hucong 7ea17fbee3 [0.11.0][BugFix] Improve the performance of prefixcache features (#4021) ...

### What this PR does / why we need it?
cherry-pick from https://github.com/vllm-project/vllm-ascend/pull/4022

The code bug caused an empty bubble. When the npu_paged_cache_load
operator was called, it forcibly transferred seq_len2 to the device,
which triggered synchronization and interrupted the CPU operator's
launch stream.


---------

Signed-off-by: underfituu <hzhucong@163.com>

2025-11-10 11:51:34 +08:00

..

attention

[0.11.0][BugFix] Improve the performance of prefixcache features (#4021)

2025-11-10 11:51:34 +08:00

compilation

[cherry-pick][v0.11.0-dev][bugfix] Fix a rare bug triggered by _npu_paged_attention in FULL_DECODE_ONLY mode (#3987)

2025-11-06 23:08:57 +08:00

core

[BugFix][Cherry-pick] Cherry-pick PR 3675 to v0.11.0-dev (#3732)

2025-10-25 09:41:51 +08:00

device_allocator

[Misc]Clean up useless import from vllm (#2049)

2025-07-28 16:01:59 +08:00

distributed

[P/D][BugFix][v0.11.0-dev]Fix proxy format processing errors & Layerwise connector performance optimization (#4069)

2025-11-09 09:55:10 +08:00

eplb

[CI]Add EPLB CI. (#3568)

2025-10-21 22:58:02 +08:00

lora

[Bugfix][LoRA] Fix forward error and shape mismatch when using LoRA (#3153)

2025-09-28 17:30:50 +08:00

models

[0.11.0][Perf] Add padding vision tower for Qwen2_5_Omni (#4041)

2025-11-08 13:56:05 +08:00

multistream

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

2025-08-15 07:35:27 +08:00

ops

[BugFix]Fix grouplist type of mc2. (#4049)

2025-11-07 17:43:23 +08:00

patch

[Bugfix][SHM] Fix weak memory ordering problem in share memory (#3988)

2025-11-04 23:07:23 +08:00

quantization

[v0.11.0][bugfix] Add 'layer_type' param to get_pergroup_param() for compatibility (#3684)

2025-10-23 21:26:50 +08:00

sample

Drop 0.10.2 (#3284)

2025-10-09 10:28:38 +08:00

spec_decode

[0.11.0][MTP][Aclgraph] Fix the support aclgraph with MTP (#3912)

2025-11-03 14:25:37 +08:00

torchair

[0.11.0][BugFix] Improve the performance of prefixcache features (#4021)

2025-11-10 11:51:34 +08:00

worker

[v0.11.0][Bugfix] fix sleepmode level2 e2e test (#4023)

2025-11-08 14:11:15 +08:00

__init__.py

[Refactor] Adapt deepseek-v3.2 to vllm 0.11.0 (#3432)

2025-10-15 17:48:58 +08:00

ascend_config.py

[main] support cpu binding (#3546)

2025-10-21 09:17:03 +08:00

ascend_forward_context.py

[0.11.0]Chery pick pta upgrade change (#3940)

2025-10-31 22:14:26 +08:00

cpu_binding.py

[main] support cpu binding (#3546)

2025-10-21 09:17:03 +08:00

envs.py

[Feat] Flash comm allgher ep (#3334)

2025-10-15 19:36:32 +08:00

meta_registration.py

Fix the bugs about operator registration by PyTorch Dispatcher (#2786)

2025-09-13 11:58:52 +08:00

platform.py

fix bug when max_seqs=14 in mtp=2 scenario and raise error when cudagraph_capture_sizes can't be an integer multiple of uniform_decode_query_len (#3909)

2025-10-31 09:25:06 +08:00

utils.py

[MM][Bugfix] Add MoE verification for multi-modal models (#3897) (#4027)

2025-11-06 20:30:40 +08:00