Andrej Karpathy mentioned Programming Massively Parallel Processors by David B. Kirk, Wen-mei W. Hwu 2 times

I read this book and then I was surprised that I still understood so little of the kernels that started to appear as llm.c contributions, beating mine. It's a pretty good 101 intro. Learning CUDA is like that horse meme, all the learning resources you can find on the left, then… pic.twitter.com/C0k1WZqkQM

— Andrej Karpathy

It’s an intermediate level resource like I mentioned. I’d first read the book, then read this and write the whole thing from scratch, then on the second reading you learn a lot.

— Andrej Karpathy