Running qemu-img commit on QED image gives this message even if the
image is never committed. This patch fixes this. Do we need to do any
L1, L2 table cleanup here ? I see that cleanup code is commented out in
the case of qcow2 (qcow2_make_empty).
before patching -
# qemu-img commit -f qed ss.qed
qemu-img: Image is already committed
even before committing the image.
after patching -
# qemu-img commit -f qed ss.qed
Image committed.
and image is actually successfully committed.
-------------------------------------------------------
Signed-off-by: Onkar N Mahajan/India/IBM
<onkar.n.maha...@linux.vnet.ibm.com>
---
block/qed.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/block/qed.c b/block/qed.c
index 6c182ca..32f1d53 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -695,7 +695,7 @@ static int coroutine_fn
bdrv_qed_co_is_allocated(BlockDriverState *bs,
static int bdrv_qed_make_empty(BlockDriverState *bs)
{
- return -ENOTSUP;
+ return 0;
}
static BDRVQEDState *acb_to_s(QEDAIOCB *acb)
--
1.7.1
--
Onkar N Mahajan
System Software Engineer
IBM Linux Technology Center
Bangalore - India
