[Bugfix] Fix triton import with local TritonPlaceholder (#17446)

Signed-off-by: Mengqing Cao <cmq0113@163.com>

vllm/triton_utils/__init__.py

@@ -1,5 +1,13 @@
 # SPDX-License-Identifier: Apache-2.0
 
-from vllm.triton_utils.importing import HAS_TRITON
+from vllm.triton_utils.importing import (HAS_TRITON, TritonLanguagePlaceholder,
+                                         TritonPlaceholder)
 
-__all__ = ["HAS_TRITON"]
+if HAS_TRITON:
+    import triton
+    import triton.language as tl
+else:
+    triton = TritonPlaceholder()
+    tl = TritonLanguagePlaceholder()
+
+__all__ = ["HAS_TRITON", "triton", "tl"]

vllm/triton_utils/importing.py

@@ -16,32 +16,34 @@ if not HAS_TRITON:
     logger.info("Triton not installed or not compatible; certain GPU-related"
                 " functions will not be available.")
 
-    class TritonPlaceholder(types.ModuleType):
 
-        def __init__(self):
-            super().__init__("triton")
-            self.jit = self._dummy_decorator("jit")
-            self.autotune = self._dummy_decorator("autotune")
-            self.heuristics = self._dummy_decorator("heuristics")
-            self.language = TritonLanguagePlaceholder()
-            logger.warning_once(
-                "Triton is not installed. Using dummy decorators. "
-                "Install it via `pip install triton` to enable kernel"
-                "compilation.")
+class TritonPlaceholder(types.ModuleType):
 
-        def _dummy_decorator(self, name):
+    def __init__(self):
+        super().__init__("triton")
+        self.jit = self._dummy_decorator("jit")
+        self.autotune = self._dummy_decorator("autotune")
+        self.heuristics = self._dummy_decorator("heuristics")
+        self.language = TritonLanguagePlaceholder()
+        logger.warning_once(
+            "Triton is not installed. Using dummy decorators. "
+            "Install it via `pip install triton` to enable kernel"
+            " compilation.")
 
-            def decorator(func=None, **kwargs):
-                if func is None:
-                    return lambda f: f
-                return func
+    def _dummy_decorator(self, name):
 
-            return decorator
+        def decorator(*args, **kwargs):
+            if args and callable(args[0]):
+                return args[0]
+            return lambda f: f
 
-    class TritonLanguagePlaceholder(types.ModuleType):
+        return decorator
 
-        def __init__(self):
-            super().__init__("triton.language")
-            self.constexpr = None
-            self.dtype = None
-            self.int64 = None
+
+class TritonLanguagePlaceholder(types.ModuleType):
+
+    def __init__(self):
+        super().__init__("triton.language")
+        self.constexpr = None
+        self.dtype = None
+        self.int64 = None