Use the manifold cache backend as the default (#144773)

Summary: Pull Request resolved: pytorch/pytorch#144773 Reviewed By: jamesjwu Differential Revision: D68155591 fbshipit-source-id: 13bf79938ca17731937dc086b0b8fac474cf8251
pytorch · Jan 15, 2025 · 68c5409 · 68c5409
1 parent d083577
commit 68c5409
Showing 1 changed file with 2 additions and 6 deletions.
diff --git a/userbenchmark/dynamo/dynamobench/_dynamo/utils.py b/userbenchmark/dynamo/dynamobench/_dynamo/utils.py
@@ -1376,14 +1376,10 @@ def record_compilation_metrics(
 ):
     if torch._inductor.utils.should_use_remote_fx_graph_cache():
         try:
-            from torch._inductor.fb.remote_cache import (
-                FbRemoteFxGraphCache,
-                REMOTE_CACHE_VERSION,
-            )
+            from torch._inductor.fb.remote_cache import REMOTE_CACHE_VERSION
 
             remote_cache_version = REMOTE_CACHE_VERSION
-            backend = FbRemoteFxGraphCache.get_remote_backend()
-            inductor_fx_remote_cache_backend_type = type(backend).__name__
+            inductor_fx_remote_cache_backend_type = "_ManifoldCache"
         except ModuleNotFoundError:
             remote_cache_version = None
             inductor_fx_remote_cache_backend_type = None