tqchen commented on code in PR #16972:
URL: https://github.com/apache/tvm/pull/16972#discussion_r1593903178
##########
src/target/target_kind.cc:
##########
@@ -340,9 +340,9 @@ TVM_REGISTER_TARGET_KIND("rocm", kDLROCM)
.set_target_parser(UpdateROCmAttrs);
TVM_REGISTER_TARGET_KIND("opencl", kDLOpenCL)
- .add_attr_option<Integer>("max_threads_per_block", Integer(256))
+ .add_attr_option<Integer>("max_threads_per_block", Integer(1024))
.add_attr_option<Integer>("max_shared_memory_per_block", Integer(16384))
- .add_attr_option<Integer>("max_num_threads", Integer(256))
+ .add_attr_option<Integer>("max_num_threads", Integer(1024))
Review Comment:
we need to be a bit careful here, since i remember generic opencl may not
have as many threads. There are some low-end gpus that only comes with 256, is
this strictly needed?
--
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]
