vinx13 commented on code in PR #14608:
URL: https://github.com/apache/tvm/pull/14608#discussion_r1164667374
##########
python/tvm/relax/backend/contrib/cutlass.py:
##########
@@ -244,6 +245,14 @@ def attention_patterns():
"cutlass.attention_bias",
*make_attention_pattern(with_bias=True),
),
+ (
+ "cutlass.stacked_attention",
Review Comment:
does the order here matter? If we have a subgraph containing both reshape
and attention, will `cutlass.attention` that matches only a single attention
operation be selected first?
##########
python/tvm/relax/backend/contrib/cutlass.py:
##########
@@ -244,6 +245,14 @@ def attention_patterns():
"cutlass.attention_bias",
*make_attention_pattern(with_bias=True),
),
+ (
+ "cutlass.stacked_attention",
Review Comment:
does the order of patterns here matter? If we have a subgraph containing
both reshape and attention, will `cutlass.attention` that matches only a single
attention operation be selected first?
--
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]
