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

[Feat] Unquantized Linear to nz and control all nz-cast (#3356)

Browse Source 
### What this PR does / why we need it?
Currently, when executing to the Linear layer of models in vLLM-Ascend,
the weights format is ND in unquantized case and skipped ascend case.
This PR supplements the execution logic for Linear layer. We use a new
global variable: VLLM_ASCEND_ENABLE_NZ. When VLLM_ASCEND_ENABLE_NZ=1 and
CANN version is 8.3, the weights of the Linear layer will be converted
to FRACTAL_NZ, in both unquantized case and skipped ascend case. We also
use VLLM_ASCEND_ENABLE_NZ to control the existing NZ conversion, such as
w8a8-quantized case.

### Does this PR introduce _any_ user-facing change?
Add a new global variable VLLM_ASCEND_ENABLE_NZ. If you want to use NZ
format, you should set VLLM_ASCEND_ENABLE_NZ=1.

### How was this patch tested?

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

Signed-off-by: anon189Ty <Stari_Falcon@outlook.com>
This commit is contained in:
anon189Ty
2025-10-14 17:39:26 +08:00
committed by GitHub
parent 5c45c227dc
commit 07e39620ea
22 changed files with 413 additions and 49 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
vllm_ascend/quantization/w4a8_dynamic.py
View File

@@ -27,7 +27,7 @@ from vllm.forward_context import get_forward_context
from vllm_ascend.ascend_config import get_ascend_config
from vllm_ascend.distributed.parallel_state import get_mc2_group
from vllm_ascend.ops.moe.experts_selector import select_experts
from vllm_ascend.utils import ACL_FORMAT_FRACTAL_NZ
from vllm_ascend.utils import ACL_FORMAT_FRACTAL_NZ, is_enable_nz
class AscendW4A8DynamicLinearMethod:
@@ -393,9 +393,10 @@ class AscendW4A8DynamicFusedMoEMethod:
self.update_bias(layer, w13_bias, w2_bias)
layer.w13_weight.data = torch_npu.npu_format_cast(
layer.w13_weight.data, ACL_FORMAT_FRACTAL_NZ)
layer.w2_weight.data = torch_npu.npu_format_cast(
layer.w2_weight.data, ACL_FORMAT_FRACTAL_NZ)
if is_enable_nz():
layer.w13_weight.data = torch_npu.npu_format_cast(
layer.w13_weight.data, ACL_FORMAT_FRACTAL_NZ)
layer.w2_weight.data = torch_npu.npu_format_cast(
layer.w2_weight.data, ACL_FORMAT_FRACTAL_NZ)
layer.w13_weight.data = self.pack_to_int32(layer.w13_weight.data)
layer.w2_weight.data = self.pack_to_int32(layer.w2_weight.data)