Re: [libvirt] [PATCH 03/30] storagefile: qcow1: Fix check for empty backing file
On 10/9/19 9:26 PM, Daniel Henrique Barboza wrote:
On 10/7/19 6:49 PM, Cole Robinson wrote:
> >From f772b3d91fd the intention of this code seems to be to set
Is this leading '>' a bogus?
That's a feature of git-send-email so that this line is not mistaken for
e-mail header line "From: Michal Privoznik ....". When Cole pushes these
patches, it won't appear in git log, nor it shows in his local branch.
> format=NONE when the image does not have a backing file. However
> 'buf' here is the whole qcow1 file header. What we want to be
> checking is 'res' which is the parsed backing file path.
> qcowXGetBackingStore sets this to NULL when there's no backing file.
>
> Signed-off-by: Cole Robinson <crobinso(a)redhat.com>
> ---
This is unusual. Either the qcow file header 'buf' is being set to \0 when
there is no backing file (which would make the current code work even
when checking the wrong thing) or this is a bug that flew under the radar
from a long time.
Since this is for qcow1 I wouldn't be surprised if it is the latter one.
I mean, does anybody even use qcow1 these days?
Michal