2009-04-21 09:04:20 +00:00
|
|
|
From: Nolan Leake <nolan <at> sigbus.net>
|
|
|
|
Subject: [PATCH] Fix (at least one cause of) qcow2 corruption.
|
|
|
|
|
|
|
|
qcow2's get_cluster_offset() scans forward in the l2 table to find other
|
|
|
|
clusters that have the same allocation status as the first cluster.
|
|
|
|
This is used by (among others) qcow_is_allocated().
|
|
|
|
|
|
|
|
Unfortunately, it was not checking to be sure that it didn't fall off
|
|
|
|
the end of the l2 table. This patch adds that check.
|
|
|
|
|
|
|
|
The symptom that motivated me to look into this was that
|
|
|
|
bdrv_is_allocated() was returning false when there was in fact data
|
|
|
|
there. This is one of many ways this bug could lead to data corruption.
|
|
|
|
|
|
|
|
I checked the other place that scans for consecutive unallocated blocks
|
|
|
|
(alloc_cluster_offset()) and it appears to be OK:
|
|
|
|
nb_clusters = MIN(nb_clusters, s->l2_size - l2_index);
|
|
|
|
appears to prevent the same problem from occurring.
|
|
|
|
|
|
|
|
Signed-off-by: Nolan Leake <nolan <at> sigbus.net>
|
|
|
|
|
|
|
|
---
|
|
|
|
|
2009-04-20 13:52:28 +00:00
|
|
|
From: Kevin Wolf <kwolf@redhat.com>
|
2009-04-21 09:04:20 +00:00
|
|
|
Subject: [PATCH] qcow2 corruption: Fix alloc_cluster_link_l2
|
2009-04-20 13:52:28 +00:00
|
|
|
|
|
|
|
This patch fixes a qcow2 corruption bug introduced in SVN Rev 5861. L2 tables
|
|
|
|
are big endian, so entries must be converted before being passed to functions.
|
|
|
|
|
|
|
|
This bug is easy to trigger. The following script will create and destroy a
|
|
|
|
qcow2 image (the header is gone after three loop iterations):
|
|
|
|
|
|
|
|
#!/bin/bash
|
|
|
|
qemu-img create -f qcow2 test.qcow 1M
|
|
|
|
for i in $(seq 1 10); do
|
|
|
|
qemu-system-x86_64 -hda test.qcow -monitor stdio > /dev/null 2>&1 <<EOF
|
|
|
|
savevm test-$i
|
|
|
|
quit
|
|
|
|
EOF
|
|
|
|
done
|
|
|
|
|
|
|
|
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
|
|
|
|
|
|
|
|
diff -up qemu-kvm-0.10/qemu/block-qcow2.c.qcow2-corruption qemu-kvm-0.10/qemu/block-qcow2.c
|
2009-04-21 09:04:20 +00:00
|
|
|
diff -up qemu-kvm-0.10/qemu/block-qcow2.c.qcow2-corruption qemu-kvm-0.10/qemu/block-qcow2.c
|
|
|
|
--- qemu-kvm-0.10/qemu/block-qcow2.c.qcow2-corruption 2009-04-21 09:57:21.000000000 +0100
|
|
|
|
+++ qemu-kvm-0.10/qemu/block-qcow2.c 2009-04-21 09:58:27.000000000 +0100
|
|
|
|
@@ -670,6 +670,10 @@ static uint64_t get_cluster_offset(Block
|
|
|
|
|
|
|
|
nb_available = (nb_available >> 9) + index_in_cluster;
|
|
|
|
|
|
|
|
+ if (nb_needed > nb_available) {
|
|
|
|
+ nb_needed = nb_available;
|
|
|
|
+ }
|
|
|
|
+
|
|
|
|
cluster_offset = 0;
|
|
|
|
|
|
|
|
/* seek the the l2 offset in the l1 table */
|
|
|
|
@@ -912,7 +916,7 @@ static int alloc_cluster_link_l2(BlockDr
|
2009-04-20 13:52:28 +00:00
|
|
|
goto err;
|
|
|
|
|
|
|
|
for (i = 0; i < j; i++)
|
|
|
|
- free_any_clusters(bs, old_cluster[i], 1);
|
|
|
|
+ free_any_clusters(bs, be64_to_cpu(old_cluster[i]), 1);
|
|
|
|
|
|
|
|
ret = 0;
|
|
|
|
err:
|