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

[main] rename device type (#5099)

Browse Source 
### What this PR does / why we need it?
Rename `_910B` to `A2`;
Rename `_910_93` to `A3`;
Rename `_910_95` to `A5`;

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

Signed-off-by: zzzzwwjj <1183291235@qq.com>
This commit is contained in:
zzzzwwjj
2025-12-17 14:08:19 +08:00
committed by GitHub
parent 4144376e88
commit 06b82e7503
16 changed files with 47 additions and 48 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm_ascend/ops/fused_moe/token_dispatcher.py
View File

@@ -99,7 +99,7 @@ class TokenDispatcherWithMC2(MoETokenDispatcher):
self.enable_dispatch_v2 = hasattr(torch_npu,
"npu_moe_distribute_dispatch_v2")
self.need_extra_args = (
get_ascend_device_type() == AscendDeviceType._910_93)
get_ascend_device_type() == AscendDeviceType.A3)
# NOTE: When in A2, setting the environment variables HCCL_INTRA_PCIE_ENABLE=1 and
# HCCL_INTRA_ROCE_ENABLE=0 can reduce cross-machine communication traffic and significantly