Files
clang-p2996/llvm/test/CodeGen/AMDGPU/vector_shuffle.packed.ll
David Majnemer be36812fb7 [TargetLowering] Be more efficient in fp -> bf16 NaN conversions
We can avoid masking completely as it is OK (and probably preferable) to
bring over some of the existant NaN payload.
2024-02-21 22:47:27 +00:00

203 KiB