On Tue, May 10, 2016 at 07:20:09 -0400, John Ferlan wrote:
On 04/29/2016 11:34 AM, John Ferlan wrote:
> As it turns out, trusting that <allocation> being 0 means it wasn't
> provided isn't such a good idea.
>
> If someone provided a <capacity> of 10 and <allocation> of 0, then
> we need to honor it.
>
> So this patch which I'll merge into the previous patch will track
> when the XML is read if the allocation was provided or not. That
> way we can determine in this code that if allocation = 0, then
> "overwrite" with the capacity value if it wasn't provided.
>
> Signed-off-by: John Ferlan <jferlan(a)redhat.com>
> ---
> src/conf/storage_conf.c | 1 +
> src/storage/storage_driver.c | 11 ++++++-----
> src/util/virstoragefile.c | 1 +
> src/util/virstoragefile.h | 2 ++
> 4 files changed, 10 insertions(+), 5 deletions(-)
>
ping - any thoughts regarding squashing this in as well?
I think it will be better if you repost the patch in the form you are
going to push it.
Peter