On 05/11/2017 01:22 PM, Markus Armbruster wrote: > Eric Blake <ebl...@redhat.com> writes: > >> On 05/11/2017 07:22 AM, Peter Xu wrote: >>> It's not very safe to assert in size_to_str(). Let's be inclusive. >>> >>> It naturally allows negative values. Now it won't even limit on the >>> size, as long as double would allow. >>> >>> Signed-off-by: Peter Xu <pet...@redhat.com> >>> --- >>> util/cutils.c | 9 +++++++-- >>> 1 file changed, 7 insertions(+), 2 deletions(-) >> >> Since you just introduced this in 2/4, I'd rather see the two squashed >> together if we decide this is the interface we want to support. > > PATCH 2 factors it out of print_type_size(). Doing just that and > improving only on top makes some sense.
Okay. (Moral of the story: I should have done more than just glance at 2 before commenting on 4) -- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3266 Virtualization: qemu.org | libvirt.org
signature.asc
Description: OpenPGP digital signature
