Hzfengsy commented on PR #80: URL: https://github.com/apache/tvm-rfcs/pull/80#issuecomment-1159395985
Many thanks, @masahi! It's an outstanding contribution to push TVM perf to another stage. My experiments show async pipeline will speed up cutlass from 80T to 93T (on fp16 gemm, 3080). I am looking forward to the following PRs. One question: Can it speed up kernels with wmma intrinsic or those using CUDA cores? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
