Nice to see other folks working on adding TensorCore support into TVM, we have also been working on enhancing TVM to incorporate TensorCore schedule support. If my understanding is correct, @Hzfengsy your solution is based on extending TVM's intrinsic while our solution put most of the complexity into TVM IR passes so that at python code level, we don't need to consider too much about TensorCore stuffs, rather the TVM core will take care of most of the TensorCore pattern recognition and schedule customization work, such as col-major/row-major pattern recognition and etc.
@minmin.sun may provide more comments. Thanks -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/dmlc/tvm/issues/4052#issuecomment-537808119
