vinx13 commented on PR #15839: URL: https://github.com/apache/tvm/pull/15839#issuecomment-1740116164
That's correct. It's relying on the current behavior of `StaticBlockPlanMemory` that doesn't rewrite the input/output tensors. So only tensors from `R.builtin.alloc_tensor` can be returned (similar to the current implementation of this pass, `StaticBlockPlanMemory` doesn't rewrite the body of `SeqExpr`). -- 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]
