On Tue, Jan 22, 2019 at 10:26 PM Matt Turner <matts...@gmail.com> wrote:
> Was this just something that you noticed by inspection?
With the patch reverted I see some validation failures in
dEQP-VK.spirv_assembly.instruction.compute.8bit_storage.push_constant_8_to_16.scalar_uint
and friends.
mov(16) g10<4>B g[a0]<VxH,1,0>B { align1 1H };
ERROR: Destination stride must be equal to the ratio of the sizes of
the execution data type to the destination type
Wow. I don't think I've ever seen an instruction like that before. I
can definitely see how the validator might not be equipped to handle
it.
I'll think about it some more.
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
