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

[OPS] support triton causal_conv1d_fn ops (#4119)

Browse Source 
### What this PR does / why we need it?
Support triton causal_conv1d_fn ops.

### Does this PR introduce _any_ user-facing change?
No.

### How was this patch tested?
CI passed with new added/existing test.

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

---------

Signed-off-by: QilaiZhang <245706640@qq.com>
This commit is contained in:
QilaiZhang
2025-12-11 15:52:39 +08:00
committed by GitHub
parent eac72f5f23
commit 78bf211539
3 changed files with 632 additions and 93 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm_ascend/patch/worker/patch_triton.py
View File

@@ -4,7 +4,7 @@ from vllm_ascend.ops.triton.fla.chunk import chunk_gated_delta_rule
from vllm_ascend.ops.triton.fla.layernorm_guard import LayerNormFn
from vllm_ascend.ops.triton.fla.sigmoid_gating import \
fused_recurrent_gated_delta_rule_fwd_kernel
from vllm_ascend.ops.triton.mamba.casual_conv1d import (
from vllm_ascend.ops.triton.mamba.causal_conv1d import (
causal_conv1d_fn, causal_conv1d_update_npu)
vllm.model_executor.layers.mamba.ops.causal_conv1d.causal_conv1d_update = causal_conv1d_update_npu