Hi everyone, I need your help, I’m new using a jetson nano and I’m trying to use pycuda in this.
[b]from pycuda.compiler import SourceModule
ImportError: No module named compiler[/b]
Hi,
Are you using python3?
Here is the instruction to get pyCuda installed with pip3:
$ sudo apt-get install python3-pip
$ pip3 install numpy pycuda --user
We can import pycuda.compiler without issue with the above installation.
Thanks.
Hi, AastaLLL
I’m using python3, and the installation correctly ends, in the importation of principal library no problem, but in the importation of Compiler o driver catch the error
This code works, but I’m trying to create new codes with pycuda.compiler this generates the error
import pycuda
import pycuda.driver as drv
drv.init()
print('CUDA device query (PyCUDA version) \n')
print('Detected {} CUDA Capable device(s) \n'.format(drv.Device.count()))
for i in range(drv.Device.count()):
gpu_device = drv.Device(i)
print('Device {}: {}'.format( i, gpu_device.name() ) )
compute_capability = float( '%d.%d' % gpu_device.compute_capability() )
print('\t Compute Capability: {}'.format(compute_capability))
print('\t Total Memory: {} megabytes'.format(gpu_device.total_memory()//(1024**2)))
# The following will give us all remaining device attributes as seen
# in the original deviceQuery.
# We set up a dictionary as such so that we can easily index
# the values using a string descriptor.
device_attributes_tuples = gpu_device.get_attributes().items()
device_attributes = {}
for k, v in device_attributes_tuples:
device_attributes[str(k)] = v
num_mp = device_attributes['MULTIPROCESSOR_COUNT']
# Cores per multiprocessor is not reported by the GPU!
# We must use a lookup table based on compute capability.
# See the following:
# http://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#compute-capabilities
cuda_cores_per_mp = { 5.0 : 128, 5.1 : 128, 5.2 : 128, 5.3 : 128, 6.0 : 64, 6.1 : 128, 6.2 : 128}[compute_capability]
print('\t ({}) Multiprocessors, ({}) CUDA Cores / Multiprocessor: {} CUDA Cores'.format(num_mp, cuda_cores_per_mp, num_mp*cuda_cores_per_mp))
device_attributes.pop('MULTIPROCESSOR_COUNT')
for k in device_attributes.keys():
print('\t {}: {}'.format(k, device_attributes[k]))
CUDA device query (PyCUDA version)
Detected 1 CUDA Capable device(s)
Device 0: NVIDIA Tegra X1
Compute Capability: 5.3
Total Memory: 3956 megabytes
(1) Multiprocessors, (128) CUDA Cores / Multiprocessor: 128 CUDA Cores
ASYNC_ENGINE_COUNT: 1
CAN_MAP_HOST_MEMORY: 1
CLOCK_RATE: 921600
COMPUTE_CAPABILITY_MAJOR: 5
COMPUTE_CAPABILITY_MINOR: 3
COMPUTE_MODE: DEFAULT
CONCURRENT_KERNELS: 1
ECC_ENABLED: 0
GLOBAL_L1_CACHE_SUPPORTED: 1
GLOBAL_MEMORY_BUS_WIDTH: 64
GPU_OVERLAP: 1
INTEGRATED: 1
KERNEL_EXEC_TIMEOUT: 1
L2_CACHE_SIZE: 262144
LOCAL_L1_CACHE_SUPPORTED: 1
MANAGED_MEMORY: 1
MAXIMUM_SURFACE1D_LAYERED_LAYERS: 2048
MAXIMUM_SURFACE1D_LAYERED_WIDTH: 16384
MAXIMUM_SURFACE1D_WIDTH: 16384
MAXIMUM_SURFACE2D_HEIGHT: 65536
MAXIMUM_SURFACE2D_LAYERED_HEIGHT: 16384
MAXIMUM_SURFACE2D_LAYERED_LAYERS: 2048
MAXIMUM_SURFACE2D_LAYERED_WIDTH: 16384
MAXIMUM_SURFACE2D_WIDTH: 65536
MAXIMUM_SURFACE3D_DEPTH: 4096
MAXIMUM_SURFACE3D_HEIGHT: 4096
MAXIMUM_SURFACE3D_WIDTH: 4096
MAXIMUM_SURFACECUBEMAP_LAYERED_LAYERS: 2046
MAXIMUM_SURFACECUBEMAP_LAYERED_WIDTH: 16384
MAXIMUM_SURFACECUBEMAP_WIDTH: 16384
MAXIMUM_TEXTURE1D_LAYERED_LAYERS: 2048
MAXIMUM_TEXTURE1D_LAYERED_WIDTH: 16384
MAXIMUM_TEXTURE1D_LINEAR_WIDTH: 134217728
MAXIMUM_TEXTURE1D_MIPMAPPED_WIDTH: 16384
MAXIMUM_TEXTURE1D_WIDTH: 65536
MAXIMUM_TEXTURE2D_ARRAY_HEIGHT: 16384
MAXIMUM_TEXTURE2D_ARRAY_NUMSLICES: 2048
MAXIMUM_TEXTURE2D_ARRAY_WIDTH: 16384
MAXIMUM_TEXTURE2D_GATHER_HEIGHT: 16384
MAXIMUM_TEXTURE2D_GATHER_WIDTH: 16384
MAXIMUM_TEXTURE2D_HEIGHT: 65536
MAXIMUM_TEXTURE2D_LINEAR_HEIGHT: 65536
MAXIMUM_TEXTURE2D_LINEAR_PITCH: 1048544
MAXIMUM_TEXTURE2D_LINEAR_WIDTH: 65536
MAXIMUM_TEXTURE2D_MIPMAPPED_HEIGHT: 16384
MAXIMUM_TEXTURE2D_MIPMAPPED_WIDTH: 16384
MAXIMUM_TEXTURE2D_WIDTH: 65536
MAXIMUM_TEXTURE3D_DEPTH: 4096
MAXIMUM_TEXTURE3D_DEPTH_ALTERNATE: 16384
MAXIMUM_TEXTURE3D_HEIGHT: 4096
MAXIMUM_TEXTURE3D_HEIGHT_ALTERNATE: 2048
MAXIMUM_TEXTURE3D_WIDTH: 4096
MAXIMUM_TEXTURE3D_WIDTH_ALTERNATE: 2048
MAXIMUM_TEXTURECUBEMAP_LAYERED_LAYERS: 2046
MAXIMUM_TEXTURECUBEMAP_LAYERED_WIDTH: 16384
MAXIMUM_TEXTURECUBEMAP_WIDTH: 16384
MAX_BLOCK_DIM_X: 1024
MAX_BLOCK_DIM_Y: 1024
MAX_BLOCK_DIM_Z: 64
MAX_GRID_DIM_X: 2147483647
MAX_GRID_DIM_Y: 65535
MAX_GRID_DIM_Z: 65535
MAX_PITCH: 2147483647
MAX_REGISTERS_PER_BLOCK: 32768
MAX_REGISTERS_PER_MULTIPROCESSOR: 65536
MAX_SHARED_MEMORY_PER_BLOCK: 49152
MAX_SHARED_MEMORY_PER_MULTIPROCESSOR: 65536
MAX_THREADS_PER_BLOCK: 1024
MAX_THREADS_PER_MULTIPROCESSOR: 2048
MEMORY_CLOCK_RATE: 12750
MULTI_GPU_BOARD: 0
MULTI_GPU_BOARD_GROUP_ID: 0
PCI_BUS_ID: 0
PCI_DEVICE_ID: 0
PCI_DOMAIN_ID: 0
STREAM_PRIORITIES_SUPPORTED: 1
SURFACE_ALIGNMENT: 512
TCC_DRIVER: 0
TEXTURE_ALIGNMENT: 512
TEXTURE_PITCH_ALIGNMENT: 32
TOTAL_CONSTANT_MEMORY: 65536
UNIFIED_ADDRESSING: 1
WARP_SIZE: 32
Traceback (most recent call last):
File "/home/ygb/Desktop/Pycuda-2.py", line 3, in <module>
import pycuda.driver as drv
File "/home/ygb/Desktop/pycuda.py", line 6, in <module>
from pycuda.compiler import SourceModule
ModuleNotFoundError: No module named 'pycuda.compiler'; 'pycuda' is not a package
I solved my problem by adding the following lines in the prompt
gedit ~/.bashrc
# Add the following at the end of the file:
export PATH="/usr/local/cuda/bin:${PATH}"
export LD_LIBRARY_PATH="/usr/local/cuda/lib64:${LD_LIBRARY_PATH}"
export CPATH=$CPATH:/usr/local/cuda-10.0/targets/aarch64-linux/include
export LIBRARY_PATH=$LIBRARY_PATH:/usr/local/cuda-10.0/targets/aarch64-linux/lib