On 9/19/19 7:56 AM, Richard W.M. Jones wrote: > b_conn_handle fields exportsize and can_* have a special meaning when > they are -1. It means that the value of the field has not been > computed and cached yet. > > The struct was being initialized using memset (_, -1, _) which does > indeed have the effect of setting all the fields to -1, although it's > somewhat non-obvious to say the least. > > This commit replaces it with ordinary field initialization, also > setting h->handle to NULL. By keeping the initialization function > next to the struct definition, hopefully they will be updated in > tandem in future. > > GCC 9.2.1 actually optimizes this back into the memset equivalent > using inline AVX instructions, so good job there! > > Simple refactoring, should have no effect on how the code works. > > See commit d60d0f4248610fc1d116dc9f249526d20913c9a3.
ACK. And affects 3 and 4 of my pending spec cleanup patches, but yours should go in first (I'll rebase mine, because I'm working on refactoring those for less code duplication anyway). -- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3226 Virtualization: qemu.org | libvirt.org
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Libguestfs mailing list Libguestfs@redhat.com https://www.redhat.com/mailman/listinfo/libguestfs
