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

[CI] Upgrade vllm to 0.9.1 (#1165)

Browse Source 
1. upgrade vllm to 0.9.1. 0.9.0 is not supported for main branch now.
keep doc to 0.9.0 until we release the first 0.9.1 release.
2. disable V0 test for PR
3. move actionlint check to lint job

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
wangxiyuan
2025-06-11 16:33:11 +08:00
committed by GitHub
parent e46dc142bf
commit 4f5964420e
19 changed files with 72 additions and 320 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
vllm_ascend/core/scheduler.py
View File

@@ -29,8 +29,6 @@ from vllm.v1.outputs import ModelRunnerOutput
from vllm.v1.request import Request, RequestStatus
from vllm.v1.structured_output import StructuredOutputManager
from vllm_ascend.utils import vllm_version_is
class AscendScheduler(Scheduler):
"""This Scheduler extends vllm's original v1 scheduler
@@ -129,12 +127,7 @@ class AscendScheduler(Scheduler):
continue
assert num_new_tokens > 0
if vllm_version_is("0.9.0"):
blocks = computed_blocks.blocks
else:
blocks = computed_blocks.blocks[0]
blocks = computed_blocks.blocks[0]
watermark = getattr(self.scheduler_config, "watermark", 0.01)
if not self._check_watermark_for_prefill(request, num_new_tokens,
blocks, watermark):
@@ -330,14 +323,8 @@ class AscendScheduler(Scheduler):
len(computed_blocks) * self.block_size)
num_required_blocks = cdiv(num_new_tokens + num_computed_tokens,
self.block_size)
if vllm_version_is("0.9.0"):
req_blocks = self.kv_cache_manager.single_type_manager.req_to_blocks[
request.request_id]
else:
req_blocks = self.kv_cache_manager.coordinator.get_blocks(
request.request_id)
req_blocks = self.kv_cache_manager.coordinator.get_blocks(
request.request_id)
num_new_blocks = (num_required_blocks - len(req_blocks) -
len(computed_blocks))
num_evictable_computed_blocks = sum(1 for blk in computed_blocks