + for (i = 0; i < j; i++) { + qcow2_free_any_clusters(bs, + be64_to_cpu(old_cluster[i]) & ~QCOW_OFLAG_COPIED, 1); + } + } ret = 0; err: -- 1.7.2.2