[Bug] mmcv-full 1.7.2 with H20GPU and torch 2.1.X Focal loss error

[Jenny0420]

Prerequisite

• I have searched Issues and Discussions but cannot get the expected help.
• The bug has not been fixed in the latest version(https://github.com/open-mmlab/mmcv).

Environment

/usr/local/lib/python3.10/dist-packages/mmcv/init.py:20: UserWarning: On January 1, 2023, MMCV will release v2.0.0, in which it will remove components related to the training process and add a data transformation module. In addition, it will rename the package names mmcv to mmcv-lite and mmcv-full to mmcv. See https://github.com/open-mmlab/mmcv/blob/master/docs/en/compatibility.md for more details.
warnings.warn(
{'sys.platform': 'linux', 'Python': '3.10.12 (main, Nov 6 2024, 20:22:13) [GCC 11.4.0]', 'CUDA available': True, 'GPU 0,1,2,3,4,5,6,7': 'NVIDIA H20', 'CUDA_HOME': '/usr/local/cuda', 'NVCC': 'Cuda compilation tools, release 12.1, V12.1.105', 'GCC': 'x86_64-linux-gnu-gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0', 'PyTorch': '2.1.2+cu121', 'PyTorch compiling details': 'PyTorch built with:\n - GCC 9.3\n - C++ Version: 201703\n - Intel(R) oneAPI Math Kernel Library Version 2022.2-Product Build 20220804 for Intel(R) 64 architecture applications\n - Intel(R) MKL-DNN v3.1.1 (Git Hash 64f6bcbcbab628e96f33a62c3e975f8535a7bde4)\n - OpenMP 201511 (a.k.a. OpenMP 4.5)\n - LAPACK is enabled (usually provided by MKL)\n - NNPACK is enabled\n - CPU capability usage: AVX512\n - CUDA Runtime 12.1\n - NVCC architecture flags: -gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_90,code=sm_90\n - CuDNN 8.9.2\n - Magma 2.6.1\n - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=12.1, CUDNN_VERSION=8.9.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -D_GLIBCXX_USE_CXX11_ABI=0 -fabi-version=11 -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOROCTRACER -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Werror=bool-operation -Wnarrowing -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=old-style-cast -Wno-invalid-partial-specialization -Wno-unused-private-field -Wno-aligned-allocation-unavailable -Wno-missing-braces -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_DISABLE_GPU_ASSERTS=ON, TORCH_VERSION=2.1.2, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=1, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, \n', 'TorchVision': '0.16.2+cu121', 'OpenCV': '4.8.1', 'MMCV': '1.7.2', 'MMCV Compiler': 'GCC 9.3', 'MMCV CUDA Compiler': '12.1'}

Reproduces the problem - code sample

the problem is on loss feedback progress.

Reproduces the problem - command or script

run the Sparse4D code with only one GPU

Reproduces the problem - error message

"
File "/usr/local/lib/python3.10/dist-packages/mmdet/models/losses/focal_loss.py", line 233, in forward
loss_cls = self.loss_weight * calculate_loss_func(
File "/usr/local/lib/python3.10/dist-packages/mmdet/models/losses/focal_loss.py", line 139, in sigmoid_focal_loss
loss = _sigmoid_focal_loss(pred.contiguous(), target.contiguous(), gamma,
File "/usr/local/lib/python3.10/dist-packages/torch/autograd/function.py", line 539, in apply
return super().apply(*args, **kwargs) # type: ignore[misc]
File "/usr/local/lib/python3.10/dist-packages/mmcv/ops/focal_loss.py", line 59, in forward
ext_module.sigmoid_focal_loss_forward(
RuntimeError: CUDA error: no kernel image is available for execution on the device
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

"

Additional information

i tried use another focal loss replace mmcv/ops/focal loss , and it is work.

[furh20]

想问下8卡的H20训练效率怎么样，相较于A100性能怎么样