EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[Quickfix] update CachedRequestState as NewRequestData changed (#2367)

Browse Source 
### What this PR does / why we need it?
1. update `CachedRequestState` as `NewRequestData` changed in
https://github.com/vllm-project/vllm/pull/22570
2. drop maintenance of vllm v0.10.0 in the branch main

### Does this PR introduce _any_ user-facing change?
N/A

### How was this patch tested?
CI passed with existing test.


- vLLM version: v0.10.0
- vLLM main:
92ff41abea

---------

Signed-off-by: MengqingCao <cmq0113@163.com>
This commit is contained in:
Mengqing Cao
2025-08-15 07:35:27 +08:00
committed by GitHub
parent 2ad7e1251e
commit 61866b8ac6
18 changed files with 77 additions and 285 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
tests/ut/kv_connector/test_remote_decode_lifecycle.py
View File

@@ -25,7 +25,6 @@ from tests.ut.kv_connector.utils import (assert_scheduler_empty,
create_model_runner_output,
create_request, create_scheduler,
create_vllm_config)
from vllm_ascend.utils import vllm_version_is
def test_basic_lifecycle():
@@ -103,13 +102,10 @@ def test_basic_lifecycle():
# (3b): execute_model()
model_runner_output = copy.deepcopy(EMPTY_MODEL_RUNNER_OUTPUT)
if vllm_version_is("0.10.0"):
model_runner_output.finished_sending = [request_id]
else:
from vllm.v1.worker.kv_connector_model_runner_mixin import \
KVConnectorOutput # type: ignore # noqa
model_runner_output.kv_connector_output = KVConnectorOutput(
finished_sending=[request_id])
from vllm.v1.worker.kv_connector_model_runner_mixin import \
KVConnectorOutput # type: ignore # noqa
model_runner_output.kv_connector_output = KVConnectorOutput(
finished_sending=[request_id])
# (3c): update_from_output()
scheduler.update_from_output(scheduler_output, model_runner_output)
@@ -164,13 +160,10 @@ def test_prefix_cache_lifecycle():
scheduler_output = scheduler.schedule()
scheduler.schedule()
model_runner_output = copy.deepcopy(EMPTY_MODEL_RUNNER_OUTPUT)
if vllm_version_is("0.10.0"):
model_runner_output.finished_sending = [request_remote.request_id]
else:
from vllm.v1.worker.kv_connector_model_runner_mixin import \
KVConnectorOutput # noqa
model_runner_output.kv_connector_output = KVConnectorOutput(
finished_sending=[request_remote.request_id])
from vllm.v1.worker.kv_connector_model_runner_mixin import \
KVConnectorOutput # noqa
model_runner_output.kv_connector_output = KVConnectorOutput(
finished_sending=[request_remote.request_id])
scheduler.update_from_output(scheduler_output, model_runner_output)
_ = scheduler.schedule()
assert_scheduler_empty(scheduler)