Hi, our cryoSPARC v0.6.5 ran well with CUDA 9.1. When we upgraded to CUDA 9.2.88, however, the experiment could not run with the following errors.
Launched job ***********
License is valid.
Traceback (most recent call last):
File “/home/cryosparc/cryosparc/cryosparc-compute/sparcjob.py”, line 84, in
import engine
File “/home/cryosparc/cryosparc/cryosparc-compute/engine/init.py”, line 8, in
from engine import *
File “/home/cryosparc/cryosparc/cryosparc-compute/engine/engine.py”, line 17, in
import gfourier
File “/home/cryosparc/cryosparc/cryosparc-compute/engine/gfourier.py”, line 14, in
import skcuda.fft
File “/home/cryosparc/cryosparc/anaconda2/lib/python2.7/site-packages/skcuda/fft.py”, line 13, in
from . import cufft
File “/home/cryosparc/cryosparc/anaconda2/lib/python2.7/site-packages/skcuda/cufft.py”, line 238, in
_libcufft.cufftSetCompatibilityMode.restype = int
File “/home/cryosparc/cryosparc/anaconda2/lib/python2.7/ctypes/init.py”, line 379, in getattr
func = self.getitem(name)
File “/home/cryosparc/cryosparc/anaconda2/lib/python2.7/ctypes/init.py”, line 384, in getitem
func = self._FuncPtr((name_or_ordinal, self))
AttributeError: /opt/cuda/cuda-9.2/lib64/libcufft.so: undefined symbol: cufftSetCompatibilityMode
The same error occurred when pycuda-2017.1.1 is reinstalled.
What could we do other than downgrade the CUDA version? Many thanks!