Fix memory leak during abort (#2238)

.github/workflows/pr-test.yml

@@ -50,7 +50,7 @@ jobs:
         timeout-minutes: 25
         run: |
           cd test/srt
-          python3 run_suite.py --suite minimal --range-begin 0 --range-end 5
+          python3 run_suite.py --suite minimal --range-begin 0 --range-end 6
 
   unit-test-backend-part-2:
     if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
@@ -67,7 +67,7 @@ jobs:
         timeout-minutes: 25
         run: |
           cd test/srt
-          python3 run_suite.py --suite minimal --range-begin 5 --range-end 14
+          python3 run_suite.py --suite minimal --range-begin 6 --range-end 15
 
   unit-test-backend-part-3:
     if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
@@ -84,7 +84,7 @@ jobs:
         timeout-minutes: 25
         run: |
           cd test/srt
-          python3 run_suite.py --suite minimal --range-begin 14 --range-end 23
+          python3 run_suite.py --suite minimal --range-begin 15 --range-end 24
 
   unit-test-backend-part-4:
     if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
@@ -101,7 +101,7 @@ jobs:
         timeout-minutes: 25
         run: |
           cd test/srt
-          python3 run_suite.py --suite minimal --range-begin 23
+          python3 run_suite.py --suite minimal --range-begin 24
 
   unit-test-backend-2-gpu-part-1:
     if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'

python/sglang/srt/managers/schedule_batch.py

@@ -231,6 +231,7 @@ def __init__(
         self.tokenizer = None
         self.finished_reason = None
         self.stream = False
+        self.to_abort = False
 
         # For incremental decoding
         # ----- | --------- read_ids -------|
@@ -368,6 +369,10 @@ def check_finished(self):
         if self.finished():
             return
 
+        if self.to_abort:
+            self.finished_reason = FINISH_ABORT()
+            return
+
         if len(self.output_ids) >= self.sampling_params.max_new_tokens:
             self.finished_reason = FINISH_LENGTH(
                 length=self.sampling_params.max_new_tokens

python/sglang/srt/managers/scheduler.py

@@ -579,6 +579,8 @@ def handle_generate_request(
                     "Image request length is longer than the KV cache pool size or "
                     "the max context length aborting because you cannot truncate the image embeds"
                 )
+                req.image_inputs = None
+                req.origin_input_ids = [0]
                 req.sampling_params.max_new_tokens = 0
                 self.waiting_queue.append(req)
                 return
@@ -1350,13 +1352,15 @@ def abort_request(self, recv_req: AbortReq):
 
         if to_del is not None:
             del self.waiting_queue[to_del]
+            logger.debug(f"Abort queued request. {req.rid=}")
+            return
 
         # Delete requests in the running batch
         if self.running_batch:
             for req in self.running_batch.reqs:
                 if req.rid == recv_req.rid and not req.finished():
-                    req.finished_reason = FINISH_ABORT()
-                    self.tree_cache.cache_finished_req(req)
+                    logger.debug(f"Abort running request. {req.rid=}")
+                    req.to_abort = True
                     break
 
     def update_weights(self, recv_req: UpdateWeightReqInput):

python/sglang/test/test_utils.py

@@ -677,8 +677,14 @@ def run_and_check_memory_leak(
     enable_mixed_chunk,
     disable_overlap,
     chunked_prefill_size,
+    assert_has_abort,
 ):
-    other_args = ["--chunked-prefill-size", str(chunked_prefill_size)]
+    other_args = [
+        "--chunked-prefill-size",
+        str(chunked_prefill_size),
+        "--log-level",
+        "debug",
+    ]
     if disable_radix_cache:
         other_args += ["--disable-radix-cache"]
     if enable_mixed_chunk:
@@ -723,14 +729,19 @@ def run_and_check_memory_leak(
     # Assert success
     has_new_server = False
     has_leak = False
+    has_abort = False
     for line in output_lines:
         if "The server is fired" in line:
             has_new_server = True
         if "leak" in line:
             has_leak = True
+        if "Abort" in line:
+            has_abort = True
 
     assert has_new_server
     assert not has_leak
+    if assert_has_abort:
+        assert has_abort
 
 
 def run_mmlu_test(
@@ -761,6 +772,7 @@ def workload_func(base_url, model):
         enable_mixed_chunk,
         disable_overlap,
         chunked_prefill_size,
+        assert_has_abort=False,
     )
 
 
@@ -800,4 +812,5 @@ def run_one(_):
         enable_mixed_chunk,
         enable_overlap,
         chunked_prefill_size,
+        assert_has_abort=False,
     )

test/srt/run_suite.py

@@ -10,6 +10,7 @@
         "models/test_lora.py",
         "models/test_reward_models.py",
         "sampling/penaltylib",
+        "test_abort.py",
         "test_chunked_prefill.py",
         "test_double_sparsity.py",
         "test_embedding_openai_server.py",

test/srt/test_abort.py

@@ -0,0 +1,54 @@
+import multiprocessing
+import time
+import unittest
+from concurrent.futures import ThreadPoolExecutor
+
+import requests
+
+from sglang.test.test_utils import run_and_check_memory_leak
+
+
+class TestAbort(unittest.TestCase):
+    def workload_func(self, base_url, model):
+        def process_func():
+            def run_one(_):
+                prompt = """
+                System: You are a helpful assistant.
+                User: What is the capital of France?
+                Assistant: The capital of France is
+                """
+
+                response = requests.post(
+                    f"{base_url}/generate",
+                    json={
+                        "text": prompt,
+                        "sampling_params": {
+                            "temperature": 0,
+                            "max_new_tokens": 2048,
+                        },
+                    },
+                )
+                ret = response.json()
+
+            with ThreadPoolExecutor(16) as executor:
+                list(executor.map(run_one, list(range(16))))
+
+        p = multiprocessing.Process(target=process_func)
+        p.start()
+        time.sleep(0.5)
+        p.terminate()
+        time.sleep(10)
+
+    def test_memory_leak(self):
+        run_and_check_memory_leak(
+            self.workload_func,
+            disable_radix_cache=False,
+            enable_mixed_chunk=False,
+            disable_overlap=False,
+            chunked_prefill_size=8192,
+            assert_has_abort=True,
+        )
+
+
+if __name__ == "__main__":
+    unittest.main()