Hi ,
Just installed pycuda and tried to run matrixmul code from <a href="MatrixmulSimple - Andreas Klöckner's Former Wiki"target=_blank>here. Got error in the phase where we compile kernel code . When i copy this in file test.py, it runs well. But when i execute step by step it gives error. After this, i might write some code on graph algorithm and get same error. so wanted to clean it now.
#!/usr/bin/env python
… #-- coding: utf-8 --
… import numpy as np
from pycuda import driver, compiler, gpuarray, tools– initialize the device
import pycuda.autoinit
kernel_code_template = “”"
… global void MatrixMulKernel (float *a, float *b, float *c)
… {
… int tx = threadIdx.x;
… int ty = threadIdx.y;
… float Pvalue=0;
… for (int k=0;k<(MATRIX_SIZE)s;++k){
… float Aelement = a[ty * %(MATRIX_SIZE)s + k];
… float Belement = b[k * %(MATRIX_SIZE)s + tx];
… Pvalue += Aelement + Belement;
… }
… c[ty * %(MATRIX_SIZE)s + tx] = Pvalue;
… }
… “”"
MATRIX_SIZE =2
a_cpu = np.random.randn(MATRIX_SIZE,MATRIX_SIZE).astype(np.float32)
b_cpu = np.random.randn(MATRIX_SIZE, MATRIX_SIZE).astype(np.float32)
c_cpu = np.dot(a_cpu, b_cpu)
a_gpu = gpuarray.to_gpu(a_cpu)
b_gpu = gpuarray.to_gpu(b_cpu)
c_gpu = gpuarray.empty((MATRIX_SIZE,MATRIX_SIZE),np.float32)
kernel_code = kernel_code_template %{
… ‘MATRIX_SIZE’:MATRIX_SIZE
… }
mod = compiler.SourceModule(kernel_code)
[i]Traceback (most recent call last):
File “”, line 1, in
File “/usr/local/lib/python2.7/dist-packages/pycuda-2011.2.2-py2.7-linux-x86_64.egg/pycuda/compiler.py”, line 283, in init
arch, code, cache_dir, include_dirs)
File “/usr/local/lib/python2.7/dist-packages/pycuda-2011.2.2-py2.7-linux-x86_64.egg/pycuda/compiler.py”, line 273, in compile
return compile_plain(source, options, keep, nvcc, cache_dir)
File “/usr/local/lib/python2.7/dist-packages/pycuda-2011.2.2-py2.7-linux-x86_64.egg/pycuda/compiler.py”, line 129, in compile_plain
cmdline, stdout=stdout, stderr=stderr)
pycuda.driver.CompileError: nvcc compilation of /tmp/tmpGuRUsm/kernel.cu failed
[command: nvcc --cubin -arch sm_11 -I/usr/local/lib/python2.7/dist-packages/pycuda-2011.2.2-py2.7-linux-x86_64.egg/pycuda/…/include/pycuda kernel.cu]
[stderr:
kernel.cu(8): error: identifier “MATRIX_SIZE” is undefined
kernel.cu(8): error: expected a “;”
2 errors detected in the compilation of “/tmp/tmpxft_00007bbf_00000000-4_kernel.cpp4.ii”.
]
[/i]
could it be some indent error which i don’t see or don’t know ?