请问有办法适配H20 么，我按照文档装完服务，总是报下面的异常，哪位大佬能帮忙解决一下，跪谢。

[devopscool]

2025-12-15 18:14:31.024221: I tensorflow/core/util/port.cc:110] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable TF_ENABLE_ONEDNN_OPTS=0.
2025-12-15 18:14:31.025673: I tensorflow/tsl/cuda/cudart_stub.cc:28] Could not find cuda drivers on your machine, GPU will not be used.
2025-12-15 18:14:31.054595: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI AVX512_BF16 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2025-12-15 18:14:31.425408: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT

skip_decoder: True

#########scale sphere:False, add_teeth:False

Render rgb: True

face_upsampled:(39904, 3), face_ori:torch.Size([9976, 3]), vertex_num_upsampled:20018, vertex_num_ori:5023

loading pretrained weight from: ./model_zoo/lam_models/releases/lam/lam-20k/step_045500/model.safetensors
finish loading pretrained weight from: ./model_zoo/lam_models/releases/lam/lam-20k/step_045500/model.safetensors

2025-12-15 18:14:38.605 | INFO | tools.flame_tracking_single_image:init:69 - Output Directory: output/tracking
2025-12-15 18:14:38.605 | INFO | tools.flame_tracking_single_image:init:72 - Loading Pre-trained Models...
2025-12-15 18:14:40.836 | INFO | tools.flame_tracking_single_image:init:119 - Finished Loading Pre-trained Models. Time: 2.23s
Running on local URL: http://0.0.0.0:7860

subdir_path and uid: raw
/tmp/tmpo05yucu8/raw.png ./assets/sample_motion/export/I_Am_Iron_Man/flame_param /tmp/tmpo05yucu8 /tmp/tmpo05yucu8/output.mp4
2025-12-15 18:14:53.203 | INFO | tools.flame_tracking_single_image:preprocess:144 - Starting Preprocessing...
2025-12-15 18:14:54.494 | INFO | tools.flame_tracking_single_image:preprocess:239 - Finished Processing Image. Time: 1.29s
2025-12-15 18:14:54.494 | INFO | tools.flame_tracking_single_image:optimize:247 - Starting Optimization...
Ignoring unknown cluster teeth.
Traceback (most recent call last):
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 2107, in _run_ninja_build
subprocess.run(
File "/root/miniconda3/envs/lam310/lib/python3.10/subprocess.py", line 526, in run
raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v', '-j', '4']' returned non-zero exit status 139.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/gradio/queueing.py", line 388, in call_prediction
output = await route_utils.call_process_api(
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/gradio/route_utils.py", line 219, in call_process_api
output = await app.get_blocks().process_api(
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/gradio/blocks.py", line 1437, in process_api
result = await self.call_function(
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/gradio/blocks.py", line 1109, in call_function
prediction = await anyio.to_thread.run_sync(
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/anyio/to_thread.py", line 61, in run_sync
return await get_async_backend().run_sync_in_worker_thread(
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/anyio/backends/asyncio.py", line 2525, in run_sync_in_worker_thread
return await future
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/anyio/backends/asyncio.py", line 986, in run
result = context.run(func, *args)
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/gradio/utils.py", line 641, in wrapper
response = f(*args, **kwargs)
File "/data/lam/LAM/app_lam.py", line 253, in core_fn
return_code = flametracking.optimize()
File "/data/lam/LAM/tools/flame_tracking_single_image.py", line 264, in optimize
tracker = GlobalTracker(config_data)
File "/data/lam/LAM/vhap/model/tracker.py", line 1290, in init
super().init(cfg)
File "/data/lam/LAM/vhap/model/tracker.py", line 68, in init
self.render = NVDiffRenderer(
File "/data/lam/LAM/vhap/util/render_nvdiffrast.py", line 72, in init
self.glctx = dr.RasterizeGLContext() if use_opengl else dr.RasterizeCudaContext()
File "/data/lam/nvdiffrast/nvdiffrast/torch/ops.py", line 177, in init
self.cpp_wrapper = get_plugin().RasterizeCRStateWrapper(cuda_device_idx)
File "/data/lam/nvdiffrast/nvdiffrast/torch/ops.py", line 118, in get_plugin
torch.utils.cpp_extension.load(name=plugin_name, sources=source_paths, extra_cflags=opts, extra_cuda_cflags=opts+['-lineinfo'], extra_ldflags=ldflags, with_cuda=True, verbose=False)
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 1309, in load
return jit_compile(
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 1719, in jit_compile
write_ninja_file_and_build_library(
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 1832, in write_ninja_file_and_build_library
run_ninja_build(
File "/root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 2123, in run_ninja_build
raise RuntimeError(message) from e
RuntimeError: Error building extension 'nvdiffrast_plugin': [1/5] /usr/bin/nvcc --generate-dependencies-with-compile --dependency-output rasterize.cuda.o.d -DTORCH_EXTENSION_NAME=nvdiffrast_plugin -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/TH -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/THC -isystem /root/miniconda3/envs/lam310/include/python3.10 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS -D__CUDA_NO_BFLOAT16_CONVERSIONS -D__CUDA_NO_HALF2_OPERATORS --expt-relaxed-constexpr -gencode=arch=compute_90,code=compute_90 -gencode=arch=compute_90,code=sm_90 --compiler-options '-fPIC' -DNVDR_TORCH -lineinfo -std=c++17 -c /data/lam/nvdiffrast/nvdiffrast/common/rasterize.cu -o rasterize.cuda.o
FAILED: [code=139] rasterize.cuda.o
/usr/bin/nvcc --generate-dependencies-with-compile --dependency-output rasterize.cuda.o.d -DTORCH_EXTENSION_NAME=nvdiffrast_plugin -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/TH -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/THC -isystem /root/miniconda3/envs/lam310/include/python3.10 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS -D__CUDA_NO_BFLOAT16_CONVERSIONS -D__CUDA_NO_HALF2_OPERATORS --expt-relaxed-constexpr -gencode=arch=compute_90,code=compute_90 -gencode=arch=compute_90,code=sm_90 --compiler-options '-fPIC' -DNVDR_TORCH -lineinfo -std=c++17 -c /data/lam/nvdiffrast/nvdiffrast/common/rasterize.cu -o rasterize.cuda.o
Segmentation fault (core dumped)
[2/5] /usr/bin/nvcc --generate-dependencies-with-compile --dependency-output interpolate.cuda.o.d -DTORCH_EXTENSION_NAME=nvdiffrast_plugin -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/TH -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/THC -isystem /root/miniconda3/envs/lam310/include/python3.10 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS -D__CUDA_NO_BFLOAT16_CONVERSIONS_ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_90,code=compute_90 -gencode=arch=compute_90,code=sm_90 --compiler-options '-fPIC' -DNVDR_TORCH -lineinfo -std=c++17 -c /data/lam/nvdiffrast/nvdiffrast/common/interpolate.cu -o interpolate.cuda.o
FAILED: [code=139] interpolate.cuda.o
/usr/bin/nvcc --generate-dependencies-with-compile --dependency-output interpolate.cuda.o.d -DTORCH_EXTENSION_NAME=nvdiffrast_plugin -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/TH -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/THC -isystem /root/miniconda3/envs/lam310/include/python3.10 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS_ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_90,code=compute_90 -gencode=arch=compute_90,code=sm_90 --compiler-options '-fPIC' -DNVDR_TORCH -lineinfo -std=c++17 -c /data/lam/nvdiffrast/nvdiffrast/common/interpolate.cu -o interpolate.cuda.o
Segmentation fault (core dumped)
[3/5] /usr/bin/nvcc --generate-dependencies-with-compile --dependency-output RasterImpl.cuda.o.d -DTORCH_EXTENSION_NAME=nvdiffrast_plugin -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/TH -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/THC -isystem /root/miniconda3/envs/lam310/include/python3.10 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS_ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_90,code=compute_90 -gencode=arch=compute_90,code=sm_90 --compiler-options '-fPIC' -DNVDR_TORCH -lineinfo -std=c++17 -c /data/lam/nvdiffrast/nvdiffrast/common/cudaraster/impl/RasterImpl.cu -o RasterImpl.cuda.o
FAILED: [code=139] RasterImpl.cuda.o
/usr/bin/nvcc --generate-dependencies-with-compile --dependency-output RasterImpl.cuda.o.d -DTORCH_EXTENSION_NAME=nvdiffrast_plugin -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/TH -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/THC -isystem /root/miniconda3/envs/lam310/include/python3.10 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS_ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_90,code=compute_90 -gencode=arch=compute_90,code=sm_90 --compiler-options '-fPIC' -DNVDR_TORCH -lineinfo -std=c++17 -c /data/lam/nvdiffrast/nvdiffrast/common/cudaraster/impl/RasterImpl.cu -o RasterImpl.cuda.o
Segmentation fault (core dumped)
[4/5] /usr/bin/nvcc --generate-dependencies-with-compile --dependency-output texture.cuda.o.d -DTORCH_EXTENSION_NAME=nvdiffrast_plugin -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/TH -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/THC -isystem /root/miniconda3/envs/lam310/include/python3.10 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS_ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_90,code=compute_90 -gencode=arch=compute_90,code=sm_90 --compiler-options '-fPIC' -DNVDR_TORCH -lineinfo -std=c++17 -c /data/lam/nvdiffrast/nvdiffrast/common/texture.cu -o texture.cuda.o
FAILED: [code=139] texture.cuda.o
/usr/bin/nvcc --generate-dependencies-with-compile --dependency-output texture.cuda.o.d -DTORCH_EXTENSION_NAME=nvdiffrast_plugin -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/TH -isystem /root/miniconda3/envs/lam310/lib/python3.10/site-packages/torch/include/THC -isystem /root/miniconda3/envs/lam310/include/python3.10 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS_ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_90,code=compute_90 -gencode=arch=compute_90,code=sm_90 --compiler-options '-fPIC' -DNVDR_TORCH -lineinfo -std=c++17 -c /data/lam/nvdiffrast/nvdiffrast/common/texture.cu -o texture.cuda.o
Segmentation fault (core dumped)
ninja: build stopped: subcommand failed.

首先是找不到gpu 改用了cpu，其次nvdiffrast 总是不能动态编译，我尝试装了各种驱动组合，torch 也用了2.4 版本，都不行，下面是我的配置：