Build pytorch failed by cuda/GcdLcmKernel.cu

SETUP ENV:
ubuntu 16.04
cuda 11.2
python 3.9

ERROR log:
[3485/5257] Generating include/renameavx.h
Generating renameavx.h: mkrename cinz_ 4 8 avx
[3487/5257] Generating include/renamefma4.h
Generating renamefma4.h: mkrename finz_ 4 8 fma4
[3488/5257] Generating include/renameavx2128.h
Generating renameavx2128.h: mkrename finz_ 2 4 avx2128
[3489/5257] Generating include/renamecuda.h
Generating renamecuda.h: mkrename finz_ 1 1 cuda
[3497/5257] Generating …/…/…/include/sleef.h
Generating sleef.h: mkrename cinz_ 2 4 __m128d __m128 __m128i _m128i SSE2
Generating sleef.h: mkrename cinz
2 4 __m128d __m128 __m128i _m128i SSE2 sse2
Generating sleef.h: mkrename cinz
2 4 __m128d __m128 __m128i _m128i SSE2 sse4
Generating sleef.h: mkrename cinz
4 8 __m256d __m256 __m128i struct\ {\ _m128i\ x,\ y;\ } AVX
Generating sleef.h: mkrename cinz
4 8 __m256d __m256 __m128i struct\ {\ _m128i\ x,\ y;\ } AVX avx
Generating sleef.h: mkrename finz
4 8 __m256d __m256 __m128i struct\ {\ _m128i\ x,\ y;\ } AVX fma4
Generating sleef.h: mkrename finz
4 8 __m256d __m256 __m128i _m256i AVX avx2
Generating sleef.h: mkrename finz
2 4 __m128d m128 m128i m128i SSE2 avx2128
Generating sleef.h: mkrename finz
8 16 m512d m512 m256i m512i AVX512F
Generating sleef.h: mkrename finz
8 16 m512d m512 m256i m512i AVX512F avx512f
Generating sleef.h: mkrename cinz
8 16 m512d m512 m256i m512i AVX512F avx512fnofma
Generating sleef.h: mkrename cinz
1 1 double float int32_t int32_t STDC purec
Generating sleef.h: mkrename finz
1 1 double float int32_t int32_t FP_FAST_FMA purecfma
[4007/5257] Building C object caffe2/CMakeFiles/torch_cpu.dir/
/third_party/miniz-2.0.8/miniz.c.o
…/third_party/miniz-2.0.8/miniz.c:3108:9: note: #pragma message: Using fopen, ftello, fseeko, stat() etc. path for file I/O - this path may not support large files.
#pragma message(“Using fopen, ftello, fseeko, stat() etc. path for file I/O - this path may not support large files.”)
^
[4561/5257] Building CUDA object caffe2/CMakeFiles/torch_cuda.dir/
/aten/src/ATen/native/cuda/GcdLcmKernel.cu.o
FAILED: caffe2/CMakeFiles/torch_cuda.dir/
/aten/src/ATen/native/cuda/GcdLcmKernel.cu.o
/usr/local/cuda-11.2/bin/nvcc -forward-unknown-to-host-compiler -DAT_PER_OPERATOR_HEADERS -DHAVE_MALLOC_USABLE_SIZE=1 -DHAVE_MMAP=1 -DHAVE_SHM_OPEN=1 -DHAVE_SHM_UNLINK=1 -DIDEEP_USE_MKL -DMAGMA_V2 -DMINIZ_DISABLE_ZIP_READER_CRC32_CHECKS -DONNXIFI_ENABLE_EXT=1 -DONNX_ML=1 -DONNX_NAMESPACE=onnx_torch -DTORCH_CUDA_BUILD_MAIN_LIB -DUSE_C10D_GLOO -DUSE_C10D_MPI -DUSE_C10D_NCCL -DUSE_CUDA -DUSE_DISTRIBUTED -DUSE_EXTERNAL_MZCRC -DUSE_NCCL -DUSE_RPC -DUSE_TENSORPIPE -D_FILE_OFFSET_BITS=64 -Dtorch_cuda_EXPORTS -Iaten/src -I…/aten/src -I. -I…/ -I…/cmake/…/third_party/benchmark/include -I…/cmake/…/third_party/cudnn_frontend/include -I…/third_party/onnx -Ithird_party/onnx -I…/third_party/foxi -Ithird_party/foxi -Iinclude -I…/torch/csrc/distributed -I…/aten/src/THC -I…/aten/src/ATen/cuda -Icaffe2/aten/src -I…/aten/…/third_party/catch/single_include -I…/aten/src/ATen/… -Icaffe2/aten/src/ATen -Inccl/include -I…/c10/cuda/…/… -I…/c10/… -I…/third_party/tensorpipe -Ithird_party/tensorpipe -I…/third_party/tensorpipe/third_party/libnop/include -I…/torch/csrc/api -I…/torch/csrc/api/include -isystem=third_party/gloo -isystem=…/cmake/…/third_party/gloo -isystem=…/cmake/…/third_party/googletest/googlemock/include -isystem=…/cmake/…/third_party/googletest/googletest/include -isystem=…/third_party/protobuf/src -isystem=/home/PJLAB/xingwenzhi/anaconda3/include -isystem=…/third_party/gemmlowp -isystem=…/third_party/neon2sse -isystem=…/third_party/XNNPACK/include -isystem=…/third_party -isystem=…/cmake/…/third_party/eigen -isystem=/home/PJLAB/xingwenzhi/anaconda3/include/python3.9 -isystem=/home/PJLAB/xingwenzhi/anaconda3/lib/python3.9/site-packages/numpy/core/include -isystem=…/cmake/…/third_party/pybind11/include -isystem=/usr/local/cuda-11.2/include -isystem=…/third_party/ideep/mkl-dnn/third_party/oneDNN/include -isystem=…/third_party/ideep/include -Xfatbin -compress-all -DONNX_NAMESPACE=onnx_torch -gencode arch=compute_75,code=sm_75 -Xcudafe --diag_suppress=cc_clobber_ignored,–diag_suppress=integer_sign_change,–diag_suppress=useless_using_declaration,–diag_suppress=set_but_not_used,–diag_suppress=field_without_dll_interface,–diag_suppress=base_class_has_different_dll_interface,–diag_suppress=dll_interface_conflict_none_assumed,–diag_suppress=dll_interface_conflict_dllexport_assumed,–diag_suppress=implicit_return_from_non_void_function,–diag_suppress=unsigned_compare_with_zero,–diag_suppress=declared_but_not_referenced,–diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -Wno-deprecated-gpu-targets --expt-extended-lambda -DCUDA_HAS_FP16=1 -D__CUDA_NO_HALF_OPERATORS
-D__CUDA_NO_HALF_CONVERSIONS
-D__CUDA_NO_HALF2_OPERATORS
-D__CUDA_NO_BFLOAT16_CONVERSIONS
-O3 -DNDEBUG -Xcompiler=-fPIC -DCAFFE2_USE_GLOO -DCUDA_HAS_FP16=1 -DHAVE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -DTH_HAVE_THREAD -Xcompiler=-Wall,-Wextra,-Wno-unused-parameter,-Wno-unused-variable,-Wno-unused-function,-Wno-unused-result,-Wno-unused-local-typedefs,-Wno-missing-field-initializers,-Wno-write-strings,-Wno-unknown-pragmas,-Wno-type-limits,-Wno-array-bounds,-Wno-unknown-pragmas,-Wno-sign-compare,-Wno-strict-overflow,-Wno-strict-aliasing,-Wno-error=deprecated-declarations,-Wno-missing-braces,-Wno-maybe-uninitialized -DTORCH_CUDA_BUILD_MAIN_LIB -Xcompiler -pthread -std=c++14 -MD -MT caffe2/CMakeFiles/torch_cuda.dir/
/aten/src/ATen/native/cuda/GcdLcmKernel.cu.o -MF caffe2/CMakeFiles/torch_cuda.dir/
/aten/src/ATen/native/cuda/GcdLcmKernel.cu.o.d -x cu -c …/aten/src/ATen/native/cuda/GcdLcmKernel.cu -o caffe2/CMakeFiles/torch_cuda.dir/
/aten/src/ATen/native/cuda/GcdLcmKernel.cu.o
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu: In lambda function:
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:19:721: error: the value of ‘at::native::gcd_name’ is not usable in a constant expression
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:16:12: note: ‘at::native::gcd_name’ was not declared ‘constexpr’
const char gcd_name[] = “gcd”;
^
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu: In lambda function:
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:19:1388: error: the value of ‘at::native::gcd_name’ is not usable in a constant expression
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:16:12: note: ‘at::native::gcd_name’ was not declared ‘constexpr’
const char gcd_name[] = “gcd”;
^
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu: In lambda function:
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:19:2050: error: the value of ‘at::native::gcd_name’ is not usable in a constant expression
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:16:12: note: ‘at::native::gcd_name’ was not declared ‘constexpr’
const char gcd_name[] = “gcd”;
^
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu: In lambda function:
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:19:2698: error: the value of ‘at::native::gcd_name’ is not usable in a constant expression
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:16:12: note: ‘at::native::gcd_name’ was not declared ‘constexpr’
const char gcd_name[] = “gcd”;
^
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu: In lambda function:
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:19:3350: error: the value of ‘at::native::gcd_name’ is not usable in a constant expression
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:16:12: note: ‘at::native::gcd_name’ was not declared ‘constexpr’
const char gcd_name[] = “gcd”;
^
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu: In lambda function:
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:38:721: error: the value of ‘at::native::lcm_name’ is not usable in a constant expression
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:35:12: note: ‘at::native::lcm_name’ was not declared ‘constexpr’
const char lcm_name[] = “lcm”;
^
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu: In lambda function:
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:38:1388: error: the value of ‘at::native::lcm_name’ is not usable in a constant expression
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:35:12: note: ‘at::native::lcm_name’ was not declared ‘constexpr’
const char lcm_name[] = “lcm”;
^
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu: In lambda function:
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:38:2050: error: the value of ‘at::native::lcm_name’ is not usable in a constant expression
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:35:12: note: ‘at::native::lcm_name’ was not declared ‘constexpr’
const char lcm_name[] = “lcm”;
^
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu: In lambda function:
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:38:2698: error: the value of ‘at::native::lcm_name’ is not usable in a constant expression
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:35:12: note: ‘at::native::lcm_name’ was not declared ‘constexpr’
const char lcm_name[] = “lcm”;
^
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu: In lambda function:
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:38:3350: error: the value of ‘at::native::lcm_name’ is not usable in a constant expression
…/aten/src/ATen/native/cuda/GcdLcmKernel.cu:35:12: note: ‘at::native::lcm_name’ was not declared ‘constexpr’
const char lcm_name[] = “lcm”;
^
[4562/5257] Building CUDA object caffe2/CMakeFiles/torch_cuda.dir/
/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu.o
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu: In lambda function:
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:794: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = double; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:855: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = double; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:916: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = long int; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:984: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = double; long unsigned int N = 3ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu: In lambda function:
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:2471: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = float; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:2531: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = float; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:2592: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = long int; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:2659: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = float; long unsigned int N = 3ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu: In lambda function:
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:4151: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = c10::Half; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:4215: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = c10::Half; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:4276: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = long int; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:177:4347: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = c10::Half; long unsigned int N = 3ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu: In member function ‘void at::native::structured_fractional_max_pool2d_backward_cuda::impl(const at::Tensor&, const at::Tensor&, c10::IntArrayRef, c10::IntArrayRef, const at::Tensor&, const at::Tensor&)’:
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:248:57: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = long int; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
auto devIndices = indices
.packed_accessor<int64_t, 4>();
^
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu: In lambda function:
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:249:816: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = double; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:249:885: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = double; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu: In lambda function:
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:249:2372: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = float; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:249:2440: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = float; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu: In lambda function:
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:249:3932: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = c10::Half; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
…/aten/src/ATen/native/cuda/FractionalMaxPool2d.cu:249:4004: warning: ‘at::GenericPackedTensorAccessor<T, N, PtrTraits, index_t> at::Tensor::packed_accessor() const & [with T = c10::Half; long unsigned int N = 4ul; PtrTraits = at::DefaultPtrTraits; index_t = long int]’ is deprecated: packed_accessor is deprecated, use packed_accessor32 or packed_accessor64 instead [-Wdeprecated-declarations]
aten/src/ATen/core/TensorBody.h:241:1: note: declared here
GenericPackedTensorAccessor<T,N,PtrTraits,index_t> packed_accessor() const & {
^
[4578/5257] Building CUDA object caffe2/CMakeFiles/torch_cuda.dir/
/aten/src/ATen/native/cuda/ForeachUnaryOp.cu.o
ninja: build stopped: subcommand failed.

If you had solved this problem, pleas reply me your solution. Thanks lot!

Which PyTorch commit are you trying to build?

Hi, thanks for your reply. My pytorch vesion is “Building wheel torch-1.9.0a0+gitdfbd030”, the latest vesion.

Hi, thanks for your reply. My pytorch vesion is “Building wheel torch-1.9.0a0+gitdfbd030”, the latest vesion.

1.9.0a0 is a pre-1.9.0 release and thus not the latest version.
In any case, this commit works for me using CUDA 11.2:

git clone https://github.com/pytorch/pytorch
cd pytorch
git checkout dfbd030
git submodule update --init --recursive 
python setup.py install
>>> import torch
>>> torch.__version__
'1.9.0a0+df837d0'
>>> torch.version.cuda
'11.2'

Thank you very much. In my case, this cannot work for me. I have to chose TensorRT for my model inference. Thanks again.