GPU-Programming

Currently learning some GPU CUDA programming.

Contents

• vecMultiply: simple vector multiplication.
• matMul: naive matrix multiplication (low CGMA).
• matMulTiled: tiled matrix multiplication.
• color2gray: turn an RGB image into a Grayscale image.
• imgBlur: blur a Grayscale image by averaging pixels in a sliding window.
• imgBlurPlus: a more efficient blurring kernel that uses shared memory.
• histogram: compute the histogram of a 1D sequence.
• convolution: apply a tiled 2D convolution on an RGB image.

References

• CUDA By Example
• Programming Massively Parallel Processors
• CS 179: GPU Programming