vLLM
/
vllm
mirror of https://github.com/vllm-project/vllm.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

[Benchmark] fix request loss if "ping" is returned (#19535) 

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
This commit is contained in:
Wang, Yi 2025-06-22 15:21:04 +08:00 committed by GitHub
parent 2bb246b8f7
commit 202c5df935
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 23 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
benchmarks/backend_request_func.py
View File

@ -404,8 +404,14 @@ async def async_request_openai_chat_completions(
chunk_bytes = chunk_bytes.strip() chunk_bytes = chunk_bytes.strip()
if not chunk_bytes: if not chunk_bytes:
continue continue
chunk_bytes = chunk_bytes.decode("utf-8")
# NOTE: SSE comments (often used as pings) start with a colon.
# These are not JSON data payload and should be skipped.
if chunk_bytes.startswith(":"):
continue
chunk = chunk_bytes.removeprefix("data: ")
chunk = chunk_bytes.decode("utf-8").removeprefix("data: ")
if chunk != "[DONE]": if chunk != "[DONE]":
timestamp = time.perf_counter() timestamp = time.perf_counter()
data = json.loads(chunk) data = json.loads(chunk)

20
vllm/benchmarks/endpoint_request_func.py
View File

@ -104,9 +104,15 @@ async def async_request_openai_completions(
chunk_bytes = chunk_bytes.strip() chunk_bytes = chunk_bytes.strip()
if not chunk_bytes: if not chunk_bytes:
continue continue
chunk_bytes = chunk_bytes.decode("utf-8")
# NOTE: SSE comments (often used as pings) start with
# a colon. These are not JSON data payload and should
# be skipped.
if chunk_bytes.startswith(":"):
continue
chunk = chunk_bytes.removeprefix("data: ")
chunk = chunk_bytes.decode("utf-8").removeprefix(
"data: ")
if chunk != "[DONE]": if chunk != "[DONE]":
data = json.loads(chunk) data = json.loads(chunk)
@ -213,9 +219,15 @@ async def async_request_openai_chat_completions(
chunk_bytes = chunk_bytes.strip() chunk_bytes = chunk_bytes.strip()
if not chunk_bytes: if not chunk_bytes:
continue continue
chunk_bytes = chunk_bytes.decode("utf-8")
# NOTE: SSE comments (often used as pings) start with
# a colon. These are not JSON data payload and should
# be skipped.
if chunk_bytes.startswith(":"):
continue
chunk = chunk_bytes.removeprefix("data: ")
chunk = chunk_bytes.decode("utf-8").removeprefix(
"data: ")
if chunk != "[DONE]": if chunk != "[DONE]":
timestamp = time.perf_counter() timestamp = time.perf_counter()
data = json.loads(chunk) data = json.loads(chunk)