LeiWang1999 commented on PR #14329: URL: https://github.com/apache/tvm/pull/14329#issuecomment-1493344819
I think this pr is ready to be merged in if I resolved the last lint ci failure (actually the link checker ci told me which file that do not follow the format rule, but I didn't find any scirpt that can auto-format the file, like clang-format or pep8, I may need some helps. @junrushao @adstraw I have a tricky fix of the dma lowering conflict between cuda backend and Hexagon, in the latest commit `e08fff9aeca293325d1c81734dea2543ddd7ce26` and `` non contiguous memory access will not be a illegal behavior and it will fall back to a case that cuda required, now it seems that this chage works, but I think in the long terms view, we should separate the asynchronous copy of gpu from the dma of cpu -- 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]
