Levi f0d41199a6 [Performance] Remove index opetation when VLLM_ASCEND_FLASHCOMM2_PARALLEL_SIZE=1 (#5936) ...

### What this PR does / why we need it?
When enable VLLM_ASCEND_FLASHCOMM2_PARALLEL_SIZE>1, we need index
operation to reorganize the batch, because that we need ensure the
correct batch-id for each rank after the reduce-scatter op in
VLLM_ASCEND_FLASHCOMM2_PARALLEL_SIZE>1. But we do not need it when
VLLM_ASCEND_FLASHCOMM2_PARALLEL_SIZE=1, which dose not need
reduce-scatter.

Signed-off-by: Levi-JQ <yujinqi2@huawei.com>
Co-authored-by: Levi-JQ <yujinqi2@huawei.com>

2026-01-19 17:12:13 +08:00

..

_310p

[Feature]: Support 310P device run qwen2.5/3 dense and qwen2.5vl models (#5776)

2026-01-17 11:49:18 +08:00

_cann_ops_custom

[Kernel] add custom op GmmSwigluQuantWeightNzTensorList (#3804)

2025-11-28 18:06:39 +08:00

attention

[Lint]Style: Convert vllm-ascend/ to ruff format(Batch #2) (#5977)

2026-01-19 08:59:46 +08:00

compilation

[Fusion] [Graph]Add Matmul Allreduce Rmsnorm fusion Pass (#5034)

2026-01-19 09:28:07 +08:00

core

[CI] fix lint (#5216)

2025-12-20 17:03:25 +08:00

device

[Lint]Style: Convert vllm-ascend/ to ruff format(Batch #2) (#5977)

2026-01-19 08:59:46 +08:00

device_allocator

[Lint]Style: Convert vllm-ascend/ to ruff format(Batch #2) (#5977)

2026-01-19 08:59:46 +08:00

distributed

[main][bugfix] fix mooncake kv cache transfer when one P has multi nodes (#5960)

2026-01-19 16:35:13 +08:00

eplb

[Feature] Adapt DispathGmmCombineDecode opertor to align with weight scale dtype of small operators. [RFC: issue 5476] (#5755)

2026-01-19 16:10:43 +08:00

kv_offload

[Main2Main] Upgrade vllm commit to 0113 (#5839)

2026-01-15 09:48:53 +08:00

lora

[BufFix]Fix the error when using Ascend custom operators with rank=128 (#5394)

2026-01-09 15:57:43 +08:00

model_loader

[BugFix] NetLoader: No backend type associated with device type npu (#5700)

2026-01-09 15:54:54 +08:00

ops

[Performance] Remove index opetation when VLLM_ASCEND_FLASHCOMM2_PARALLEL_SIZE=1 (#5936)

2026-01-19 17:12:13 +08:00

patch

[Fusion] [Graph]Add Matmul Allreduce Rmsnorm fusion Pass (#5034)

2026-01-19 09:28:07 +08:00

quantization

[Feature] Adapt DispathGmmCombineDecode opertor to align with weight scale dtype of small operators. [RFC: issue 5476] (#5755)

2026-01-19 16:10:43 +08:00

sample

[Feature] add the magicmtp speculative decoding acceleration algorithm (#5542)

2026-01-08 09:15:55 +08:00

spec_decode

Eagle3 mm support, enablement on qwen3vl (#4848)

2026-01-19 08:58:07 +08:00

worker

[Refactor] Move AttentionSpec initialization to Attention module (#5834)

2026-01-19 14:22:18 +08:00

xlite

[BugFix] Xlite: Bypass the padding of the graph mode in non-MTP cases to obtain the correct decode num. (#5711)

2026-01-09 15:55:30 +08:00

__init__.py

[Lint]Style: Convert vllm-ascend/compilation to ruff format (#5912)

2026-01-16 20:57:46 +08:00

ascend_config.py

[Fusion] [Graph]Add Matmul Allreduce Rmsnorm fusion Pass (#5034)

2026-01-19 09:28:07 +08:00

ascend_forward_context.py

[Lint]Style: Convert vllm-ascend/compilation to ruff format (#5912)

2026-01-16 20:57:46 +08:00

batch_invariant.py

[Lint]Style: Convert vllm-ascend/ to ruff format(Batch #2) (#5977)

2026-01-19 08:59:46 +08:00

cpu_binding.py

[Lint]Style: Convert vllm-ascend/compilation to ruff format (#5912)

2026-01-16 20:57:46 +08:00

envs.py

[Lint]Style: Convert vllm-ascend/ to ruff format(Batch #2) (#5977)

2026-01-19 08:59:46 +08:00

flash_common3_context.py

[Lint]Style: Convert vllm-ascend/compilation to ruff format (#5912)

2026-01-16 20:57:46 +08:00

meta_registration.py

[Lint]Style: Convert vllm-ascend/compilation to ruff format (#5912)

2026-01-16 20:57:46 +08:00

platform.py

[CI]fix for lint CI (#5982)

2026-01-19 09:49:28 +08:00

profiling_config.py

[Lint]Style: Convert vllm-ascend/compilation to ruff format (#5912)

2026-01-16 20:57:46 +08:00

utils.py

[Feature]: Support 310P device run qwen2.5/3 dense and qwen2.5vl models (#5776)

2026-01-17 11:49:18 +08:00