From ed77e23b6808065c927e626f471d5bf588dd2cc2 Mon Sep 17 00:00:00 2001
From: PyTorch MergeBot <pytorchmergebot@users.noreply.github.com>
Date: Tue, 9 Sep 2025 00:21:05 +0000
Subject: [PATCH] Revert "[dynamo] Constant fold
 torch.autograd._profiler_enabled (#158482)"

This reverts commit d7e1b8b11d7430c7633dcad6f6596b5df8fa02f7.

Reverted https://github.com/pytorch/pytorch/pull/158482 on behalf of https://github.com/borgstrom due to NCCL hangs in S560336 ([comment](https://github.com/pytorch/pytorch/pull/158482#issuecomment-3268426781))
---
 test/dynamo/test_profiler.py     | 41 --------------------------------
 torch/_dynamo/trace_rules.py     |  2 +-
 torch/_dynamo/variables/torch.py |  1 -
 3 files changed, 1 insertion(+), 43 deletions(-)

diff --git a/test/dynamo/test_profiler.py b/test/dynamo/test_profiler.py
index c155613eb9e..61dc63ed2d5 100644
--- a/test/dynamo/test_profiler.py
+++ b/test/dynamo/test_profiler.py
@@ -192,47 +192,6 @@ class DynamoProfilerTests(torch._dynamo.test_case.TestCase):
             ],
         )
 
-    def test_profiler_enabled(self):
-        def fn(x):
-            x = torch.sin(x)
-            if torch.autograd._profiler_enabled():
-                return torch.cos(x)
-            else:
-                return torch.sigmoid(x)
-
-        opt_fn = torch.compile(fn, backend="eager", fullgraph=True)
-        x = torch.randn(4)
-
-        ref = fn(x)
-        res = opt_fn(x)
-        self.assertEqual(ref, res)
-
-        with torch.autograd.profiler.profile():
-            ref = fn(x)
-            res = opt_fn(x)
-            self.assertEqual(ref, res)
-
-    def test_profiler_record_function_ignore(self):
-        def fn(x):
-            x = torch.sin(x)
-            if torch.autograd._profiler_enabled():
-                with torch.autograd.profiler.record_function("dummy"):
-                    return torch.cos(x)
-            else:
-                return torch.sigmoid(x)
-
-        opt_fn = torch.compile(fn, backend="eager", fullgraph=True)
-        x = torch.randn(4)
-
-        ref = fn(x)
-        res = opt_fn(x)
-        self.assertEqual(ref, res)
-
-        with torch.autograd.profiler.profile():
-            ref = fn(x)
-            res = opt_fn(x)
-            self.assertEqual(ref, res)
-
 
 if __name__ == "__main__":
     from torch._dynamo.test_case import run_tests
diff --git a/torch/_dynamo/trace_rules.py b/torch/_dynamo/trace_rules.py
index 8dd5a309253..47ad8cda0c9 100644
--- a/torch/_dynamo/trace_rules.py
+++ b/torch/_dynamo/trace_rules.py
@@ -177,6 +177,7 @@ manual_torch_name_rule_map: dict[
     "torch.compiler.is_compiling": TorchInGraphFunctionVariable,
     "torch.compiler.is_dynamo_compiling": TorchInGraphFunctionVariable,
     "torch.compiler.is_exporting": TorchInGraphFunctionVariable,
+    "torch.autograd._profiler_enabled": SkipFunctionVariable,
     "torch._C._to_dlpack": SkipFunctionVariable,
     "torch.to_dlpack": SkipFunctionVariable,
     # We graph break on RNG state setters or getters like
@@ -2440,7 +2441,6 @@ torch_non_c_binding_in_graph_functions = dict.fromkeys(
         "torch.atleast_3d",
         "torch.autograd._calculate_shape",
         "torch.autograd._is_checkpoint_valid",
-        "torch.autograd._profiler_enabled",
         "torch.autograd._make_grads",
         "torch.autograd._register_py_tensor_class_for_device",
         "torch.autograd._tensor_or_tensors_to_tuple",
diff --git a/torch/_dynamo/variables/torch.py b/torch/_dynamo/variables/torch.py
index ee53ee4f568..bfebedc88d6 100644
--- a/torch/_dynamo/variables/torch.py
+++ b/torch/_dynamo/variables/torch.py
@@ -149,7 +149,6 @@ constant_fold_functions_need_guards = [
     torch.cuda.is_initialized,
     torch.xpu.current_device,
     torch.xpu.is_initialized,
-    torch.autograd._profiler_enabled,
 ]
 
 constant_fold_functions = [