On 13.11.2015 16:32, Lucian Petrut wrote: > > At the moment, qemu-img extends new image virtual sizes based > on the CHS algorithm provided by the VHD specs in order to > ensure that the disk geometry (and payload as seen by some > guests which use the CHS value) can fit in the requested disk. > > This patch drops this behavior, as it breaks compatibility with > Azure, which requires the MB alignment to be preserved. > > Signed-off-by: Lucian Petrut <lpet...@cloudbasesolutions.com> > --- > Proposed fix for https://bugs.launchpad.net/qemu/+bug/1490611 > > block/vpc.c | 32 ++++++++------------------------ > 1 file changed, 8 insertions(+), 24 deletions(-) >
CC-ing Peter. Looks right, considering page 7 of the spec: (http://download.microsoft.com/download/f/f/e/ffef50a5-07dd-4cf8-aaa3-442c0673a029/Virtual%20Hard%20Disk%20Format%20Spec_10_18_06.doc) "When the user creates a hard disk of a certain size, the size of the hard disk image in the virtual machine is smaller than that created by the user. This is because CHS value calculated from the hard disk size is rounded down." However, this patch looks incomplete, note the comment you are removing: "This ensures that qemu-img convert doesn't truncate images". This is because vpc_open() prefers the CHS size over the current_size value, and as far as I remember there was indeed a good reason for that (some application had invalid values for current_size, or something like that). So at the very least, we will have to change vpc_open() to always use the value from footer->current_size and ignore footer->{cyls,heads,secs_per_cyl}, otherwise qemu will always consider these images smaller than they actually are (which is bad, at least when converting images). But I'd like to know from Peter whether he remembers the reason why vpc_open() tries to ignore footer->current_size in the first place... Max PS: If possible, the "From: " header in an emailed patch should match the Signed-off-by line. I think. It doesn't state that in http://wiki.qemu.org/Contribute/SubmitAPatch, but the commit looks strange otherwise. So I guess if it isn't reasonably possible, it's not too bad. :-)
signature.asc
Description: OpenPGP digital signature
