| Message ID | 1375775095-19456-2-git-send-email-famz@redhat.com |
|---|---|
| State | New |
| Headers | show |
Am 06.08.2013 um 09:44 hat Fam Zheng geschrieben: > A inner scope "ret" variable hides the real return code, it will always > return VMDK_OK for bs->backing_hd. Fix this by removing the declaration. > > Signed-off-by: Fam Zheng <famz@redhat.com> In the original code there is no problem. > --- > block/vmdk.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/block/vmdk.c b/block/vmdk.c > index 3756333..f42657b 100644 > --- a/block/vmdk.c > +++ b/block/vmdk.c > @@ -813,8 +813,8 @@ static int get_whole_cluster(BlockDriverState *bs, > /* we will be here if it's first write on non-exist grain(cluster). > * try to read from parent image, if exist */ > if (bs->backing_hd) { > - int ret; > - > + whole_grain = > + qemu_blockalign(bs, extent->cluster_sectors << BDRV_SECTOR_BITS); > if (!vmdk_is_cid_valid(bs)) { > return VMDK_ERROR; > } After this patch, there are two: block/vmdk.c: In function 'get_whole_cluster': block/vmdk.c:816:21: error: incompatible types when assigning to type 'uint8_t[(sizetype)(extent->cluster_sectors * 512u)]' from type 'void *' block/vmdk.c:824:9: error: 'ret' undeclared (first use in this function) Kevin
On Tue, Aug 06, 2013 at 02:15:13PM +0200, Kevin Wolf wrote: > Am 06.08.2013 um 09:44 hat Fam Zheng geschrieben: > > A inner scope "ret" variable hides the real return code, it will always > > return VMDK_OK for bs->backing_hd. Fix this by removing the declaration. > > > > Signed-off-by: Fam Zheng <famz@redhat.com> > > In the original code there is no problem. > Right, this whole patch should be squashed into patch 9. Looks like a bad rebase of the patches. > > --- > > block/vmdk.c | 4 ++-- > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > diff --git a/block/vmdk.c b/block/vmdk.c > > index 3756333..f42657b 100644 > > --- a/block/vmdk.c > > +++ b/block/vmdk.c > > @@ -813,8 +813,8 @@ static int get_whole_cluster(BlockDriverState *bs, > > /* we will be here if it's first write on non-exist grain(cluster). > > * try to read from parent image, if exist */ > > if (bs->backing_hd) { > > - int ret; > > - > > + whole_grain = > > + qemu_blockalign(bs, extent->cluster_sectors << BDRV_SECTOR_BITS); ^^^^^^^^^^^^^^^^^^ This hunk is from patch 9. > > if (!vmdk_is_cid_valid(bs)) { > > return VMDK_ERROR; > > } > > After this patch, there are two: > > block/vmdk.c: In function 'get_whole_cluster': > block/vmdk.c:816:21: error: incompatible types when assigning to type > 'uint8_t[(sizetype)(extent->cluster_sectors * 512u)]' from type 'void *' > block/vmdk.c:824:9: error: 'ret' undeclared (first use in this function) > > Kevin
diff --git a/block/vmdk.c b/block/vmdk.c index 3756333..f42657b 100644 --- a/block/vmdk.c +++ b/block/vmdk.c @@ -813,8 +813,8 @@ static int get_whole_cluster(BlockDriverState *bs, /* we will be here if it's first write on non-exist grain(cluster). * try to read from parent image, if exist */ if (bs->backing_hd) { - int ret; - + whole_grain = + qemu_blockalign(bs, extent->cluster_sectors << BDRV_SECTOR_BITS); if (!vmdk_is_cid_valid(bs)) { return VMDK_ERROR; }
A inner scope "ret" variable hides the real return code, it will always return VMDK_OK for bs->backing_hd. Fix this by removing the declaration. Signed-off-by: Fam Zheng <famz@redhat.com> --- block/vmdk.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)