From ce216c80dc413e404b7f6f63a151436253c8b837 Mon Sep 17 00:00:00 2001
From: Qubitium <417764+Qubitium@users.noreply.github.com>
Date: Sun, 24 Mar 2024 01:15:16 +0800
Subject: [PATCH] Cleanup codebase: removed unnecessary code/logic (#298)

---
 python/sglang/srt/managers/router/manager.py    | 8 ++++++--
 python/sglang/srt/managers/router/model_rpc.py  | 1 -
 python/sglang/srt/managers/tokenizer_manager.py | 7 ++-----
 3 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/python/sglang/srt/managers/router/manager.py b/python/sglang/srt/managers/router/manager.py
index 4dc7d1f1c..7562a3022 100644
--- a/python/sglang/srt/managers/router/manager.py
+++ b/python/sglang/srt/managers/router/manager.py
@@ -41,12 +41,16 @@ class RouterManager:
                 self.send_to_detokenizer.send_pyobj(obj)
 
             # async sleep for receiving the subsequent request and avoiding cache miss
+            slept = False
             if len(out_pyobjs) != 0:
                 has_finished = any([obj.finished for obj in out_pyobjs])
                 if has_finished:
-                    await asyncio.sleep(self.extend_dependency_time)
+                    if self.extend_dependency_time > 0:
+                        slept = True
+                        await asyncio.sleep(self.extend_dependency_time)
 
-            await asyncio.sleep(0.0006)
+            if not slept:
+                await asyncio.sleep(0.0006)
 
     async def loop_for_recv_requests(self):
         while True:
diff --git a/python/sglang/srt/managers/router/model_rpc.py b/python/sglang/srt/managers/router/model_rpc.py
index 68f8423ed..6468c2d5f 100644
--- a/python/sglang/srt/managers/router/model_rpc.py
+++ b/python/sglang/srt/managers/router/model_rpc.py
@@ -90,7 +90,6 @@ class ModelRpcServer(rpyc.Service):
                 tokenizer_mode=server_args.tokenizer_mode,
                 trust_remote_code=server_args.trust_remote_code,
             )
-        self.eos_token_id = self.tokenizer.eos_token_id
         self.max_total_num_token = self.model_runner.max_total_num_token
         self.max_num_running_seq = self.max_total_num_token // 2
         self.max_prefill_num_token = max(
diff --git a/python/sglang/srt/managers/tokenizer_manager.py b/python/sglang/srt/managers/tokenizer_manager.py
index 814141e1f..7947ca2ff 100644
--- a/python/sglang/srt/managers/tokenizer_manager.py
+++ b/python/sglang/srt/managers/tokenizer_manager.py
@@ -36,7 +36,6 @@ class ReqState:
     out_list: List
     finished: bool
     event: asyncio.Event
-    lock: asyncio.Lock
 
 
 global global_processor
@@ -178,9 +177,8 @@ class TokenizerManager:
             )
             self.send_to_router.send_pyobj(tokenized_obj)
 
-            lock = asyncio.Lock()
             event = asyncio.Event()
-            state = ReqState([], False, event, lock)
+            state = ReqState([], False, event)
             self.rid_to_state[rid] = state
 
             while True:
@@ -221,9 +219,8 @@ class TokenizerManager:
                 )
                 self.send_to_router.send_pyobj(tokenized_obj)
 
-                lock = asyncio.Lock()
                 event = asyncio.Event()
-                state = ReqState([], False, event, lock)
+                state = ReqState([], False, event)
                 self.rid_to_state[rid] = state
 
             output_list = []