Eager Compiler for Torch Compile (#11803)

Signed-off-by: Oasis-Git <ayw.sirius19@gmail.com>

python/sglang/srt/compilation/backend.py

@@ -17,24 +17,30 @@ from torch._dispatch.python import enable_python_dispatcher
 
 from sglang.srt.compilation.compilation_config import CompilationConfig
 from sglang.srt.compilation.compilation_counter import compilation_counter
-from sglang.srt.compilation.compiler_interface import InductorAdaptor
+from sglang.srt.compilation.compiler_interface import EagerAdapter, InductorAdaptor
 from sglang.srt.compilation.cuda_piecewise_backend import CUDAPiecewiseBackend
 from sglang.srt.compilation.pass_manager import PostGradPassManager
 
 logger = logging.getLogger(__name__)
 
 
-def make_compiler():
-    return InductorAdaptor()
+def make_compiler(config: CompilationConfig):
+    if config.compiler == "eager":
+        return EagerAdapter()
+    elif config.compiler == "inductor":
+        return InductorAdaptor()
+    else:
+        raise ValueError(f"Unknown compiler: {config.compiler}")
 
 
 class CompilerManager:
     def __init__(
         self,
+        config: CompilationConfig,
     ):
         self.cache = dict()
         self.is_cache_updated = False
-        self.compiler = make_compiler()
+        self.compiler = make_compiler(config)
 
     def compute_hash(self):
         return self.compiler.compute_hash()
@@ -348,7 +354,7 @@ class SGLangBackend:
         self.sym_tensor_indices = []
         self.input_buffers = []
 
-        self.compiler_manager = CompilerManager()
+        self.compiler_manager = CompilerManager(config)
         self.inductor_config = {
             "enable_auto_functionalized_v2": False,
         }

python/sglang/srt/compilation/compilation_config.py

@@ -5,9 +5,10 @@ from typing import List
 
 # TODO(Yuwei): support better compile config support
 class CompilationConfig:
-    def __init__(self, capture_sizes: List[int]):
+    def __init__(self, capture_sizes: List[int], compiler: str = "eager"):
         self.traced_files = set()
         self.capture_sizes = capture_sizes
+        self.compiler = compiler
 
     def add_traced_file(self, file_path: str):
         self.traced_files.add(file_path)

python/sglang/srt/compilation/compiler_interface.py

@@ -475,3 +475,29 @@ def set_inductor_config(config, runtime_shape):
         # can be beneficial
         config["max_autotune"] = True
         config["coordinate_descent_tuning"] = True
+
+
+class EagerAdapter(CompilerInterface):
+    name = "eager"
+
+    def compile(
+        self,
+        graph: fx.GraphModule,
+        example_inputs: list[Any],
+        compiler_config: dict[str, Any],
+        runtime_shape: Optional[int] = None,
+        key: Optional[str] = None,
+        num_graphs: int = 1,
+    ) -> tuple[Optional[Callable], Optional[Any]]:
+        return graph, None
+
+    def load(
+        self,
+        handle: Any,
+        graph: fx.GraphModule,
+        example_inputs: list[Any],
+        graph_index: int,
+        runtime_shape: Optional[int] = None,
+        num_graphs: int = 1,
+    ) -> Callable:
+        raise NotImplementedError("eager compilation is not supported")

python/sglang/srt/compilation/cuda_piecewise_backend.py

@@ -9,6 +9,7 @@ from unittest.mock import patch
 import torch
 import torch.fx as fx
 
+import sglang.srt.compilation.weak_ref_tensor_jit  # noqa: F401
 from sglang.srt.compilation.compilation_config import CompilationConfig
 from sglang.srt.compilation.compilation_counter import compilation_counter
 

python/sglang/srt/model_executor/piecewise_cuda_graph_runner.py

@@ -103,9 +103,10 @@ def _to_torch(model: torch.nn.Module, reverse: bool, num_tokens: int):
 
 
 @contextmanager
-def patch_model(model: torch.nn.Module):
+def patch_model(model: torch.nn.Module, compiler: str):
     try:
-        _to_torch(model, reverse=False, num_tokens=16)
+        if compiler != "eager":
+            _to_torch(model, reverse=False, num_tokens=16)
         yield model
     finally:
         _to_torch(model, reverse=True, num_tokens=16)
@@ -144,8 +145,13 @@ class PiecewiseCudaGraphRunner:
         assert (
             self.model_runner.server_args.piecewise_cuda_graph_tokens is not None
         ), "piecewise_cuda_graph_tokens is not set"
+        assert self.model_runner.server_args.piecewise_cuda_graph_compiler in [
+            "eager",
+            "inductor",
+        ], "By now, only eager and inductor are supported for piecewise cuda graph compiler."
         self.compile_config = CompilationConfig(
-            self.model_runner.server_args.piecewise_cuda_graph_tokens
+            self.model_runner.server_args.piecewise_cuda_graph_tokens,
+            self.model_runner.server_args.piecewise_cuda_graph_compiler,
         )
 
         # Batch sizes to capture
@@ -179,7 +185,9 @@ class PiecewiseCudaGraphRunner:
         # Set graph pool id globally to be able to use symmetric memory
         set_graph_pool_id(get_global_graph_memory_pool())
 
-        with patch_model(self.model_runner.model.model) as patched_model:
+        with patch_model(
+            self.model_runner.model.model, self.compile_config.compiler
+        ) as patched_model:
             install_torch_compiled(
                 patched_model,
                 fullgraph=True,
@@ -191,14 +199,14 @@ class PiecewiseCudaGraphRunner:
             with set_compiled(True):
                 self.warmup_and_capture()
 
-        # Capture
-        try:
-            with model_capture_mode():
-                self.capture()
-        except RuntimeError as e:
-            raise Exception(
-                f"Capture cuda graph failed: {e}\n{PIECEWISE_CUDA_GRAPH_CAPTURE_FAILED_MSG}"
-            )
+            # Capture
+            try:
+                with model_capture_mode():
+                    self.capture()
+            except RuntimeError as e:
+                raise Exception(
+                    f"Capture cuda graph failed: {e}\n{PIECEWISE_CUDA_GRAPH_CAPTURE_FAILED_MSG}"
+                )
 
         self.raw_num_tokens = 0
 

python/sglang/srt/server_args.py

@@ -436,6 +436,7 @@ class ServerArgs:
     torch_compile_max_bs: int = 32
     piecewise_cuda_graph_max_tokens: int = 4096
     piecewise_cuda_graph_tokens: Optional[List[int]] = None
+    piecewise_cuda_graph_compiler: str = "eager"
     torchao_config: str = ""
     enable_nan_detection: bool = False
     enable_p2p_check: bool = False
@@ -2815,6 +2816,13 @@ class ServerArgs:
             default=ServerArgs.piecewise_cuda_graph_tokens,
             help="Set the list of tokens when using piecewise cuda graph.",
         )
+        parser.add_argument(
+            "--piecewise-cuda-graph-compiler",
+            type=str,
+            default=ServerArgs.piecewise_cuda_graph_compiler,
+            help="Set the compiler for piecewise cuda graph. Choices are: eager, inductor.",
+            choices=["eager", "inductor"],
+        )
         parser.add_argument(
             "--torch-compile-max-bs",
             type=int,