jinhongyii opened a new pull request, #16094: URL: https://github.com/apache/tvm/pull/16094
1. PropagateSharding can now handle tuples. 2. A better sharding propagation algorithm in TIR which can cover all the operators in llama. There's no need to register how to propagate sharding through a Relax high-level operator. An easier way is to run LegalizeOps first, and then this pass will automatically analyze TIR and figure out a way to propagate. -- 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]
