Viewing profile — tspeterkim

Detailed writeup here: https://tspeterkim.github.io/posts/mixed-precision-from-scra...

I tried this out today. While it works (no longer a pre-split step required), it makes the CUDA kernel run ridiculously slow. I believe it's because of the while loop: while (i Com…

Please do. My hope with this blog was to rile up other CUDA enthusiasts. Making them wanna bring in bigger and better hardware that I don't have access to. + Someone in the CUDA MO…

This is the possible optimization that I mention at the end of the blog - using a private map for each thread block. The catch is that this map must fit in shared memory, which is …

I misspoke. (got confused with the key limit in my link above) Atomics work up to 128-bits ( https://docs.nvidia.com/cuda/cuda-c-programming-guide/#atomi... ). Regardless, it's sti…

for (int i = 0; buffer[offset+i] != '\n'; i++) { This would only process the current line, though. Here, each thread processes ~split_size bytes (multiple lines). Even if were to r…

By "launcher", do you mean the CUDA kernel? How can it avoid accessing the input data since it needs access to the characters based on the offsets? I also already pass these offset…

The PMPP book is great. I reread the histogram chapter after finishing the blog, and realized I could use privatization. You got me! By coarsening, do you mean making the threads h…

So performance would increase since hashing is faster than binary-searching. However, the problem of collisions across threads and dealing with concurrent map key insertions still …

Agreed. I tried reducing across cities first. The problem was that the work of gathering all the temperatures for each city (before I could launch the reduction CUDA kernels) requi…

Thanks Daniel. The main blocker is me not able to fully grasp the backward pass. (trying to understand Appendix B.2 in the original paper) I need to get more comfortable with matri…