mbaret commented on a change in pull request #7925: URL: https://github.com/apache/tvm/pull/7925#discussion_r634334158
########## File path: python/tvm/tir/transform/inject_rolling_buffer.py ########## @@ -0,0 +1,214 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. +"""Inject rolling buffers through a TIR transformation.""" +# pylint: disable=invalid-name,unused-argument,inconsistent-return-statements +from collections import defaultdict + +import tvm +from tvm import arith + + +def InjectRollingBuffer(): Review comment: The necessary TIR snippets would be very large for this particular pass. I have a before/after TIR example in the tests so hopefully these could be used as a reference. In lieu of TIR in the documentation, I've improved it somewhat and included a link to the RFC which has a far more detailed explanation of the pass + diagrams. Let me know if you think this is sufficient. -- 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. For queries about this service, please contact Infrastructure at: [email protected]
