biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[CI] Fix mypy for vllm/v1/core and vllm/v1/engine (#27108)

Browse Source 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Wentao Ye
2025-10-30 07:32:17 -04:00
committed by GitHub
parent c7d2a554ba
commit c01f6e525f
12 changed files with 91 additions and 61 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
vllm/v1/core/sched/scheduler.py
View File

@@ -167,7 +167,7 @@ class Scheduler(SchedulerInterface):
self.kv_cache_manager = KVCacheManager(
kv_cache_config=kv_cache_config,
max_model_len=self.max_model_len,
enable_caching=self.cache_config.enable_prefix_caching,
enable_caching=bool(self.cache_config.enable_prefix_caching),
use_eagle=self.use_eagle,
log_stats=self.log_stats,
enable_kv_cache_events=self.enable_kv_cache_events,
@@ -407,13 +407,13 @@ class Scheduler(SchedulerInterface):
# Get externally-cached tokens if using a KVConnector.
if self.connector is not None:
num_external_computed_tokens, load_kv_async = (
ext_tokens, load_kv_async = (
self.connector.get_num_new_matched_tokens(
request, num_new_local_computed_tokens
)
)
if num_external_computed_tokens is None:
if ext_tokens is None:
# The request cannot be scheduled because
# the KVConnector couldn't determine
# the number of matched tokens.
@@ -421,6 +421,8 @@ class Scheduler(SchedulerInterface):
skipped_waiting_requests.prepend_request(request)
continue
num_external_computed_tokens = ext_tokens
# Total computed tokens (local + external).
num_computed_tokens = (
num_new_local_computed_tokens + num_external_computed_tokens
@@ -905,13 +907,13 @@ class Scheduler(SchedulerInterface):
outputs: dict[int, list[EngineCoreOutput]] = defaultdict(list)
spec_decoding_stats: SpecDecodingStats | None = None
kv_connector_stats = (
kv_connector_stats: KVConnectorStats | None = (
kv_connector_output.kv_connector_stats if kv_connector_output else None
)
if kv_connector_stats and self.connector:
stats = self.connector.get_kv_connector_stats()
if stats:
kv_connector_stats = kv_connector_stats.aggregate(stats)
kv_stats = self.connector.get_kv_connector_stats()
if kv_stats:
kv_connector_stats = kv_connector_stats.aggregate(kv_stats)
failed_kv_load_req_ids = None
if kv_connector_output and kv_connector_output.invalid_block_ids: