metakernel v1: device dossier, throttle-rejected bench, ULP compare, fuzz, sweep, stamps
0c6aadc verified | [general] | |
| name = "metakernel" | |
| version = 1 | |
| edition = 5 | |
| license = "Apache-2.0" | |
| backends = [ | |
| "cuda", | |
| ] | |
| [general.hub] | |
| repo-id = "phanerozoic/metakernel" | |
| [torch] | |
| src = [ | |
| "torch-ext/torch_binding.cpp", | |
| "torch-ext/torch_binding.h", | |
| ] | |
| [kernel.metakernel] | |
| backend = "cuda" | |
| cuda-capabilities = ["8.0", "8.6", "8.9", "9.0", "10.0", "12.0"] | |
| depends = ["torch"] | |
| src = [ | |
| "metakernel_cuda/probes.cu", | |
| "metakernel_cuda/probes_launch.h", | |
| "metakernel_cuda/stamps.cuh", | |
| ] | |