kvcache-ai/sglang
1
0
Fork 0
mirror of https://github.com/kvcache-ai/sglang.git synced 2026-07-13 10:56:58 +00:00
Code Issues Packages Projects Releases Wiki Activity

[sgl-kernel Code Clean] Remove useless lightning_attention kernel (#13819)

Browse Source
This commit is contained in:
Xiaoyu Zhang
2025-11-24 18:26:25 +08:00
committed by GitHub
parent aeac622058
commit ecefc7904f
10 changed files with 0 additions and 1754 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
sgl-kernel/python/sgl_kernel/__init__.py
View File

@@ -13,7 +13,6 @@ from sgl_kernel.allreduce import *
from sgl_kernel.attention import (
cutlass_mla_decode,
cutlass_mla_get_workspace_size,
lightning_attention_decode,
merge_state,
merge_state_v2,
)

6
sgl-kernel/python/sgl_kernel/attention.py
View File

@@ -3,12 +3,6 @@ from typing import Optional, Tuple
import torch
def lightning_attention_decode(q, k, v, past_kv, slope, output, new_kv):
torch.ops.sgl_kernel.lightning_attention_decode.default(
q, k, v, past_kv, slope, output, new_kv
)
def merge_state(
v_a: torch.Tensor,
s_a: torch.Tensor,