diff options
| author | ravil-mobile <ravil.aviva.com@gmail.com> | 2025-10-02 16:14:19 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-10-02 10:14:19 -0400 |
| commit | b92ff6b2099fa5103a0350028c6752ad953b9b8d (patch) | |
| tree | e16948b5caafcb66febf69dba45eeb08af4b39e9 /llvm/lib/CodeGen/StackFrameLayoutAnalysisPass.cpp | |
| parent | b147019f8b11cd491f331bd707f764786792665e (diff) | |
| download | llvm-b92ff6b2099fa5103a0350028c6752ad953b9b8d.zip llvm-b92ff6b2099fa5103a0350028c6752ad953b9b8d.tar.gz llvm-b92ff6b2099fa5103a0350028c6752ad953b9b8d.tar.bz2 | |
[ROCDL] Added `rocdl.cvt.scale.pk8` ops (#161411)
This patch introduces some missing FP conversion instructions in the
ROCDL dialect
Specifically:
- Downscaling 8x packed F16, Bf16, Fp32 values to Fp8, Bf8, Fp4
Tests:
- Added lit-tests to check MLIR -> LLVM lowering
Diffstat (limited to 'llvm/lib/CodeGen/StackFrameLayoutAnalysisPass.cpp')
0 files changed, 0 insertions, 0 deletions