enable piecewise cuda graph for prefill server (#13377)

Co-authored-by: serverance.fu <serverance.fu@temu.com>

python/sglang/srt/server_args.py

@@ -1825,8 +1825,12 @@ class ServerArgs:
 
             self.disaggregation_prefill_pp = self.pp_size
             self.validate_disagg_tp_size(self.tp_size, self.disaggregation_decode_tp)
-            self.disable_cuda_graph = True
-            logger.warning("Cuda graph is disabled for prefill server")
+
+            if not self.enable_piecewise_cuda_graph:
+                self.disable_cuda_graph = True
+                logger.warning(
+                    "Cuda graph is disabled for prefill server when piecewise cuda graph is not enabled."
+                )
 
     def _handle_tokenizer_batching(self):
         if self.enable_tokenizer_batch_encode and self.enable_dynamic_batch_tokenizer:

test/srt/test_disaggregation_piecewise_cuda_graph.py

@@ -0,0 +1,86 @@
+import unittest
+from types import SimpleNamespace
+
+from sglang.test.few_shot_gsm8k import run_eval as run_eval_few_shot_gsm8k
+from sglang.test.test_disaggregation_utils import TestDisaggregationBase
+from sglang.test.test_utils import (
+    DEFAULT_MODEL_NAME_FOR_TEST,
+    DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
+    popen_launch_pd_server,
+)
+
+
+class TestDisaggregationPiecewiseCudaGraph(TestDisaggregationBase):
+    """Test piecewise CUDA graph support in disaggregation prefill server"""
+
+    @classmethod
+    def setUpClass(cls):
+        super().setUpClass()
+        cls.model = DEFAULT_MODEL_NAME_FOR_TEST
+
+        # Start servers
+        cls.start_prefill()
+        cls.start_decode()
+
+        # Wait for both to be ready
+        cls.wait_server_ready(cls.prefill_url + "/health")
+        cls.wait_server_ready(cls.decode_url + "/health")
+
+        cls.launch_lb()
+
+    @classmethod
+    def start_prefill(cls):
+        prefill_args = [
+            "--trust-remote-code",
+            "--disaggregation-mode",
+            "prefill",
+            "--tp",
+            "1",
+            "--enable-piecewise-cuda-graph",
+        ]
+        prefill_args += cls.transfer_backend + cls.rdma_devices
+        cls.process_prefill = popen_launch_pd_server(
+            cls.model,
+            cls.prefill_url,
+            timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
+            other_args=prefill_args,
+        )
+
+    @classmethod
+    def start_decode(cls):
+        decode_args = [
+            "--trust-remote-code",
+            "--disaggregation-mode",
+            "decode",
+            "--tp",
+            "1",
+            "--base-gpu-id",
+            "1",
+        ]
+        decode_args += cls.transfer_backend + cls.rdma_devices
+        cls.process_decode = popen_launch_pd_server(
+            cls.model,
+            cls.decode_url,
+            timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
+            other_args=decode_args,
+        )
+
+    def test_gsm8k_accuracy(self):
+        """Verify that piecewise cuda graph works correctly in prefill server"""
+        args = SimpleNamespace(
+            num_shots=5,
+            data_path=None,
+            num_questions=200,
+            max_new_tokens=512,
+            parallel=128,
+            host=f"http://{self.base_host}",
+            port=int(self.lb_port),
+        )
+        metrics = run_eval_few_shot_gsm8k(args)
+        print(f"GSM8K accuracy with piecewise cuda graph: {metrics['accuracy']:.3f}")
+
+        self.assertGreater(metrics["accuracy"], 0.62)
+
+
+if __name__ == "__main__":
+    unittest.main()