search

melsonGit / serial_cpp_vs_parallel_cuda

Code work for completion of MSc ICT. The original version of this project was submitted alongside a 15,000 word thesis paper and put into practice the theoretical CPU and GPGPU/GPU elements explored in the write-up. I continued to develop this project post-graduation.
0 stars 1 forks source link

matMultiFunc() kernel is skipped; potentially CUDA file separate compilation issue #3

Closed melsonGit closed 2 years ago

melsonGit commented 2 years ago

Issue

The matMultiFunc() kernel is skipped when called in matMultiCore(). This results in the program not functioning correctly according to output in matMultiCheck().

Additional Info

From my investigation, I've been able to identify that it may be a result of a CUDA file separate compilation issue.

Current workaround is to push CUDA code all into matMultiCore() (and for all other CUDA opertions), however this approach isn't in line with CpuArithmeticApp file structure (and I'd prefer consistency across the two programs).

melsonGit commented 2 years ago

Fixed in commit 37abde4

Closing.