https://github.com/arsenm commented:
You'd probably have better luck with 16-bit element vectors for AMDGPU (particularly cases that aren't even aligned). Or testing on another target https://github.com/llvm/llvm-project/pull/200935 _______________________________________________ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
