-
Notifications
You must be signed in to change notification settings - Fork 6
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
nvcc fatal : Unsupported gpu architecture 'compute_30' #8
Comments
I just pushed an update to |
After git pull, I tried a series of build/test/install. I ran build successfully but I faced error message on test; (miniconda3-4.7.12) sekigh@779812fb3431:~/Windowsfolder/sporco-cuda$ python setup.py test error: invalid command 'pytest' I am confused and wonder if I need to install pytest prior to test, So I install pytest by conda install pytest. But I have the same error as the above. |
But at least the build succeeds now, whereas it failed before? I suspect the error that you're seeing on running |
Thank you for your quick response. one is AssertionError under test_cbpdn.py: line 90 several depreciation warnings for su.rndmask(s.shape, frc, dtype=np.float32) Full log statements are as follows: (miniconda3-4.7.12) sekigh@779812fb3431:~/Windowsfolder/sporco-cuda$ python setup.py test sporco_cuda/tests/test_cbpdn.py ...F.......... [ 77%] =================================================================================================== FAILURES =================================================================================================== self = <test_cbpdn.TestSet01 object at 0x7f58dac38350>
E assert 3.1156701478864617e-07 < 1e-10 sporco_cuda/tests/test_cbpdn.py:90: AssertionError sporco_cuda/tests/test_cbpdn.py::TestSet01::test_11 sporco_cuda/tests/test_cbpdn.py::TestSet01::test_12 sporco_cuda/tests/test_cbpdn.py::TestSet01::test_13 sporco_cuda/tests/test_cbpdn.py::TestSet01::test_14 -- Docs: https://docs.pytest.org/en/stable/warnings.html end of text |
Thanks for the log. The deprecation warnings are because |
If you do a |
Thank you for your immediate fix. Yes, python setup.py test works fine without any errors and warnings. I tried following cuda-related scripts under sporco/examples/csc and sporco-cuda/examples and confirmed their working fine; sporco/examples/csc: sporco-cuda/examples: This is closed. I also would like to establish sporco-cuda environment on the first machine mentioned on #7. The cause looks different from this issue. I already tried out your updated setup.py on the first machine but failed in the same errors. I put that errors on #7. |
In parallel to #7 , I try to set up both sporco and sporco-cuda on the second machine by another way, saying commands
conda install sporco
pip install sporco-cuda
conda install sporco successfully install sporco but second command issues another problem saying:
nvcc fatal : Unsupported gpu architecture 'compute_30'
It looks cuda3 architecture should be removed from nvcc option. Could you tell how to detour this problem ?
My environment is:
ubuntu 18.04
python 3.7.4
cuda-11.2
gpu RX3090 (architecture 8.6)
docker 20.10.5
Full error message is as below:
(miniconda3-4.7.12) sekigh@779812fb3431:~$ pip install sporco-cuda
Collecting sporco-cuda
Using cached sporco-cuda-0.0.5.tar.gz (119 kB)
Requirement already satisfied: future in ./.pyenv/versions/miniconda3-4.7.12/lib/python3.7/site-packages (from sporco-cuda) (0.18.2)
Requirement already satisfied: numpy in ./.pyenv/versions/miniconda3-4.7.12/lib/python3.7/site-packages (from sporco-cuda) (1.19.2)
Building wheels for collected packages: sporco-cuda
Building wheel for sporco-cuda (setup.py) ... error
ERROR: Command errored out with exit status 1:
command: /home/sekigh/.pyenv/versions/miniconda3-4.7.12/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-8ag8y8ks/sporco-cuda_b9aca92322af4e80b4d58c47b0bf8e3c/setup.py'"'"'; file='"'"'/tmp/pip-install-8ag8y8ks/sporco-cuda_b9aca92322af4e80b4d58c47b0bf8e3c/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(file);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' bdist_wheel -d /tmp/pip-wheel-k6hvhw7h
cwd: /tmp/pip-install-8ag8y8ks/sporco-cuda_b9aca92322af4e80b4d58c47b0bf8e3c/
Complete output (13 lines):
running bdist_wheel
running build
running build_ext
skipping 'sporco_cuda/util.c' Cython extension (up-to-date)
skipping 'sporco_cuda/cbpdn.c' Cython extension (up-to-date)
building 'sporco_cuda.util' extension
creating build
creating build/temp.linux-x86_64-3.7
creating build/temp.linux-x86_64-3.7/sporco_cuda
creating build/temp.linux-x86_64-3.7/sporco_cuda/src
/usr/local/cuda-11.2/bin/nvcc -I/home/sekigh/.pyenv/versions/miniconda3-4.7.12/lib/python3.7/site-packages/numpy/core/include -I/usr/local/cuda-11.2/include -Isporco_cuda/src -I/home/sekigh/.pyenv/versions/miniconda3-4.7.12/include/python3.7m -c sporco_cuda/src/utils.cu -o build/temp.linux-x86_64-3.7/sporco_cuda/src/utils.o -Xcompiler '-D__builtin_stdarg_start=__builtin_va_start' --compiler-options '-fno-inline' --compiler-options '-fno-strict-aliasing' --compiler-options '-Wall' -gencode arch=compute_30,code=sm_30 -gencode arch=compute_32,code=sm_32 -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_53,code=sm_53 -Xcompiler '-fPIC'
nvcc fatal : Unsupported gpu architecture 'compute_30'
error: command '/usr/local/cuda-11.2/bin/nvcc' failed with exit status 1
ERROR: Failed building wheel for sporco-cuda
Running setup.py clean for sporco-cuda
Failed to build sporco-cuda
Installing collected packages: sporco-cuda
Running setup.py install for sporco-cuda ... error
ERROR: Command errored out with exit status 1:
command: /home/sekigh/.pyenv/versions/miniconda3-4.7.12/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-8ag8y8ks/sporco-cuda_b9aca92322af4e80b4d58c47b0bf8e3c/setup.py'"'"'; file='"'"'/tmp/pip-install-8ag8y8ks/sporco-cuda_b9aca92322af4e80b4d58c47b0bf8e3c/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(file);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' install --record /tmp/pip-record-sybdeo_g/install-record.txt --single-version-externally-managed --compile --install-headers /home/sekigh/.pyenv/versions/miniconda3-4.7.12/include/python3.7m/sporco-cuda
cwd: /tmp/pip-install-8ag8y8ks/sporco-cuda_b9aca92322af4e80b4d58c47b0bf8e3c/
Complete output (19 lines):
running install
running build
running build_ext
cythoning sporco_cuda/util.pyx to sporco_cuda/util.c
/home/sekigh/.pyenv/versions/miniconda3-4.7.12/lib/python3.7/site-packages/Cython/Compiler/Main.py:369: FutureWarning: Cython directive 'language_level' not set, using 2 for now (Py2). This will change in a later release! File: /tmp/pip-install-8ag8y8ks/sporco-cuda_b9aca92322af4e80b4d58c47b0bf8e3c/sporco_cuda/util.pyx
tree = Parsing.p_module(s, pxd, full_module_name)
cythoning sporco_cuda/cbpdn.pyx to sporco_cuda/cbpdn.c
/home/sekigh/.pyenv/versions/miniconda3-4.7.12/lib/python3.7/site-packages/Cython/Compiler/Main.py:369: FutureWarning: Cython directive 'language_level' not set, using 2 for now (Py2). This will change in a later release! File: /tmp/pip-install-8ag8y8ks/sporco-cuda_b9aca92322af4e80b4d58c47b0bf8e3c/sporco_cuda/cbpdn.pyx
tree = Parsing.p_module(s, pxd, full_module_name)
warning: sporco_cuda/cbpdn.pyx:128:47: the result of using negative indices inside of code sections marked as 'wraparound=False' is undefined
warning: sporco_cuda/cbpdn.pyx:250:47: the result of using negative indices inside of code sections marked as 'wraparound=False' is undefined
building 'sporco_cuda.util' extension
creating build
creating build/temp.linux-x86_64-3.7
creating build/temp.linux-x86_64-3.7/sporco_cuda
creating build/temp.linux-x86_64-3.7/sporco_cuda/src
/usr/local/cuda-11.2/bin/nvcc -I/home/sekigh/.pyenv/versions/miniconda3-4.7.12/lib/python3.7/site-packages/numpy/core/include -I/usr/local/cuda-11.2/include -Isporco_cuda/src -I/home/sekigh/.pyenv/versions/miniconda3-4.7.12/include/python3.7m -c sporco_cuda/src/utils.cu -o build/temp.linux-x86_64-3.7/sporco_cuda/src/utils.o -Xcompiler '-D__builtin_stdarg_start=__builtin_va_start' --compiler-options '-fno-inline' --compiler-options '-fno-strict-aliasing' --compiler-options '-Wall' -gencode arch=compute_30,code=sm_30 -gencode arch=compute_32,code=sm_32 -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_53,code=sm_53 -Xcompiler '-fPIC'
nvcc fatal : Unsupported gpu architecture 'compute_30'
error: command '/usr/local/cuda-11.2/bin/nvcc' failed with exit status 1
----------------------------------------
ERROR: Command errored out with exit status 1: /home/sekigh/.pyenv/versions/miniconda3-4.7.12/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-8ag8y8ks/sporco-cuda_b9aca92322af4e80b4d58c47b0bf8e3c/setup.py'"'"'; file='"'"'/tmp/pip-install-8ag8y8ks/sporco-cuda_b9aca92322af4e80b4d58c47b0bf8e3c/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(file);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' install --record /tmp/pip-record-sybdeo_g/install-record.txt --single-version-externally-managed --compile --install-headers /home/sekigh/.pyenv/versions/miniconda3-4.7.12/include/python3.7m/sporco-cuda Check the logs for full command output.
(miniconda3-4.7.12) sekigh@779812fb3431:~$
end of message
The text was updated successfully, but these errors were encountered: