Commits · 3959c943f769d7c6f77fe781dae48a96b2949b9a · jerrrrry / infinicore

25 Sep, 2025 1 commit

zhangyue authored Sep 25, 2025

* issue/472: p800 ccl

* issue/472: 删掉无用操作

* issue/472: fix format

* issue/472: memcpy h2h case

3959c943

04 Sep, 2025 1 commit
- feat: implement GEMM with MUBLAS and MUDNN backends in moore gpu · 5ffd00ea
  zhushuang authored Sep 03, 2025
  
  5ffd00ea
27 Aug, 2025 2 commits
- issue/340: xblas · b92ecc31
  zhangyue authored Aug 27, 2025
  
  b92ecc31
- issue/294：摩尔通信库接入 · f2bb97ad
  spike-zhu authored Aug 27, 2025
  
  f2bb97ad
26 Aug, 2025 1 commit
- issue/404: kunlun_common.h 和 kunlun_handle.h 解耦合 · 2fe7052f
  zhangyue authored Aug 26, 2025
  
  2fe7052f
18 Aug, 2025 1 commit
- feat: Add BF16 support gemm in moore gpu and rename existing 'musa' to 'moore' in some files · 66a8eb93
  zhushuang authored Aug 18, 2025
  
  66a8eb93
13 Aug, 2025 1 commit
- issue/214 - Elementwise Support for Cambricon Bang · adbda4c4
  thatPepe authored Aug 13, 2025
  
  adbda4c4
07 Aug, 2025 1 commit

issue/349 P800-elementwise · 72c4dc7c

zhangyue authored Aug 07, 2025

* 适配p800软件栈

* 支持p800上编译手写算子，重构elementwise 算子组件

* fix format

* delete include xtdk_io

* 支持 elementwise float16 bfloat16 数据类型

* 修改自定义 size_t 和 ptrdiff_t 中基础数据类型

* p800 拆分 kernel

72c4dc7c

06 Aug, 2025 2 commits
- 支持p800上编译手写算子，重构elementwise 算子组件 · f7e7c7ba
  zhangyue authored Aug 06, 2025
  
  f7e7c7ba
- 适配p800软件栈 · 7d3ca92d
  zhangyue authored Jul 30, 2025
  
  7d3ca92d
30 Jul, 2025 1 commit
- issue/112: infinirt测试框架 · 3c74a2e3
  Catheriany authored Jul 30, 2025
  
  3c74a2e3
18 Jul, 2025 2 commits
- Merge pull request #332 from InfiniTensor/issue/277 · 11dd5850
  Jiacheng Huang authored Jul 18, 2025
```
Issue/277: 为 ReLU 的九齿实现接入天数设备
```
  11dd5850
- issue/333- fix compile error. · 0981ae87
  pengcheng888 authored Jul 18, 2025
  
  0981ae87
15 Jul, 2025 1 commit
- issue/277: 添加 ReLU 算子的九齿实现 · 400fad38
  Jiacheng Huang authored Jul 15, 2025
  
  400fad38
14 Jul, 2025 3 commits
- issue/158/refactor: 分离天数和英伟达的算子定义 · e09a0b7c
  YdrMaster authored Jul 14, 2025
```
Signed-off-by: YdrMaster <ydrml@hotmail.com>
```
  e09a0b7c
- issue/291/style: xmake target 中 cuda 改为 nvidia · 1d8d9623
  YdrMaster authored Jul 14, 2025
```
Signed-off-by: YdrMaster <ydrml@hotmail.com>
```
  1d8d9623
- issue/291/style: 根据实际情况将 cuda 改为 nvidia · ceb57c2a
  YdrMaster authored Jul 14, 2025
```
Signed-off-by: YdrMaster <ydrml@hotmail.com>
```
  ceb57c2a
11 Jul, 2025 3 commits
- issue/291/style: 所有 maca 改为 metax · 507be07e
  YdrMaster authored Jul 11, 2025
```
Signed-off-by: YdrMaster <ydrml@hotmail.com>
```
  507be07e
- issue/158/feat: 编译天数 ccl · 31d70fdb
  YdrMaster authored Jul 11, 2025
```
Signed-off-by: YdrMaster <ydrml@hotmail.com>
```
  31d70fdb
- issue/291/style: 一系列 maca 改为 metax 并格式化 · d15aaa3b
  PanZezhong authored Jul 11, 2025
```
Signed-off-by: PanZezhong <panzezhong@qiyuanlab.com>
```
  d15aaa3b
10 Jul, 2025 2 commits
- issue/291/build: nvidia 和 metax 共用 softmax kernel 实现 · f0c5a569
  YdrMaster authored Jul 01, 2025
```
Signed-off-by: YdrMaster <ydrml@hotmail.com>
```
  f0c5a569
- issue/291/fix: 动态加载 cuda 路径，避免依赖环境变量 · e8d427f3
  YdrMaster authored Jul 09, 2025
```
Signed-off-by: YdrMaster <ydrml@hotmail.com>
```
  e8d427f3
09 Jul, 2025 3 commits
- issue/314/fix: cuda compilation and test · f72b8653
  YdrMaster authored Jul 10, 2025
```
Signed-off-by: YdrMaster <ydrml@hotmail.com>
```
  f72b8653
- issue/291/build: 为 cudnn 增加一个编译选项，并允许 cuda 使用标记弃用的函数 · f789a52f
  YdrMaster authored Jun 30, 2025
```
Signed-off-by: YdrMaster <ydrml@hotmail.com>
```
  f789a52f
- issue/232: 接入九齿 · a19efb54
  Jiacheng Huang authored Jul 09, 2025
  
  a19efb54
04 Jul, 2025 1 commit
- issue/304 fix clip · 6e7fe25f
  PanZezhong authored Jul 04, 2025
  
  6e7fe25f
03 Jul, 2025 1 commit
- issue/300 - InfiniCCL Cambricon · f0300ff3
  wooway777 authored Jul 03, 2025
  
  f0300ff3
02 Jul, 2025 2 commits
- issue/158/fix: 修改天数上的其他编译问题 · 09ed53f7
  YdrMaster authored Jun 06, 2025
```
Signed-off-by: YdrMaster <ydrml@hotmail.com>
```
  09ed53f7
- issue/158/feat: 支持天数编译 · 29089d99
  YdrMaster authored Apr 08, 2025
```
Signed-off-by: YdrMaster <ydrml@hotmail.com>
```
  29089d99
19 Jun, 2025 1 commit
- fix: xmake modify in moore gpu · a1fedf0d
  zhushuang authored Jun 19, 2025
  
  a1fedf0d
11 Jun, 2025 1 commit
- issue/256 沐曦通信库 · f119c32e
  PanZezhong authored Jun 11, 2025
  
  f119c32e
04 Jun, 2025 1 commit
- issue/247 接入昇腾通信库 · 4ab20362
  Pan Zezhong authored Jun 04, 2025
  
  4ab20362
16 May, 2025 1 commit
- fix: 修复寒武纪平台 infinirt 编译问题 · fd574720
  bolun authored May 16, 2025
  
  fd574720
14 May, 2025 1 commit
- feat: 添加昇腾swiglu算子 · 6e1491bd
  zhangyunze authored Apr 02, 2025
  
  6e1491bd
25 Apr, 2025 1 commit
- issue/193 接入通信库，支持英伟达平台，增加通信库测试，修复运行时bug · c9bd8c5f
  PanZezhong authored Apr 25, 2025
  
  c9bd8c5f
14 Apr, 2025 2 commits
- issue/127: refactor elementwise framework, complete CUDA implementation,... · 6292da00
  Zimin Li authored Apr 07, 2025
```
issue/127: refactor elementwise framework, complete CUDA implementation, refactor swiglu using the generic elementwise framework
```
  6292da00
- issue/40: 实现沐曦rms_norm算子 · 7847868e
  qinyiqun authored Apr 08, 2025
  
  7847868e
10 Apr, 2025 1 commit
- issue/111: 修改 kunlun xmake 支持自动获取机器架构 · 98a72ab3
  zhangyue authored Apr 10, 2025
  
  98a72ab3
03 Apr, 2025 1 commit
- issue/87/musa: fix 使用公共的Pool和Check Macro，添加缺失的运行时接口 · 2d759447
  qinyiqun authored Mar 25, 2025
  
  2d759447
01 Apr, 2025 1 commit
- issue/111:添加rmsnorm以及算子编译流程 · 777fd6af
  zhangyue authored Apr 01, 2025
  
  777fd6af