Jetson TX2 problem with pytorch install

Robotics & Edge Computing Jetson & Embedded Systems Jetson TX2
1

I tried to install pytorch in TX2 board. but it didn’t work…
DO u guys know the reason why???
The Errors below this sentence occurred when i used python3 setup.py install
80 calls function ‘_Z29ncclReduceScatterRing_max_f64P14CollectiveArgs’ with regcount of 96
nvlink error : entry function ‘_Z33ncclAllReduceRingLLKernel_sum_i328ncclColl’ with max regcount of 80 calls function ‘_Z29ncclReduceScatterRing_max_f64P14CollectiveArgs’ with regcount of 96
nvlink error : entry function ‘_Z32ncclAllReduceTreeLLKernel_sum_u88ncclColl’ with max regcount of 80 calls function ‘_Z29ncclReduceScatterRing_max_f64P14CollectiveArgs’ with regcount of 96
nvlink error : entry function ‘_Z32ncclAllReduceRingLLKernel_sum_u88ncclColl’ with max regcount of 80 calls function ‘_Z29ncclReduceScatterRing_max_f64P14CollectiveArgs’ with regcount of 96
nvlink error : entry function ‘_Z32ncclAllReduceTreeLLKernel_sum_i88ncclColl’ with max regcount of 80 calls function ‘_Z29ncclReduceScatterRing_max_f64P14CollectiveArgs’ with regcount of 96
nvlink error : entry function ‘_Z32ncclAllReduceRingLLKernel_sum_i88ncclColl’ with max regcount of 80 calls function ‘_Z29ncclReduceScatterRing_max_f64P14CollectiveArgs’ with regcount of 96
Makefile:68: recipe for target ‘/pytorch/build/nccl/obj/collectives/device/devlink.o’ failed
make[2]: *** [/pytorch/build/nccl/obj/collectives/device/devlink.o] Error 255
make[2]: Leaving directory ‘/pytorch/third_party/nccl/nccl/src/collectives/device’
Makefile:49: recipe for target ‘/pytorch/build/nccl/obj/collectives/device/colldevice.a’ failed
make[1]: *** [/pytorch/build/nccl/obj/collectives/device/colldevice.a] Error 2
make[1]: Leaving directory ‘/pytorch/third_party/nccl/nccl/src’
Makefile:25: recipe for target ‘src.build’ failed
make: *** [src.build] Error 2
[4/2908] Building CXX object third_party/protobuf…f.dir/__/src/google/protobuf/compiler/parser.cc.o
ninja: build stopped: subcommand failed.
Traceback (most recent call last):
File “setup.py”, line 759, in
build_deps()
File “setup.py”, line 321, in build_deps
cmake=cmake)
File “/pytorch/tools/build_pytorch_libs.py”, line 63, in build_caffe2
cmake.build(my_env)
File “/pytorch/tools/setup_helpers/cmake.py”, line 330, in build
self.run(build_args, my_env)
File “/pytorch/tools/setup_helpers/cmake.py”, line 143, in run
check_call(command, cwd=self.build_dir, env=env)
File “/usr/lib/python3.5/subprocess.py”, line 581, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command ‘[‘cmake’, ‘–build’, ‘.’, ‘–target’, ‘install’, ‘–config’, ‘Release’, ‘–’, ‘-j’, ‘4’]’ returned non-zero exit status 1

2

Hi,

NVLINK doesn’t support Jetson platform. Please turn it off when building pyTorch.

$ export USE_NCCL=0
$ export USE_DISTRIBUTED=0
$ export TORCH_CUDA_ARCH_LIST="5.3;6.2;7.2"

Here is our tutorial for building pyTorch from source for your reference:
https://devtalk.nvidia.com/default/topic/1049071/jetson-nano/pytorch-for-jetson-nano/

Thanks.