jcf94 commented on a change in pull request #8527:
URL: https://github.com/apache/tvm/pull/8527#discussion_r675935513
##########
File path: src/relay/transforms/combine_parallel_batch_matmul.cc
##########
@@ -68,6 +68,13 @@ class ParallelBatchMatmulCombiner : public
ParallelOpCombiner {
// shape[2] is the contraction axis and automatically consistent
// if it were valid batch_matmul ops
+ // This pass only support the original NT format now
+ // TODO(jcf94): Add full support of layout format
+ if (attrs_a->transpose_a == true || attrs_a->transpose_b == false ||
+ attrs_b->transpose_a == true || attrs_b->transpose_b == false) {
+ return false;
Review comment:
Warning added.
--
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]
