[PATCH] D42885: [AMDGPU] intrintrics for byte/short load/store

[Tim Corringham via Phabricator via llvm-commits]

timcorringham added a comment.

Matt, we do actually need these intrinsics as we have an urgent requirement for them Open Vulkan (which is of course my motivation for implementing them).

As Tim commented, the load ubyte and load short instructions extend to 32 bits. While float is a little odd, it does match the behavior of the other buffer_load instructions. Also I think changing it would require a disproportionate amount of effort.


Repository:
  rL LLVM

https://reviews.llvm.org/D42885