diff options
| author | ravil-mobile <ravil.aviva.com@gmail.com> | 2025-10-07 17:11:36 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-10-07 11:11:36 -0400 |
| commit | 3f6240775501945fd36f7b66fb23de0648b07f90 (patch) | |
| tree | 9a0cb619fa5aa31658504494f187d7ca75ad6e7a /llvm/test/tools/llvm-offload-binary/llvm-offload-binary.ll | |
| parent | b256d0a7aa00079e7ff0e64d52b8055ed6440682 (diff) | |
| download | llvm-3f6240775501945fd36f7b66fb23de0648b07f90.zip llvm-3f6240775501945fd36f7b66fb23de0648b07f90.tar.gz llvm-3f6240775501945fd36f7b66fb23de0648b07f90.tar.bz2 | |
[ROCDL] Added rocdl.cvt.scale.sr.pk8 ops (#162244)
This patch introduces some missing FP conversion instructions in the
ROCDL dialect for the GFX1250 arch.
Specifically:
Downscaling 8x packed F16, Bf16, Fp32 values to Fp8, Bf8, Fp4 with
stochastic rounding
Tests:
Added lit-tests to check MLIR -> LLVM lowering
Diffstat (limited to 'llvm/test/tools/llvm-offload-binary/llvm-offload-binary.ll')
0 files changed, 0 insertions, 0 deletions