Cryosparc installation failing with pycuda

Install, Configure and Update
1

Hi,

I am trying to install Cryosparc v4.2 on my ububtu 22.04.2 LTS workstation. But it’s failing with the following error.

This is the error at the end.

      ***error: command '/usr/bin/gcc' failed with exit code 1***
***      [end of output]***
***  ***
***  note: This error originates from a subprocess, and is likely not a problem with pip.***
***error: legacy-install-failure***

***× Encountered error while trying to install package.***
***╰─> pycuda***

***note: This is an issue with the package mentioned above, not pip.***
***hint: See above for output from the failure.***
***check_install_deps.sh: 66: ERROR: installing python failed.***

What I understand is the installation is failing to build "pycuda" wheel. below is the error log 
 ***note: This error originates from a subprocess, and is likely not a problem with pip.***
***  ERROR: Failed building wheel for pycuda***
***  Running setup.py clean for pycuda***
***Failed to build pycuda***

Does anybody know how to solve this issue?
Thanks for any help.

2

Which cuda version are you using. CryoSPARC only supports up to Cuda 11.8, so if you’re using Cuda 12, you will run into this issue

3

Welcome to the forum @vijaypkumar90.
Please can you post

  • additional lines from the error message that precede
  • the full command that resulted in this output
  • the output of nvidia-smi on the computer where you ran the command

You may also want to check out related topics.

5

Thank you @wtempel

The command I used for installing is:

./install.sh --standalone
–license $LICENSE_ID
–worker_path /Software/cryosparc/cryosparc_worker/
–cudapath /usr/local/cuda
–ssdpath /scratch/cryosparc_cache
–initial_email “xxxxx@gmail.com”
–initial_password “xxxxxx”
–initial_username “vijay”
–initial_firstname “Vijay”
–initial_lastname “kumar”
–port 61000

The output of above command is really big and may be repetitive. So I am pasting last part of the output.

  src/cpp/cuda.hpp:143:27: note: in definition of macro ‘CUDAPP_CALL_GUARDED’
    143 |     cu_status_code = NAME ARGLIST; \
        |                           ^~~~~~~
  In file included from src/cpp/cuda.hpp:23,
                   from src/wrapper/wrap_cudadrv.cpp:3:
  /usr/local/cuda/include/cudaProfiler.h:143:36: note: declared here
    143 | __CUDA_DEPRECATED CUresult CUDAAPI cuProfilerInitialize(const char *configFile, const char *outputFile, CUoutput_mode outputMode);
        |                                    ^~~~~~~~~~~~~~~~~~~~
  src/wrapper/wrap_cudadrv.cpp: In function ‘pycudaboost::python::tuple {anonymous}::mem_alloc_pitch_wrap(size_t, size_t, unsigned int)’:
  src/wrapper/wrap_cudadrv.cpp:121:10: warning: ‘template<class> class std::auto_ptr’ is deprecated: use 'std::unique_ptr' instead [-Wdeprecated-declarations]
    121 |     std::auto_ptr<device_allocation> da;
        |          ^~~~~~~~
  In file included from /usr/include/c++/11/memory:76,
                   from bpl-subset/bpl_subset/boost/config/no_tr1/memory.hpp:21,
                   from bpl-subset/bpl_subset/boost/smart_ptr/shared_ptr.hpp:27,
                   from bpl-subset/bpl_subset/boost/shared_ptr.hpp:17,
                   from src/cpp/cuda.hpp:30,
                   from src/wrapper/wrap_cudadrv.cpp:3:
  /usr/include/c++/11/bits/unique_ptr.h:57:28: note: declared here
     57 |   template<typename> class auto_ptr;
        |                            ^~~~~~~~
  src/wrapper/wrap_cudadrv.cpp: In function ‘pycuda::module* {anonymous}::module_from_buffer(pycudaboost::python::api::object, pycudaboost::python::api::object, pycudaboost::python::api::object)’:
  src/wrapper/wrap_cudadrv.cpp:357:30: warning: ‘int PyObject_AsCharBuffer(PyObject*, const char**, Py_ssize_t*)’ is deprecated [-Wdeprecated-declarations]
    357 |     if (PyObject_AsCharBuffer(buffer.ptr(), &mod_buf, &len))
        |         ~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  In file included from /Software/cryosparc/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/include/python3.8/Python.h:147,
                   from bpl-subset/bpl_subset/boost/python/detail/wrap_python.hpp:142,
                   from bpl-subset/bpl_subset/boost/python/detail/prefix.hpp:13,
                   from bpl-subset/bpl_subset/boost/python/args.hpp:8,
                   from bpl-subset/bpl_subset/boost/python.hpp:11,
                   from src/cpp/cuda.hpp:36,
                   from src/wrapper/wrap_cudadrv.cpp:3:
  /Software/cryosparc/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/include/python3.8/abstract.h:320:17: note: declared here
    320 | PyAPI_FUNC(int) PyObject_AsCharBuffer(PyObject *obj,
        |                 ^~~~~~~~~~~~~~~~~~~~~
  src/wrapper/wrap_cudadrv.cpp: In member function ‘void {anonymous}::Linker::add_data(pycudaboost::python::api::object, CUjitInputType, pycudaboost::python::str)’:
  src/wrapper/wrap_cudadrv.cpp:502:34: warning: ‘int PyObject_AsCharBuffer(PyObject*, const char**, Py_ssize_t*)’ is deprecated [-Wdeprecated-declarations]
    502 |         if (PyObject_AsCharBuffer(py_data.ptr(), &data_buf, &data_buf_len) != 0) {
        |             ~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  In file included from /Software/cryosparc/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/include/python3.8/Python.h:147,
                   from bpl-subset/bpl_subset/boost/python/detail/wrap_python.hpp:142,
                   from bpl-subset/bpl_subset/boost/python/detail/prefix.hpp:13,
                   from bpl-subset/bpl_subset/boost/python/args.hpp:8,
                   from bpl-subset/bpl_subset/boost/python.hpp:11,
                   from src/cpp/cuda.hpp:36,
                   from src/wrapper/wrap_cudadrv.cpp:3:
  /Software/cryosparc/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/include/python3.8/abstract.h:320:17: note: declared here
    320 | PyAPI_FUNC(int) PyObject_AsCharBuffer(PyObject *obj,
        |                 ^~~~~~~~~~~~~~~~~~~~~
  src/wrapper/wrap_cudadrv.cpp: In function ‘pycudaboost::python::handle<> {anonymous}::numpy_empty(pycudaboost::python::api::object, pycudaboost::python::api::object, pycudaboost::python::api::object, unsigned int)’:
  src/wrapper/wrap_cudadrv.cpp:574:10: warning: ‘template<class> class std::auto_ptr’ is deprecated: use 'std::unique_ptr' instead [-Wdeprecated-declarations]
    574 |     std::auto_ptr<Allocation> alloc(
        |          ^~~~~~~~
  In file included from /usr/include/c++/11/memory:76,
                   from bpl-subset/bpl_subset/boost/config/no_tr1/memory.hpp:21,
                   from bpl-subset/bpl_subset/boost/smart_ptr/shared_ptr.hpp:27,
                   from bpl-subset/bpl_subset/boost/shared_ptr.hpp:17,
                   from src/cpp/cuda.hpp:30,
                   from src/wrapper/wrap_cudadrv.cpp:3:
  /usr/include/c++/11/bits/unique_ptr.h:57:28: note: declared here
     57 |   template<typename> class auto_ptr;
        |                            ^~~~~~~~
  src/wrapper/wrap_cudadrv.cpp: In function ‘pycudaboost::python::handle<> {anonymous}::register_host_memory(pycudaboost::python::api::object, unsigned int)’:
  src/wrapper/wrap_cudadrv.cpp:615:10: warning: ‘template<class> class std::auto_ptr’ is deprecated: use 'std::unique_ptr' instead [-Wdeprecated-declarations]
    615 |     std::auto_ptr<registered_host_memory> regmem(
        |          ^~~~~~~~
  In file included from /usr/include/c++/11/memory:76,
                   from bpl-subset/bpl_subset/boost/config/no_tr1/memory.hpp:21,
                   from bpl-subset/bpl_subset/boost/smart_ptr/shared_ptr.hpp:27,
                   from bpl-subset/bpl_subset/boost/shared_ptr.hpp:17,
                   from src/cpp/cuda.hpp:30,
                   from src/wrapper/wrap_cudadrv.cpp:3:
  /usr/include/c++/11/bits/unique_ptr.h:57:28: note: declared here
     57 |   template<typename> class auto_ptr;
        |                            ^~~~~~~~
  src/wrapper/wrap_cudadrv.cpp: In function ‘void init_module__driver()’:
  src/wrapper/wrap_cudadrv.cpp:997:26: error: ‘CU_TARGET_COMPUTE_20’ was not declared in this scope; did you mean ‘CU_TARGET_COMPUTE_90’?
    997 |     .value("COMPUTE_20", CU_TARGET_COMPUTE_20)
        |                          ^~~~~~~~~~~~~~~~~~~~
        |                          CU_TARGET_COMPUTE_90
  src/wrapper/wrap_cudadrv.cpp:1000:26: error: ‘CU_TARGET_COMPUTE_21’ was not declared in this scope; did you mean ‘CU_TARGET_COMPUTE_61’?
   1000 |     .value("COMPUTE_21", CU_TARGET_COMPUTE_21)
        |                          ^~~~~~~~~~~~~~~~~~~~
        |                          CU_TARGET_COMPUTE_61
  error: command '/usr/bin/gcc' failed with exit code 1
  [end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: legacy-install-failure

× Encountered error while trying to install package.
╰─> pycuda

note: This is an issue with the package mentioned above, not pip.
hint: See above for output from the failure.
check_install_deps.sh: 66: ERROR: installing python failed.

The output of nvidia-smi command

$ nvidia-smi
Fri Jul 14 10:58:21 2023
±--------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.54.03 Driver Version: 535.54.03 CUDA Version: 12.2 |
|-----------------------------------------±---------------------±---------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+======================+======================|
| 0 NVIDIA GeForce RTX 2080 Ti Off | 00000000:18:00.0 Off | N/A |
| 30% 29C P8 10W / 250W | 6MiB / 11264MiB | 0% Default |
| | | N/A |
±----------------------------------------±---------------------±---------------------+
| 1 NVIDIA GeForce RTX 2080 Ti Off | 00000000:3B:00.0 Off | N/A |
| 30% 32C P8 12W / 250W | 6MiB / 11264MiB | 0% Default |
| | | N/A |
±----------------------------------------±---------------------±---------------------+
| 2 NVIDIA GeForce RTX 2080 Ti Off | 00000000:86:00.0 Off | N/A |
| 30% 31C P8 5W / 250W | 6MiB / 11264MiB | 0% Default |
| | | N/A |
±----------------------------------------±---------------------±---------------------+
| 3 NVIDIA GeForce RTX 2080 Ti Off | 00000000:AF:00.0 On | N/A |
| 30% 31C P8 28W / 250W | 321MiB / 11264MiB | 1% Default |
| | | N/A |
±----------------------------------------±---------------------±---------------------+

±--------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=======================================================================================|
| 0 N/A N/A 3453 G /usr/lib/xorg/Xorg 4MiB |
| 1 N/A N/A 3453 G /usr/lib/xorg/Xorg 4MiB |
| 2 N/A N/A 3453 G /usr/lib/xorg/Xorg 4MiB |
| 3 N/A N/A 3453 G /usr/lib/xorg/Xorg 122MiB |
| 3 N/A N/A 3771 G /usr/bin/gnome-shell 29MiB |
| 3 N/A N/A 34650 G …irefox/2908/usr/lib/firefox/firefox 167MiB |
±--------------------------------------------------------------------------------------+

6

Thanks @vijaypkumar90 for this information.

What is the output of the command

/usr/local/cuda/bin/nvcc --version

?

7

@wtempel
Here is the output of /usr/local/cuda/bin/nvcc --version

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Mon_Apr__3_17:16:06_PDT_2023
Cuda compilation tools, release 12.1, V12.1.105
Build cuda_12.1.r12.1/compiler.32688072_0

and the output of nvcc --version

nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Thu_Nov_18_09:45:30_PST_2021
Cuda compilation tools, release 11.5, V11.5.119
Build cuda_11.5.r11.5/compiler.30672275_0

8

The install.sh --cudapath parameter must point to a version of the toolkit that is supported by CryoSPARC, but version 11.8 is the newest version currently supported.
Suggestion:
Let /path/to/cryosparc be the directory that holds the cryosparc_master and cryosparc_worker. If there is enough space inside /path/to/cryosparc, you could run these commands under the non-root Linux account under which you will also install CryoSPARC (details):

cd /path/to/cryosparc/
wget https://developer.download.nvidia.com/compute/cuda/11.8.0/local_installers/cuda_11.8.0_520.61.05_linux.run
bash cuda_11.8.0_520.61.05_linux.run --silent --toolkit --toolkitpath=/path/to/cryosparc/cuda-11.8.0 --defaultroot=/path/to/cryosparc/cuda-11.8.0

and later use

--cudapath /path/to/cryosparc/cuda-11.8.0

during installation of CryoSPARC.

1 Like
9

@wtempel Thank you so much.

i was not able install the rpm. But I followed a different method to downgrade my cuda version.
After that the installation was successful.
thank you again.