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

[feat][torchair] support super kernel feat for quantized dsr1 (#3485)

Browse Source 
### What this PR does / why we need it?
Port #1916 and #2157 to master branch to fuse operators in deepseek moe
layers, which can reduce scheduling overhead on devices. Note that this
feature is valid only when `tp_size = 1` and
`multistream_overlap_shared_expert` is enabled with torchair graph mode.

### Does this PR introduce _any_ user-facing change?
Users can enable this feature with `--additional-config
'{"torchair_graph_config":{"enabled":true, "enable_super_kernel":true},
"multistream_overlap_shared_expert":true}'`.

### How was this patch tested?
E2E deepseek serving with 2P1D disaggregated prefill scenarios.


- vLLM version: v0.11.0rc3
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0

---------

Signed-off-by: linfeng-yuan <1102311262@qq.com>
This commit is contained in:
linfeng-yuan
2025-10-20 20:04:37 +08:00
committed by GitHub
parent 70bef33f13
commit 068ed706c8
8 changed files with 138 additions and 86 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
docs/source/user_guide/configuration/additional_config.md
View File

@@ -58,6 +58,7 @@ The details of each config option are as follows:
| `graph_batch_sizes` | list[int] | `[]` | The batch size for torchair graph cache |
| `graph_batch_sizes_init` | bool | `False` | Init graph batch size dynamically if `graph_batch_sizes` is empty |
| `enable_kv_nz`| bool | `False` | Whether to enable kvcache NZ layout. This option only takes effects on models using MLA (e.g., DeepSeek). |
| `enable_super_kernel` | bool | `False` | Whether to enable super kernel to fuse operators in deepseek moe layers. This option only takes effects on moe models using dynamic w8a8 quantization.|
**ascend_scheduler_config**