These will be widened in the DAG. In the meanwhile early
widening prevents otherwise possible vectorization of
such loads.
Details
Details
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Paths
| Differential D77835
[AMDGPU] Disable sub-dword scralar loads IR widening ClosedPublic Authored by rampitec on Apr 9 2020, 3:31 PM.
Details Summary These will be widened in the DAG. In the meanwhile early
Diff Detail
Event TimelineHerald added subscribers: kerbowa, hiraditya, t-tye and 6 others. · View Herald TranscriptApr 9 2020, 3:31 PM This revision is now accepted and ready to land.Apr 9 2020, 6:59 PM Closed by commit rG44920e856652: [AMDGPU] Disable sub-dword scralar loads IR widening (authored by rampitec). · Explain WhyApr 10 2020, 8:28 AM This revision was automatically updated to reflect the committed changes.
Revision Contents
Diff 256577 llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp
llvm/test/CodeGen/AMDGPU/vectorize-loads.ll
llvm/test/CodeGen/AMDGPU/widen_extending_scalar_loads.ll
|