On 25/11/2020 11:36, Richard Biener wrote:
This makes sure to lower VECTOR_BOOLEAN_TYPE_P typed VEC_COND_EXPRs so we don't try to use vcond to expand those. That's especially improtant for x86 integer mode boolean vectors but eventually as well for aarch64 / gcn VnBImode ones.
GCN does not have vector-boolean modes. It uses a DImode mask for the VCC and EXEC registers, and doesn't ever store them as vectors.
I presume, therefore, that this patch ought to be safe. Andrew
