[Fix] [torch.compile] Improve UUID system for custom passes (#15249)

Signed-off-by: luka <luka@neuralmagic.com>

tests/compile/test_pass_manager.py

 # SPDX-License-Identifier: Apache-2.0
-
-import pickle
+import copy
 
 import pytest
 import torch
@@ -10,32 +9,63 @@ from vllm.compilation.pass_manager import PostGradPassManager
 from vllm.config import CompilationConfig
 
 
+# dummy custom pass that doesn't inherit
 def simple_callable(graph: torch.fx.Graph):
     pass
 
 
-callable_uuid = CallableInductorPass(simple_callable,
-                                     InductorPass.hash_source(__file__))
+# Should fail to add directly to the pass manager
+def test_bad_callable():
+    config = CompilationConfig().pass_config
+
+    pass_manager = PostGradPassManager()
+    pass_manager.configure(config)
+
+    with pytest.raises(AssertionError):
+        pass_manager.add(simple_callable)  # noqa, type wrong on purpose
+
+
+# Pass that inherits from InductorPass
+class ProperPass(InductorPass):
+
+    def __call__(self, graph: torch.fx.graph.Graph) -> None:
+        pass
 
 
 @pytest.mark.parametrize(
-    "works, callable",
+    "callable",
     [
-        (False, simple_callable),
-        (True, callable_uuid),
-        (True, CallableInductorPass(simple_callable)),
+        ProperPass(),
+        # Can also wrap callables in CallableInductorPass for compliance
+        CallableInductorPass(simple_callable),
+        CallableInductorPass(simple_callable,
+                             InductorPass.hash_source(__file__))
     ],
 )
-def test_pass_manager(works: bool, callable):
+def test_pass_manager_uuid(callable):
     config = CompilationConfig().pass_config
 
     pass_manager = PostGradPassManager()
     pass_manager.configure(config)
 
-    # Try to add the callable to the pass manager
-    if works:
+    # Check that UUID is different if the same pass is added 2x
     pass_manager.add(callable)
-        pickle.dumps(pass_manager)
-    else:
-        with pytest.raises(AssertionError):
+    uuid1 = pass_manager.uuid()
     pass_manager.add(callable)
+    uuid2 = pass_manager.uuid()
+    assert uuid1 != uuid2
+
+    # UUID should be the same as the original one,
+    # as we constructed in the same way.
+    pass_manager2 = PostGradPassManager()
+    pass_manager2.configure(config)
+    pass_manager2.add(callable)
+    assert uuid1 == pass_manager2.uuid()
+
+    # UUID should be different due to config change
+    config2 = copy.deepcopy(config)
+    config2.enable_fusion = not config2.enable_fusion
+    pass_manager3 = PostGradPassManager()
+    pass_manager3.configure(config2)
+    pass_manager3.add(callable)
+    assert uuid1 != pass_manager3.uuid()

vllm/compilation/inductor_pass.py

 # SPDX-License-Identifier: Apache-2.0
 
 import hashlib
+import importlib.metadata
 import inspect
+import json
 import types
-from abc import ABC, abstractmethod
-from typing import Any, Callable, Optional, Union
+from typing import Any, Callable, Dict, Optional, Union
 
 import torch
+from packaging.version import Version
 from torch import fx
 
+if Version(importlib.metadata.version('torch')) >= Version("2.6"):
+    from torch._inductor.custom_graph_pass import CustomGraphPass
+else:
+    # CustomGraphPass is not present in 2.5 or lower, import our version
+    from .torch25_custom_graph_pass import (  # noqa: yapf
+        Torch25CustomGraphPass as CustomGraphPass)
 
-class InductorPass(ABC):
-    """
-    General custom inductor pass interface.
-    """
 
-    @abstractmethod
-    def __call__(self, graph: torch.fx.Graph):
+class InductorPass(CustomGraphPass):
     """
-        Execute the pass on the given graph.
+    A custom graph pass that uses a hash of its source as the UUID.
+    This is defined as a convenience and should work in most cases.
     """
-        raise NotImplementedError
 
     def uuid(self) -> Any:
         """
@@ -48,7 +51,16 @@ class InductorPass(ABC):
             else:
                 src_str = inspect.getsource(src.__class__)
             hasher.update(src_str.encode("utf-8"))
-        return hasher.digest()
+        return hasher.hexdigest()
+
+    @staticmethod
+    def hash_dict(dict_: Dict[Any, Any]):
+        """
+        Utility method to hash a dictionary, can alternatively be used for uuid.
+        :return: A sha256 hash of the json rep of the dictionary.
+        """
+        encoded = json.dumps(dict_, sort_keys=True).encode("utf-8")
+        return hashlib.sha256(encoded).hexdigest()
 
 
 class CallableInductorPass(InductorPass):
@@ -61,25 +73,10 @@ class CallableInductorPass(InductorPass):
                  callable: Callable[[fx.Graph], None],
                  uuid: Optional[Any] = None):
         self.callable = callable
-        if uuid is None:
-            uuid = InductorPass.hash_source(callable)
-        self._uuid = uuid
+        self._uuid = self.hash_source(callable) if uuid is None else uuid
 
     def __call__(self, graph: torch.fx.Graph):
         self.callable(graph)
 
     def uuid(self) -> Any:
         return self._uuid
-
-    def __getstate__(self):
-        """
-        Pickling occurs in the Inductor code cache if a pass is not given to
-        the pass manager but is instead directly added to config as a pass.
-        See PostGradPassManager for more.
-
-        TODO(torch==2.6), use the `uuid` method in CustomGraphPass instead.
-        """
-        return self._uuid
-
-    def __setstate__(self, state):
-        raise ValueError("Cannot unpickle CallableInductorPass")

vllm/compilation/pass_manager.py

 # SPDX-License-Identifier: Apache-2.0
 
-from typing import Any, Dict, List
+from typing import List
 
-import torch
 from torch import fx as fx
 
 from vllm.config import CompilationConfig
@@ -10,29 +9,18 @@ from vllm.logger import init_logger
 
 from .fix_functionalization import FixFunctionalizationPass
 from .fusion import FusionPass
-from .inductor_pass import InductorPass
+from .inductor_pass import CustomGraphPass, InductorPass
 from .noop_elimination import NoOpEliminationPass
 
 logger = init_logger(__name__)
 
 
-class PlaceHolder:
-    pass
-
-
-if torch.__version__ < "2.6":
-    Parent = PlaceHolder  # type: ignore
-else:
-    Parent = torch._inductor.custom_graph_pass.CustomGraphPass  # type: ignore
-
-
-class PostGradPassManager(Parent):
+class PostGradPassManager(CustomGraphPass):
     """
     The pass manager for post-grad passes.
     It handles configuration, adding custom passes, and running passes.
-    It also supports pickling, which is used by the Inductor code cache.
-    TODO(torch==2.6), use CustomGraphPass
-    (torch._inductor.custom_graph_pass.CustomGraphPass)
+    It supports uuid for the Inductor code cache. That includes torch<2.6
+    support using pickling (in .inductor_pass.CustomGraphPass).
 
     The order of the post-grad post-passes is:
     1. passes (constructor parameter)
@@ -67,27 +55,13 @@ class PostGradPassManager(Parent):
         self.passes.append(pass_)
 
     def uuid(self):
-        return self.__getstate__()
-
-    def __getstate__(self) -> Dict[str, List[Any]]:
         """
-        Custom pickling for the pass manager, as some passes cannot be pickled.
-        Pickling occurs because the pass manager is set as the value of
-        `config["post_grad_custom_post_pass"]` in the Inductor config.
-        The config is pickled to act as a key in the Inductor code cache.
-        Any other passes in the config are pickled as well.
-
-        TODO(torch==2.6), use the `uuid` method in CustomGraphPass instead.
+        The PostGradPassManager is set as a custom pass in the Inductor and
+        affects compilation caching. Its uuid depends on the UUIDs of all
+        dependent passes and the pass config. See InductorPass for more info.
         """
         state = {"pass_config": self.pass_config.uuid(), "passes": []}
         for pass_ in self.passes:
             state["passes"].append(pass_.uuid())
         state["passes"].append(self.fix_functionalization.uuid())
-        return state
-
-    def __setstate__(self, state):
-        """
-        Do not allow unpickling of the pass manager.
-        If this is needed in the future, it should properly pickle the passes.
-        """
-        raise ValueError("Cannot unpickle PostGradPassManager")
+        return InductorPass.hash_dict(state)

vllm/compilation/torch25_custom_graph_pass.py

+# SPDX-License-Identifier: Apache-2.0
+from abc import ABC, abstractmethod
+from typing import Any, Optional
+
+import torch
+
+
+class Torch25CustomGraphPass(ABC):  # noqa (redefinition)
+    """
+    This class replaces CustomGraphPass from torch==2.6 when using torch<2.6.
+    It conforms to the 2.6 interface but also supports pickling, as that's what
+    the inductor code cache uses to determine the cache key before 2.6.
+    (in 2.6 and above, uuid() is used.)
+
+    Subclasses can just "pretend" that uuid is used.
+    """
+
+    @abstractmethod
+    def __call__(self, graph: torch.fx.graph.Graph) -> None:
+        """
+        Implementation of the custom pass.
+        """
+
+    @abstractmethod
+    def uuid(self) -> Optional[Any]:
+        """
+        Return an ID to uniquely identify your custom pass implementation.
+        Return None to skip inductor code caching entirely.
+        """
+
+    def __getstate__(self):
+        """
+        Pickling is used instead of uuid() in torch<2.6. Just return uuid()
+         to enable subclasses to only have to implement uuid.
+        """
+        return self.uuid()
+
+    def __setstate__(self, state):
+        raise ValueError("Cannot unpickle CustomGraphPass because pickling"
+                         " is used for cache key uuid. Use torch>=2.6 with"
+                         " native uuid support for custom passes.")

vllm/config.py

@@ -4,6 +4,7 @@ import ast
 import copy
 import enum
 import hashlib
+import importlib.metadata
 import json
 import sys
 import warnings
@@ -17,6 +18,7 @@ from typing import (TYPE_CHECKING, Any, Callable, ClassVar, Final, Literal,
                     Optional, Protocol, Union)
 
 import torch
+from packaging.version import Version
 from pydantic import BaseModel, Field, PrivateAttr
 from torch.distributed import ProcessGroup, ReduceOp
 from transformers import PretrainedConfig
@@ -52,8 +54,6 @@ if TYPE_CHECKING:
 else:
     QuantizationConfig = None
 
-from packaging.version import Version
-
 logger = init_logger(__name__)
 
 # This value is chosen to have a balance between ITL and TTFT. Note it is
@@ -3088,8 +3088,7 @@ class CompilationConfig(BaseModel):
             compilation.
             """
             dict_ = self.model_dump(include={"enable_fusion", "enable_noop"})
-            encoded = json.dumps(dict_, sort_keys=True).encode("utf-8")
-            return hashlib.sha256(encoded).digest()
+            return InductorPass.hash_dict(dict_)
 
         def model_post_init(self, __context: Any) -> None:
             if not self.enable_noop and self.enable_fusion:
@@ -3178,7 +3177,7 @@ class CompilationConfig(BaseModel):
         #    and it is not yet a priority. RFC here:
         #    https://github.com/vllm-project/vllm/issues/14703
 
-        if Version(torch.__version__) >= Version("2.6"):
+        if Version(importlib.metadata.version('torch')) >= Version("2.6"):
             KEY = 'enable_auto_functionalized_v2'
             if KEY not in self.inductor_compile_config:
                 self.inductor_compile_config[KEY] = False