manupa-arm commented on a change in pull request #9950:
URL: https://github.com/apache/tvm/pull/9950#discussion_r791776941
##########
File path: src/driver/driver_api.cc
##########
@@ -47,6 +47,7 @@ TVM_REGISTER_PASS_CONFIG_OPTION("tir.disable_vectorize",
Bool);
TVM_REGISTER_PASS_CONFIG_OPTION("tir.is_entry_func", Bool);
TVM_REGISTER_PASS_CONFIG_OPTION("tir.add_lower_pass", Array<Array<ObjectRef>>);
TVM_REGISTER_PASS_CONFIG_OPTION("tir.debug_keep_trivial_loop", Bool);
+TVM_REGISTER_PASS_CONFIG_OPTION("tir.max_stack_alloca", Integer);
Review comment:
Hi @SebastianBoblestETAS ,
I guess you are anwsering the the question, I have asked in the title.
Let me cite an similar example :
https://github.com/apache/tvm/blob/6720d3593d4dac6015418d4b7e9ad875bbf0b0a2/tests/python/unittest/test_tir_transform_inject_double_buffer.py#L56-L60
We could use a similiar tvm.tir.stmt_functor.post_order_visit that counts
the remaining Allocate nodes.
For this thread,
What do you think about moving the PassConfig option to be part of
configuration of lower_tvm_builtin pass rather than here in the driver_api ?
(because this configuration now essential becomes a configuration of the pass)
--
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]
