On 05/23/2017 06:22 AM, Alberto Garcia wrote: > qcow2_encrypt_sectors() does not need an Error parameter, and we're > not checking its value anyway, so we can safely remove it.
Misleading. You are NOT removing the Error parameter from qcow2_encrypt_sectors(), but rather are explicitly ignoring any errors by passing NULL. I'd update the commit message to something like: We are relying on the return value of qcow2_encrypt_sectors() to flag problems, but have no way to report that error to the end user. Since we are just throwing away the error, we can pass NULL instead for simpler code. A more robust solution would figure out how to pass the original error (rather than a new message related to our -EIO return) back to the caller, but that is more invasive. > > Signed-off-by: Alberto Garcia <be...@igalia.com> > --- > block/qcow2-cluster.c | 4 +--- > 1 file changed, 1 insertion(+), 3 deletions(-) > With a better commit message, Reviewed-by: Eric Blake <ebl...@redhat.com> -- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3266 Virtualization: qemu.org | libvirt.org
signature.asc
Description: OpenPGP digital signature