comaniac commented on a change in pull request #6919:
URL: https://github.com/apache/incubator-tvm/pull/6919#discussion_r523711459
##########
File path: python/tvm/auto_scheduler/relay_integration.py
##########
@@ -214,6 +214,8 @@ def auto_schedule_topi(outs):
if env is None: # in the final build mode
state = DispatchContext.current.query(tvm.target.Target.current(), key)
if state is None:
+ if "gpu" in tvm.target.Target.current().keys:
+ raise RuntimeError("Cannot compile for GPU targets if no valid
schedule is found.")
Review comment:
The task extraction PR fixes this issue, but it's fine to let this in
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.
For queries about this service, please contact Infrastructure at:
[email protected]
