frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

Improve Scheduler init duration (#161491)

Browse Source 
Early exit merge_loops() if config.loop_ordering_after_fusion is false.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/161491
Approved by: https://github.com/jansel
This commit is contained in:
Son Nguyen
2025-08-28 00:27:51 +00:00
committed by PyTorch MergeBot
parent fd60117051
commit 9b02435e9f
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
torch/_inductor/scheduler.py
View File

@ -2755,10 +2755,10 @@ class Scheduler:
node.max_order = order
def merge_loops(self) -> None:
for node in self.nodes:
if not config.loop_ordering_after_fusion:
continue
return
for node in self.nodes:
# Even for CPU, if we are using the halide backend, we still need
# the merge loops steps below
if not isinstance(node, (SchedulerNode, FusedSchedulerNode)) or (