Star
862
Fork
1359
代码
介绍
代码
Issues
1721
Pull Requests
2232
流水线
Actions
项目讨论
Wiki
项目成员
325
分析
项目设置
Star
862
Fork
1359
已合并
import triton backend for mm and triton cv fusion to master branch
#44574
shi-yufeng99
创建于
28 天前
从
shi-yufeng99/pytorch:
0814_master_triton_cv_branch1
合入到
Ascend/pytorch:
master
import triton backend for mm and triton cv fusion to master branch
#44574
已合并
shi-yufeng99
创建于
28 天前
从
shi-yufeng99/pytorch
0814_master_triton_cv_branch1
合入到
Ascend/pytorch
master
讨论
0
提交
0
检查
0
文件改动
0
克隆/下载
0
全选
2026-08-17
add "npu" to GPU_BENCHMARK_DEVICE_TYPES
4344938c
4344938c807e12e49e024b8acf1f61f61be2c83d
triton mm template support dynamic shape
ce42338b
ce42338b6cd81aa46e103cf9159380b9aff0687e
2026-08-15
remove redundent import
d96f17c3
d96f17c391bf351eb084e520b47e8b712a46fa1e
2026-08-14
fix lintunner 3
d113b6a7
d113b6a7316e50bcd6fec8e6deddecac68a84b74
fix lintrunner 2
235f6ba5
235f6ba5ebb2bf5a072f1df7c7aac0172dfe2708
fix lintrunner error
a9d8442e
a9d8442e227b61c19cca051626745455a172496a
skip fixed indexer for size=1, to avoid TA tt.store pass ERROR
9a8fb975
9a8fb97547c31df5e8cced84448a48e4fe4ae86d
[inductor] [feat] optimized bmm template
e704344f
e704344f327e415e73a8f11ea36d2160685cd9b3
enhance triton mm template and enable precompile parrellel for triton mm
e3f47668
e3f476689d9e44121742377ddea0acf0ecf5cd32
cherry-pick 8d09d258bd4ba5f8051247e7c258c9fbda255d13
3ac0d6e7
3ac0d6e7cf258192c58b723e343a33a658e75ca7
10
总计: 12
1
2