trunk/d2e122ddd079e2386d88ca97681fd1d5a9c28060: Parametrize FP4 scaled_mm tests by recipe instead of platform. (#197740)
- PyTorch: 1462 events in the last 90 days
- PyTorch: 1443th Release in the last 90 days
- Previous: earlier the same day · trunk/db9987a77da7446854969c5980234079339de26d
What happened
test_blockwise_nvfp4_compile picked its recipe off torch.version.hip, so on ROCm it ran MXFP4 under an NVFP4 name; ScaledBlas.cpp rejects NVFP4 on ROCm, so the test never could have covered it. The same ternary in test_blockwise_mxfp8_nvfp4_error_messages made mxfp4 unselectable on CUDA, leaving its "recipe == mxfp4 and SM120OrLater" skip permanently dead and MXFP4 untested there. Both tests now parametrize the full…
Summary assembled by rule from the sources below