The fdiv lowering is currently split between an IR pass and codegen,
so make sure this works end to end. We also currently differ from the
DAG on some edge cases, which this will show in a future change.
Details
Details
Diff Detail
Diff Detail
Event Timeline
Comment Actions
Seems obvious.
llvm/test/CodeGen/AMDGPU/GlobalISel/fdiv.f16.ll | ||
---|---|---|
54–74 | Would update_llc_test_checks common these up if you added a GFX89 prefix? Same for loads of other tests. |
llvm/test/CodeGen/AMDGPU/GlobalISel/fdiv.f16.ll | ||
---|---|---|
54–74 | Apparently it is (I wasn't smart enough to order the checks from most preferred to least) |
Would update_llc_test_checks common these up if you added a GFX89 prefix? Same for loads of other tests.