OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity

Revert "[Inductor] Add Additional Configs for persistent+TMA version of Triton mm and addmm (#150587)"

Browse Source 
This reverts commit 99aeee2c5f.

Reverted https://github.com/pytorch/pytorch/pull/150587 on behalf of https://github.com/ZainRizvi due to Sorry but this is breaking internally (see D73410693). To validate your fixes internally, you can follow the instructions here: https://fburl.com/fixing-ghfirst-reverts ([comment](https://github.com/pytorch/pytorch/pull/150587#issuecomment-2821828926))
This commit is contained in:
PyTorch MergeBot 2025-04-22 16:15:55 +00:00
parent 4504910843
commit 3804aed32e
1 changed files with 0 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
torch/_inductor/template_heuristics.py
View File

@ -170,9 +170,6 @@ class BaseConfigHeuristic(metaclass=BaseHeuristicSingleton):
GemmConfig(128, 128, 128, 3, 8),
GemmConfig(128, 128, 128, 3, 4),
GemmConfig(128, 128, 64, 4, 8),
GemmConfig(128, 128, 64, 5, 8),
GemmConfig(256, 128, 64, 4, 8),
GemmConfig(128, 128, 64, 5, 4),
]
self.scaled_mm_configs: list[BaseConfig] = [