On Tue, Oct 20, 2015 at 09:36:29 -0400, John Ferlan wrote: > > > On 10/16/2015 08:11 AM, Peter Krempa wrote: > > Later on, we will need to know whether the user specified the target > > node or not. Turn the variable into a signed value. We already treat it > > as signed in various parts of the qemu driver. > > --- > > src/conf/domain_conf.c | 10 ++++++---- > > src/conf/domain_conf.h | 2 +- > > 2 files changed, 7 insertions(+), 5 deletions(-) > > > > Seeing domaincommon.rng change in patch 8 got me to thinking - should it > change here? That is from unsignedInt to integer. Especially since > you're checking the return value determined < 0.
Not really. This change shouldn't change the existing semantics of the field. It only modifies the data type but does not change what is parsed from the user. Mixing in the domaincommon.rng change would actually mix up the non-change part with the semantic change. Also the rng change merely allows that this field is omitted. Also patch 8 still will be required since the semantics need to be kept until the last patch that adds the checks. Peter
signature.asc
Description: Digital signature
-- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
