A Python tool for quick and dirty translation of CUDA kernels to OpenCL

You could include a link to the full documentation of your project here.

As a command-line tool:

cutocl vector_add.cu

Or use from Python:

from cutocl import cutocl
kernel_string = """
__global__ void vector_add(float *c, float *a, float *b, int n) {
    int i = blockIdx.x * block_size_x + threadIdx.x;
    if (i<n) {
        c[i] = a[i] + b[i];
    }
}
"""
opencl_kernel = cutocl.translate(kernel_string)
print(opencl_kernel)

This should produce:

__kernel void vector_add(__global float *c, __global float *a, __global float *b, int n) {
    int i = get_group_id(0) * block_size_x + get_local_id(0);
    if (i<n) {
        c[i] = a[i] + b[i];
    }
}

clone the repository
git clone [email protected]:benvanwerkhoven/cutocl.git
change into the top-level directory
cd cutocl
install using
pip install .

• Python 3

Copyright (c) 2016, Ben van Werkhoven

Apache Software License 2.0

Contributing authors so far:

• Ben van Werkhoven