NvFuserScheduler_Matmul/nvfuser_nosplitk_timmshapes_TT/M:8/N:128/K:8/warps:4/stages:3/manual_time fails #1746Description
root@54f6edae576c:/opt/pytorch/nvfuser# bin/nvfuser_bench --benchmark_filter='NvFuserScheduler_Matmul/nvfuser_nosplitk_timmshapes_TT/M:8/N:128/K:8/warps:4/stages:3/manual_time'
...
Running bin/nvfuser_bench
...
terminate called after throwing an instance of 'nvfuser::nvfError'
what(): Fusion returns wrong results! The result tensor has shape [..., 8,128]. Mismatch happens at region result[...,0:8,0:128]
Exception raised from checkMatch at /opt/pytorch/nvfuser/benchmark/matmul.cpp:106 (most recent call first):
frame #0: nvfuser::nvfCheckFail(char const*, char const*, unsigned int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) + 0xf3 (0x7fa8587b1e0d in /opt/pytorch/nvfuser/build/libnvfuser_codegen.so)
frame #1: <unknown function> + 0xcac0c (0x5636afd43c0c in bin/nvfuser_bench)
frame #2: <unknown function> + 0xcf12c (0x5636afd4812c in bin/nvfuser_bench)
frame #3: <unknown function> + 0x14f6f8 (0x5636afdc86f8 in bin/nvfuser_bench)
frame #4: <unknown function> + 0x12e87b (0x5636afda787b in bin/nvfuser_bench)
frame #5: <unknown function> + 0x130485 (0x5636afda9485 in bin/nvfuser_bench)
frame #6: <unknown function> + 0x130aa9 (0x5636afda9aa9 in bin/nvfuser_bench)
frame #7: <unknown function> + 0x11f3fb (0x5636afd983fb in bin/nvfuser_bench)
frame #8: <unknown function> + 0x12183e (0x5636afd9a83e in bin/nvfuser_bench)
frame #9: <unknown function> + 0x121ade (0x5636afd9aade in bin/nvfuser_bench)
frame #10: <unknown function> + 0x5db84 (0x5636afcd6b84 in bin/nvfuser_bench)
frame #11: <unknown function> + 0x29d90 (0x7fa82235ad90 in /usr/lib/x86_64-linux-gnu/libc.so.6)
frame #12: __libc_start_main + 0x80 (0x7fa82235ae40 in /usr/lib/x86_64-linux-gnu/libc.so.6)
frame #13: <unknown function> + 0x7a555 (0x5636afcf3555 in bin/nvfuser_bench)
Aborted (core dumped)
environment: build-20240209
GPU: H100, A100
Metadata
Metadata
Assignees
Labels
No labels