search

traveller59 / spconv

Spatial Sparse Convolution Library
Apache License 2.0
1.86k stars 363 forks source link

Spconv2.x.x from source on Jetson Xavier (cuda 10.2) #565

Open YoushaaMurhij opened 1 year ago

YoushaaMurhij commented 1 year ago

I follow the instructions in the readme:

ENV CUMM_CUDA_ARCH_LIST="7.2"
ENV CUMM_DISABLE_JIT=1
RUN git clone https://github.com/FindDefinition/cumm &&  cd ./cumm   && pip install -e .

RUN git clone https://github.com/traveller59/spconv && cd spconv && rm pyproject.toml setup.py
COPY pyproject.toml ./spconv/pyproject.toml 
COPY setup.py ./spconv/setup.py
RUN SPCONV_DISABLE_JIT="1"
RUN cd ./spconv && pip install -e .

And got this error:

[inference.py-1] [1/39] [GCC][c++/pch]/tmp/cumm/cumm/build/core_cc/include/csrc/arrayref/ArrayPtr.h.gch
[inference.py-1] /tmp/cumm/cumm/build/core_cc/include/csrc/arrayref/ArrayPtr.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [2/39] [GCC][c++/pch]/tmp/cumm/cumm/build/core_cc/include/tensorview_bind/TensorViewBind.h.gch
[inference.py-1] /tmp/cumm/cumm/build/core_cc/include/tensorview_bind/TensorViewBind.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [3/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/core_cc_pybind_main/core_cc_pybind_main.cc.o
[inference.py-1] [4/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/tensorview_bind/PyBindTensorViewBind/PyBindTensorViewBind_bind_TensorViewBind.cc.o
[inference.py-1] [5/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_get_length.cc.o
[inference.py-1] [6/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_ArrayPtr.cc.o
[inference.py-1] [7/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_copy.cc.o
[inference.py-1] [8/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr___repr__.cc.o
[inference.py-1] [9/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_shadow_copy.cc.o
[inference.py-1] [10/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/tensorview_bind/TensorViewBind/TensorViewBind_hello.cc.o
[inference.py-1] [11/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_clear.cc.o
[inference.py-1] [12/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/PyBindArrayPtr/PyBindArrayPtr_bind_ArrayPtr.cc.o
[inference.py-1] [13/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/tensorview_bind/TensorViewBind/TensorViewBind_run_nvrtc_conv_kernel.cc.o
[inference.py-1] [14/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/tensorview_bind/TensorViewBind/TensorViewBind_bind_nvrtc_params.cc.o
[inference.py-1] [15/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_get_byte_length.cc.o
[inference.py-1] [16/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/tensorview_bind/TensorViewBind/TensorViewBind_run_nvrtc_gemm_kernel.cc.o
[inference.py-1] [17/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/tensorview_bind/TensorViewBind/TensorViewBind_bind_gemm_algo_desp.cc.o
[inference.py-1] [18/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/tensorview_bind/TensorViewBind/TensorViewBind_bind_conv_algo_desp.cc.o
[inference.py-1] [19/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/tensorview_bind/TensorViewBind/TensorViewBind_bind_conv_params.cc.o
[inference.py-1] [20/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/tensorview_bind/TensorViewBind/TensorViewBind_bind_gemm_params.cc.o
[inference.py-1] [21/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/tensorview_bind/TensorViewBind/TensorViewBind_bind_enums.cc.o
[inference.py-1] [22/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_get_offset.cc.o
[inference.py-1] [23/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_get_access_offset.cc.o
[inference.py-1] [24/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_get_access_size.cc.o
[inference.py-1] [25/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr___add__.cc.o
[inference.py-1] [26/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr___iadd__.cc.o
[inference.py-1] [27/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_get_num_access.cc.o
[inference.py-1] [28/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr___sub__.cc.o
[inference.py-1] [29/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr___isub__.cc.o
[inference.py-1] [30/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_change_access_size.cc.o
[inference.py-1] [31/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_change_access_byte_size.cc.o
[inference.py-1] [32/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr___getitem__.cc.o
[inference.py-1] [33/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr___setitem__.cc.o
[inference.py-1] [34/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_get_data.cc.o
[inference.py-1] [35/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_get_meta_data.cc.o
[inference.py-1] [36/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_check_smem_bank_conflicit.cc.o
[inference.py-1] [37/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/csrc/arrayref/ArrayPtr/ArrayPtr_astype.cc.o
[inference.py-1] [38/39] [GCC][c++]/tmp/cumm/cumm/build/core_cc/src/tensorview_bind/TensorViewBind/TensorViewBind_bind_tensorview.cc.o
[inference.py-1] [39/39] [GCC][Link]/tmp/cumm/cumm/build/core_cc/core_cc.cpython-36m-aarch64-linux-gnu.so
[inference.py-1] [1/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/cumm/common/CompileInfo.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/cumm/common/CompileInfo.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m"  -O3 -std=c++14 -fPIC -x c++-header  -c /tmp/spconv/spconv/build/core_cc/include/cumm/common/CompileInfo.h -o /tmp/spconv/spconv/build/core_cc/include/cumm/common/CompileInfo.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/cumm/common/CompileInfo.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [2/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/ExternalSpconvMatmul.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/ExternalSpconvMatmul.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/ExternalSpconvMatmul.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/ExternalSpconvMatmul.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/ExternalSpconvMatmul.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [3/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ExternalAllocatorGuard.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ExternalAllocatorGuard.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ExternalAllocatorGuard.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ExternalAllocatorGuard.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ExternalAllocatorGuard.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [4/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ExternalAllocator.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ExternalAllocator.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ExternalAllocator.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ExternalAllocator.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ExternalAllocator.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [5/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/gather/GatherCPU.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/gather/GatherCPU.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/gather/GatherCPU.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/gather/GatherCPU.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/gather/GatherCPU.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [6/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/ConvTuneResult.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/ConvTuneResult.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/ConvTuneResult.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/ConvTuneResult.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/ConvTuneResult.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [7/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/GemmTuneResult.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/GemmTuneResult.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/GemmTuneResult.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/GemmTuneResult.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/GemmTuneResult.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [8/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/gemmops/GemmTunerSimple.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/gemmops/GemmTunerSimple.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/gemmops/GemmTunerSimple.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/gemmops/GemmTunerSimple.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/gemmops/GemmTunerSimple.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [9/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/cumm/gemm/main/GemmMainUnitTest.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/cumm/gemm/main/GemmMainUnitTest.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/cumm/gemm/main/GemmMainUnitTest.h -o /tmp/spconv/spconv/build/core_cc/include/cumm/gemm/main/GemmMainUnitTest.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/cumm/gemm/main/GemmMainUnitTest.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [10/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/cumm/conv/main/ConvMainUnitTest.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/cumm/conv/main/ConvMainUnitTest.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/cumm/conv/main/ConvMainUnitTest.h -o /tmp/spconv/spconv/build/core_cc/include/cumm/conv/main/ConvMainUnitTest.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/cumm/conv/main/ConvMainUnitTest.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [11/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/convops/ConvTunerSimple.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/convops/ConvTunerSimple.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/convops/ConvTunerSimple.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/convops/ConvTunerSimple.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/convops/ConvTunerSimple.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [12/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ThrustCustomAllocatorV2.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ThrustCustomAllocatorV2.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ThrustCustomAllocatorV2.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ThrustCustomAllocatorV2.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ThrustCustomAllocatorV2.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [13/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ThrustAllocator.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ThrustAllocator.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ThrustAllocator.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ThrustAllocator.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/alloc/ThrustAllocator.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [14/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/maxpool/IndiceMaxPoolCPU.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/maxpool/IndiceMaxPoolCPU.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/maxpool/IndiceMaxPoolCPU.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/maxpool/IndiceMaxPoolCPU.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/maxpool/IndiceMaxPoolCPU.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [15/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu1d/SparseConvIndicesCPU.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu1d/SparseConvIndicesCPU.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu1d/SparseConvIndicesCPU.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu1d/SparseConvIndicesCPU.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu1d/SparseConvIndicesCPU.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [16/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops1d/Point2Voxel.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops1d/Point2Voxel.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops1d/Point2Voxel.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops1d/Point2Voxel.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops1d/Point2Voxel.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [17/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu2d/Point2VoxelCPU.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu2d/Point2VoxelCPU.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu2d/Point2VoxelCPU.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu2d/Point2VoxelCPU.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu2d/Point2VoxelCPU.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [18/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu2d/SparseConvIndicesCPU.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu2d/SparseConvIndicesCPU.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu2d/SparseConvIndicesCPU.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu2d/SparseConvIndicesCPU.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu2d/SparseConvIndicesCPU.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [19/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu1d/Point2VoxelCPU.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu1d/Point2VoxelCPU.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu1d/Point2VoxelCPU.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu1d/Point2VoxelCPU.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu1d/Point2VoxelCPU.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [20/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops2d/Point2Voxel.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops2d/Point2Voxel.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops2d/Point2Voxel.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops2d/Point2Voxel.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops2d/Point2Voxel.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [21/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/spops/ConvGemmOps.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/spops/ConvGemmOps.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/spops/ConvGemmOps.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/spops/ConvGemmOps.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/spops/ConvGemmOps.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [22/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu3d/Point2VoxelCPU.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu3d/Point2VoxelCPU.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu3d/Point2VoxelCPU.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu3d/Point2VoxelCPU.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu3d/Point2VoxelCPU.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [23/810] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] FAILED: /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] In file included from /usr/local/cuda/include/cuda_runtime.h:83,
[inference.py-1]                  from <command-line>:
[inference.py-1] /usr/local/cuda/include/crt/host_config.h:138:2: error: #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]   138 | #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]       |  ^~~~~
[inference.py-1] [24/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu3d/SparseConvIndicesCPU.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu3d/SparseConvIndicesCPU.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu3d/SparseConvIndicesCPU.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu3d/SparseConvIndicesCPU.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu3d/SparseConvIndicesCPU.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [25/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops3d/Point2Voxel.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops3d/Point2Voxel.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops3d/Point2Voxel.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops3d/Point2Voxel.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops3d/Point2Voxel.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [26/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu4d/Point2VoxelCPU.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu4d/Point2VoxelCPU.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu4d/Point2VoxelCPU.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu4d/Point2VoxelCPU.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu4d/Point2VoxelCPU.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [27/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu4d/SparseConvIndicesCPU.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu4d/SparseConvIndicesCPU.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu4d/SparseConvIndicesCPU.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu4d/SparseConvIndicesCPU.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops_cpu4d/SparseConvIndicesCPU.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [28/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/utils/boxops/BoxOps.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/utils/boxops/BoxOps.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/utils/boxops/BoxOps.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/utils/boxops/BoxOps.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/utils/boxops/BoxOps.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [29/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops4d/Point2Voxel.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops4d/Point2Voxel.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops4d/Point2Voxel.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops4d/Point2Voxel.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/ops4d/Point2Voxel.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [30/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/SimpleExternalSpconvMatmul.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/SimpleExternalSpconvMatmul.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/SimpleExternalSpconvMatmul.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/SimpleExternalSpconvMatmul.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/convops/SimpleExternalSpconvMatmul.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [31/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/utils/pcc/PointCloudCompress.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/utils/pcc/PointCloudCompress.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/utils/pcc/PointCloudCompress.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/utils/pcc/PointCloudCompress.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/utils/pcc/PointCloudCompress.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] [32/810] g++ -MMD -MT /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/SpconvOps.h.gch -MF /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/SpconvOps.h.gch.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" -O3 -std=c++14 -fPIC -x c++-header -DTV_CUDA -c /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/SpconvOps.h -o /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/SpconvOps.h.gch
[inference.py-1] /tmp/spconv/spconv/build/core_cc/include/csrc/sparse/all/SpconvOps.h:1:9: warning: #pragma once in main file
[inference.py-1]     1 | #pragma once
[inference.py-1]       |         ^~~~
[inference.py-1] ninja: build stopped: subcommand failed.
[inference.py-1] [1/779] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] FAILED: /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] In file included from /usr/local/cuda/include/cuda_runtime.h:83,
[inference.py-1]                  from <command-line>:
[inference.py-1] /usr/local/cuda/include/crt/host_config.h:138:2: error: #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]   138 | #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]       |  ^~~~~
[inference.py-1] [2/779] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] FAILED: /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n512k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] In file included from /usr/local/cuda/include/cuda_runtime.h:83,
[inference.py-1]                  from <command-line>:
[inference.py-1] /usr/local/cuda/include/crt/host_config.h:138:2: error: #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]   138 | #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]       |  ^~~~~
[inference.py-1] [3/779] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] FAILED: /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] In file included from /usr/local/cuda/include/cuda_runtime.h:83,
[inference.py-1]                  from <command-line>:
[inference.py-1] /usr/local/cuda/include/crt/host_config.h:138:2: error: #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]   138 | #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]       |  ^~~~~
[inference.py-1] [4/779] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] FAILED: /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] In file included from /usr/local/cuda/include/cuda_runtime.h:83,
[inference.py-1]                  from <command-line>:
[inference.py-1] /usr/local/cuda/include/crt/host_config.h:138:2: error: #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]   138 | #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]       |  ^~~~~
[inference.py-1] [5/779] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] FAILED: /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n256k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] In file included from /usr/local/cuda/include/cuda_runtime.h:83,
[inference.py-1]                  from <command-line>:
[inference.py-1] /usr/local/cuda/include/crt/host_config.h:138:2: error: #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]   138 | #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]       |  ^~~~~
[inference.py-1] [6/779] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] FAILED: /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] In file included from /usr/local/cuda/include/cuda_runtime.h:83,
[inference.py-1]                  from <command-line>:
[inference.py-1] /usr/local/cuda/include/crt/host_config.h:138:2: error: #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]   138 | #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]       |  ^~~~~
[inference.py-1] [7/779] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] FAILED: /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m64n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] In file included from /usr/local/cuda/include/cuda_runtime.h:83,
[inference.py-1]                  from <command-line>:
[inference.py-1] /usr/local/cuda/include/crt/host_config.h:138:2: error: #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]   138 | #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]       |  ^~~~~
[inference.py-1] [8/779] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] FAILED: /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m128n128k8m32n64k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] In file included from /usr/local/cuda/include/cuda_runtime.h:83,
[inference.py-1]                  from <command-line>:
[inference.py-1] /usr/local/cuda/include/crt/host_config.h:138:2: error: #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]   138 | #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]       |  ^~~~~
[inference.py-1] [9/779] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] FAILED: /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32tnt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] In file included from /usr/local/cuda/include/cuda_runtime.h:83,
[inference.py-1]                  from <command-line>:
[inference.py-1] /usr/local/cuda/include/crt/host_config.h:138:2: error: #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]   138 | #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]       |  ^~~~~
[inference.py-1] [10/779] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] FAILED: /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] nvcc -MMD -MT /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o -MF /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o.d -I "/tmp/spconv/spconv/build/core_cc/include" -I "/usr/local/lib/python3.6/dist-packages/pybind11/include" -I "/usr/include/python3.6m" -I "/usr/local/cuda/include" -I "/tmp/cumm/include" --expt-relaxed-constexpr -std=c++14 -O3 -Xcompiler="-fPIC" -Xcompiler='-O3' -DTV_CUDA -gencode=arch=compute_72,code=sm_72 -Xcudafe "--diag_suppress=implicit_return_from_non_void_function" -c /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu -o /tmp/spconv/spconv/build/core_cc/src/cumm/gemm/main/Simt_f32f32f32f32f32ttt_m32n128k16m32n32k8A1_200_S1/GemmKernel/GemmKernel_gemm_kernel.cu.o
[inference.py-1] In file included from /usr/local/cuda/include/cuda_runtime.h:83,
[inference.py-1]                  from <command-line>:
[inference.py-1] /usr/local/cuda/include/crt/host_config.h:138:2: error: #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]   138 | #error -- unsupported GNU version! gcc versions later than 8 are not supported!
[inference.py-1]       |  ^~~~~
[inference.py-1] ninja: build stopped: subcommand failed.
[inference.py-1] Traceback (most recent call last):
[inference.py-1]     import spconv.pytorch as spconv
[inference.py-1]   File "/tmp/spconv/spconv/__init__.py", line 15, in <module>
[inference.py-1]     from . import build as _build
[inference.py-1]   File "/tmp/spconv/spconv/build.py", line 74, in <module>
[inference.py-1]     verbose=True)
[inference.py-1]   File "/usr/local/lib/python3.6/dist-packages/pccm/builder/pybind.py", line 137, in build_pybind
[inference.py-1]     for p, v in path_to_meta.items()})
[inference.py-1]   File "/usr/local/lib/python3.6/dist-packages/ccimport/core.py", line 171, in ccimport
[inference.py-1]     linker=linker)
[inference.py-1]   File "/usr/local/lib/python3.6/dist-packages/ccimport/buildtools/writer.py", line 1100, in build_simple_ninja
[inference.py-1]     raise subprocess.CalledProcessError(proc.returncode, cmds)
[inference.py-1] subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.
[inference.py-1]
[inference.py-1] During handling of the above exception, another exception occurred:
[inference.py-1]
[inference.py-1] Traceback (most recent call last):
[inference.py-1]     import spconv
[inference.py-1]   File "/tmp/spconv/spconv/__init__.py", line 15, in <module>
[inference.py-1]     from . import build as _build
[inference.py-1]   File "/tmp/spconv/spconv/build.py", line 74, in <module>
[inference.py-1]     verbose=True)
[inference.py-1]   File "/usr/local/lib/python3.6/dist-packages/pccm/builder/pybind.py", line 137, in build_pybind
[inference.py-1]     for p, v in path_to_meta.items()})
[inference.py-1]   File "/usr/local/lib/python3.6/dist-packages/ccimport/core.py", line 171, in ccimport
[inference.py-1]     linker=linker)
[inference.py-1]   File "/usr/local/lib/python3.6/dist-packages/ccimport/buildtools/writer.py", line 1100, in build_simple_ninja
[inference.py-1]     raise subprocess.CalledProcessError(proc.returncode, cmds)
[inference.py-1] subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.
[ERROR] [inference.py-1]: process has died [pid 1751, exit code 1, cmd

Any suggestion? Thanks

YoushaaMurhij commented 1 year ago

Follow up: @traveller59, I managed to build spconv v2.x.x (master) with gcc < 8 and cuda 10.2 on Xavier from source following your instructions. but I noticed that the initialization takes a long time. I have an inference code and I just replaced spconv 1.2.1 with spconv v2.x.x (master) and now it takes too long to start and once it starts, everything is ok. It seems that it builds something additional while invoking spconv library. Could you please explain to me what to do to avoid such a long-time initialization so, I can add it to the environment setup before launching the script? Thanks

Alexmsit commented 1 year ago

Hey @YoushaaMurhij ,

Any updates on this? Were you able to solve this issue?

YoushaaMurhij commented 1 year ago

Hey! Unfortunately, no. I am still looking for updates on this.