sh make.sh

[Markovcom]

x86_64-conda_cos6-linux-gnu-gcc -pthread -fno-strict-aliasing -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O3 -pipe -DNDEBUG -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -DWITH_CUDA -I/home/zhoulinmao/anaconda2/lib/python2.7/site-packages/torch/utils/ffi/../../lib/include -I/home/zhoulinmao/anaconda2/lib/python2.7/site-packages/torch/utils/ffi/../../lib/include/TH -I/home/zhoulinmao/anaconda2/lib/python2.7/site-packages/torch/utils/ffi/../../lib/include/THC -I/usr/local/cuda/include -I/home/zhoulinmao/anaconda2/include/python2.7 -c /home/code/zhoulinmao/faster-rcnn.pytorch/lib/model/roi_crop/src/roi_crop_cuda.c -o ./home/code/zhoulinmao/faster-rcnn.pytorch/lib/model/roi_crop/src/roi_crop_cuda.o
x86_64-conda_cos6-linux-gnu-gcc -pthread -shared -Wl,-O2,--sort-common,--as-needed,-z,relro,-z,now -Wl,-rpath,/home/zhoulinmao/anaconda2/lib -L/home/zhoulinmao/anaconda2/lib -Wl,--no-as-needed ./_roi_crop.o ./home/code/zhoulinmao/faster-rcnn.pytorch/lib/model/roi_crop/src/roi_crop.o ./home/code/zhoulinmao/faster-rcnn.pytorch/lib/model/roi_crop/src/roi_crop_cuda.o /home/code/zhoulinmao/faster-rcnn.pytorch/lib/model/roi_crop/src/roi_crop_cuda_kernel.cu.o -L/home/zhoulinmao/anaconda2/lib -lpython2.7 -o ./_roi_crop.so
x86_64-conda_cos6-linux-gnu-gcc: error: /home/code/zhoulinmao/faster-rcnn.pytorch/lib/model/roi_crop/src/roi_crop_cuda_kernel.cu.o: No such file or directory
Traceback (most recent call last):
File "build.py", line 36, in
ffi.build()
File "/home/zhoulinmao/anaconda2/lib/python2.7/site-packages/torch/utils/ffi/init.py", line 167, in build
_build_extension(ffi, cffi_wrapper_name, target_dir, verbose)
File "/home/zhoulinmao/anaconda2/lib/python2.7/site-packages/torch/utils/ffi/init.py", line 103, in _build_extension
ffi.compile(tmpdir=tmpdir, verbose=verbose, target=libname)
File "/home/zhoulinmao/anaconda2/lib/python2.7/site-packages/cffi/api.py", line 697, in compile
compiler_verbose=verbose, debug=debug, **kwds)
File "/home/zhoulinmao/anaconda2/lib/python2.7/site-packages/cffi/recompiler.py", line 1520, in recompile
compiler_verbose, debug)
File "/home/zhoulinmao/anaconda2/lib/python2.7/site-packages/cffi/ffiplatform.py", line 22, in compile
outputfilename = _build(tmpdir, ext, compiler_verbose, debug)
File "/home/zhoulinmao/anaconda2/lib/python2.7/site-packages/cffi/ffiplatform.py", line 58, in _build
raise VerificationError('%s: %s' % (e.class.name, e))
cffi.error.VerificationError: LinkError: command 'x86_64-conda_cos6-linux-gnu-gcc' failed with exit status 1

[jwyang]

Hi, @Markovcom , you might want to check which arch your GPU is, and modify sh make.sh on line

faster-rcnn.pytorch/lib/make.sh

Line 8 in 1dbb072

CUDA_ARCH="-gencode arch=compute_30,code=sm_30 \

[Markovcom]

@jwyang thanks,it works well

I'm getting the same error on a Tesla P100, despite setting the right gencode. I tried both:

CUDA_ARCH="-gencode arch=compute_60,code=sm_60 \ -gencode arch=compute_35,code=sm_35 \ -gencode arch=compute_37,code=sm_37 \ -gencode arch=compute_50,code=sm_50 \ -gencode arch=compute_52,code=sm_52 \ -gencode arch=compute_61,code=sm_61 "

and

CUDA_ARCH="-gencode arch=compute_60,code=sm_60"

[AndresPMD]

@Codewithsk Did you manage to solve this issue?

Yeah! Turns out my CUDA path was set wrong.

[ControllableGeneration]

@jwyang What if I do not have a CUDA-enabled GPU? What do I have to change in make.sh? I already visited #25 from tf-faster-rcnn. But basically I am getting an error as in #114 during compilation which they did not encounter.