yaoyaoding commented on PR #283: URL: https://github.com/apache/tvm-ffi/pull/283#issuecomment-3568481289
> The last two args of triton kernel are scratch memory, it need to be allocated using the size in metadata. See https://github.com/triton-lang/triton/blob/main/third_party/nvidia/backend/driver.py#L700-L714 you will also need to handle constexpr args that are not in the cubin args Cool, thanks for the information! -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
