comaniac commented on a change in pull request #8669:
URL: https://github.com/apache/tvm/pull/8669#discussion_r685519889
##########
File path: python/tvm/topi/x86/dense_alter_op.py
##########
@@ -39,6 +40,17 @@ def _alter_dense_layout(attrs, inputs, tinfos, out_type):
relay.op.get("nn.dense"), attrs, tinfos, out_type, target
)
workload = autotvm.task.get_workload(outs)
+
+ data_type =
relay.transform.InferType()(IRModule.from_expr(inputs[0]))["main"].body.checked_type
Review comment:
What I meant was this line creates a new module with only `%1`. I'm not
sure if the function body will be included in this new module or just the
function symbol (i.e., `%global_var`). The latter case will crash InferType due
to fail to loop up the function body. If this is not the case then we're good.
--
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]
