chenyu f798b60338 add METAL_FAST_MATH env var to disable metal fast math (#3369) ...

* env var METAL_FAST_MATH to disable fastmath for metal

use this to test impact of fast math. might need to disable compiler cache with DISABLE_COMPILER_CACHE

* failed onnx test with fast math

METAL_FAST_MATH=0 DISABLE_COMPILER_CACHE=1 NOOPT=1 python -m pytest -n=auto test/external/external_test_onnx_backend.py -k test_MaxPool3d_stride_padding_cpu

2024-02-11 04:26:09 -05:00

..

autogen

rebuild ocelot (#3259)

2024-01-26 18:46:36 -08:00

compiler

use comgr to compile (#3248)

2024-01-26 18:27:49 -08:00

graph

move autogen to runtime/autogen (#3254)

2024-01-26 12:44:19 -08:00

ops_clang.py

compiler support (#3260)

2024-01-26 23:36:40 -08:00

ops_cpu.py

clean up einsum_mulacc (#3312)

2024-02-04 06:21:19 -05:00

ops_cuda.py

compiler support (#3260)

2024-01-26 23:36:40 -08:00

ops_disk.py

add device name to device, all are constructed (#3221)

2024-01-23 20:34:56 -08:00

ops_gpu.py

compiler support (#3260)

2024-01-26 23:36:40 -08:00

ops_hip.py

full sync to fix HIP memory leak (#3364)

2024-02-10 11:50:27 +01:00

ops_llvm.py

compiler support (#3260)

2024-01-26 23:36:40 -08:00

ops_metal.py

add METAL_FAST_MATH env var to disable metal fast math (#3369)

2024-02-11 04:26:09 -05:00

ops_python.py

ops_python: add HIP tensor core mock and refactor METAL (#3354)

2024-02-09 12:46:06 +01:00

ops_torch.py

fix neg logical_not inconsistencies (#3222)

2024-01-24 11:48:40 -05:00