240
edits
Changes
→Attempt 1:
== PHASE 3 ==
We have tried to optimize the maze program. However, the result by removing some of optimization did not improve the speed of if statements to reduce thread divergence, however the program. In fact, it attempt made the program slower than the parallelization but phase 2 version and only a little bit faster than the serialversion.
__global__ void k_drawWalls(png_byte* rows, const short* cells, const int width, const int height, const int len, const int size) {
int i = blockIdx.x * blockDim.x + threadIdx.x;
}
// Set pixels to white according to the pattern the cell belongs to
__global__ void k_drawPaths(png_byte* rows, const short* cells, const int width, const int height, const int len) {
int i = blockIdx.x * blockDim.x + threadIdx.x;
}
[[File:SPODiagram.PNG]]
We decided to use shared memory in the GPU to improve the speed of the program. However, the maze image does not show correctly: the paths were showing as randomly coloured pixels; the cause is due to the threads setting only a part (1/3, 2/3) of a pixel'''3s values to hexadecimal F. Optimize'''