[PATCH] D124387: AMDGPU: Fold out readfirstlane between vgpr to vgpr copies

[Jay Foad via Phabricator via llvm-commits]

foad added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/SIFoldOperands.cpp:1858
+        //
+        // => %2 = COPY %0
+        //
----------------
This transformation only makes sense if you know that %0 is uniform. I think @nhaehnle has suggested introducing a "readanylane" pseudo and/or intrinsic for that kind of use case.

I'm not sure if there is any existing code that deliberately uses readfirstlane on a non-uniform argument, but if there is then this will break it.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D124387/new/

https://reviews.llvm.org/D124387