From d3695811a19bef754e77bf0bf290aa1e98c5d039 Mon Sep 17 00:00:00 2001 From: Kevin Wolf Date: Wed, 12 Aug 2015 15:22:01 +0200 Subject: [PATCH 16/16] qcow2: Flush pending discards before allocating cluster Message-id: <1439392921-2301-2-git-send-email-kwolf@redhat.com> Patchwork-id: 67521 O-Subject: [RHEV-7.2 qemu-kvm-rhev PATCH 1/1] qcow2: Flush pending discards before allocating cluster Bugzilla: 1226297 RH-Acked-by: Stefan Hajnoczi RH-Acked-by: Laurent Vivier RH-Acked-by: Max Reitz Before a freed cluster can be reused, pending discards for this cluster must be processed. The original assumption was that this was not a problem because discards are only cached during discard/write zeroes operations, which are synchronous so that no concurrent write requests can cause cluster allocations. However, the discard/write zeroes operation itself can allocate a new L2 table (and it has to in order to put zero flags there), so make sure we can cope with the situation. This fixes https://bugs.launchpad.net/bugs/1349972. Cc: qemu-stable@nongnu.org Signed-off-by: Kevin Wolf Reviewed-by: Max Reitz (cherry picked from commit ecbda7a22576591a84f44de1be0150faf6001f1c) Signed-off-by: Kevin Wolf Signed-off-by: Miroslav Rezanina --- block/qcow2-refcount.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c index 6cbae1d..63c0085 100644 --- a/block/qcow2-refcount.c +++ b/block/qcow2-refcount.c @@ -833,6 +833,11 @@ static int64_t alloc_clusters_noref(BlockDriverState *bs, uint64_t size) uint64_t i, nb_clusters, refcount; int ret; + /* We can't allocate clusters if they may still be queued for discard. */ + if (s->cache_discards) { + qcow2_process_discards(bs, 0); + } + nb_clusters = size_to_clusters(s, size); retry: for(i = 0; i < nb_clusters; i++) { -- 1.8.3.1