DeepSeek/DeepGEMM
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepGEMM synced 2025-05-31 00:48:12 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
116 Commits 3 Branches 0 Tags 3.4 MiB
5acb5a9ec5
Commit Graph

116 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Gabriel Wu
5acb5a9ec5
Merge 8aff6309d4 into d374456787 2025-04-29 20:28:42 +08:00
Zihua Wu
8aff6309d4 Merge remote-tracking branch 'upstream/main' into nvrtc 2025-04-29 05:28:07 -07:00
Chenggang Zhao
d374456787 Less stages for small shape K 2025-04-28 10:36:08 +08:00
Chenggang Zhao
86afd0c212 Add two more optimization TODOs 2025-04-27 17:51:11 +08:00
Zihua Wu
69852c465f doc: update README 
Signed-off-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
 2025-04-25 18:57:22 -07:00
Zihua Wu
d473f594be Merge remote-tracking branch 'upstream/main' into nvrtc 2025-04-25 18:56:49 -07:00
Zihua Wu
f6198492cb feat: drop support for CUDA<12.3 
Signed-off-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
 2025-04-25 18:56:40 -07:00
Chenggang Zhao
33e0c3ce40 Update plans 2025-04-24 14:37:53 +08:00
Zihua Wu
46762b6903 feat: make API more general 
Signed-off-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
 2025-04-23 02:34:23 -07:00
Zihua Wu
6c982791eb Merge remote-tracking branch 'upstream/main' into nvrtc 2025-04-23 00:17:28 -07:00
Gabriel Wu
6f0a17cb10
Merge pull request #1 from lucifer1004/nvrtc-compat 
feat: add compat for older drivers and Windows
 2025-04-23 15:06:10 +08:00
yukuai26
95e81b3dd6
Indivisible TMA (#90) 
Fix indivisible shapes for TMA multicast

---------

Co-authored-by: yukuai <yukuai@deepseek.com>
Co-authored-by: Chenggang Zhao <chenggangz@deepseek.com>
 2025-04-23 14:55:14 +08:00
Gabriel Wu
2d8c4f22d5 fix: windows compat 
Signed-off-by: Gabriel Wu <13583761+lucifer1004@users.noreply.github.com>
 2025-04-23 14:47:15 +08:00
Zihua Wu
40c09fb883 feat: fix win compat 
Signed-off-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
 2025-04-22 23:01:23 -07:00
Zihua Wu
a3210ac850 feat: save kernel name to file 
Signed-off-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
 2025-04-22 22:23:47 -07:00
Zihua Wu
767793bf95 feat: compat for old drivers 
Signed-off-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
 2025-04-22 20:57:44 -07:00
Zihua Wu
78c7fa347e fix: compiler version 
Signed-off-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
 2025-04-23 00:06:18 +00:00
Zihua Wu
c14cad0c06 refactor: compile to .cubin and add NVRTC option 
Signed-off-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
 2025-04-22 10:17:52 +00:00
Zihua Wu
27cd276e19 [wip] refactor: compile to .cubin 
Signed-off-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
 2025-04-22 08:08:40 +00:00
yukuai26
891f35adf5
Support TMA multicast on B with m_grouped_gemm_contiguous. (#88) 2025-04-21 09:43:17 +08:00
Chenggang Zhao
83aa960b9b Fix bugs 2025-04-18 11:55:51 +08:00
Chenggang Zhao
fea9309c1e Update README 2025-04-18 11:38:52 +08:00
Chenggang Zhao
340d9880f4 Overlap TMA store 2025-04-18 11:18:23 +08:00
Zhean Xu
4499c4ccbb
Refactor MMA template with CUTLASS (#87) 
* Refactor MMA with cutlass

* Update README.md

---------

Co-authored-by: Zhean Xu <xza@deepseek.com>
 2025-04-14 17:06:49 +08:00
Chenggang Zhao
37aa127451
Use swizzling instead of padding (#86) 
* Add swizzling params

* Add TMA D descriptor

* Always use STSMx2

* Swizzling draft

* Compatible with padding

* Fix bugs

* Optimize swizzle performance

* Optimize expression

* Optimize TMA issues

* Fix README

* Stricter assertions
 2025-04-14 15:20:58 +08:00
Chenggang Zhao
2e7e58011b
Merge pull request #83 from deepseek-ai/tma-1d-store 
Use 1D TMA store instead of 3D
 2025-04-11 11:25:01 +08:00
Chenggang Zhao
b0d64817a7 OOB bugs fixed 2025-04-11 11:00:47 +08:00
Chenggang Zhao
99eb6ec563 Remove useless STSM 2025-04-11 10:45:36 +08:00
Chenggang Zhao
8041ed7164 Use 1D TMA store 2025-04-11 10:42:01 +08:00
Chenggang Zhao
a77009cb14 Make partition pipelined 2025-04-10 18:07:25 +08:00
Chenggang Zhao
5bda27244b Add CMake support for CLion indexing 2025-04-10 09:57:54 +08:00
Chenggang Zhao
327ec92f69 Update roadmap 2025-04-09 11:44:30 +08:00
Chenggang Zhao
677143be64 Update roadmap 2025-04-09 11:41:36 +08:00
Chenggang Zhao
fed3e4d701
Merge pull request #81 from deepseek-ai/blocktile-256x128 
Performance: BlockTile 256x128 optimizations enable 1500+ TF FP8
 2025-04-09 11:26:40 +08:00
Chenggang Zhao
989c9e3694 Update README 2025-04-09 11:17:47 +08:00
Chenggang Zhao
a9967bc27c Update README 2025-04-09 11:14:45 +08:00
Chenggang Zhao
5a80e4bb96 Fix indent x2 2025-04-09 11:00:10 +08:00
Chenggang Zhao
bdca8b0624 Fix indent 2025-04-09 10:59:07 +08:00
Chenggang Zhao
4c0cc290c7 Refactor M repetition with loops 2025-04-09 10:50:44 +08:00
Chenggang Zhao
a6524d411a Larger block N candidates 2025-04-09 10:11:43 +08:00
Chenggang Zhao
48a5f071be Clean up config heuristics 2025-04-09 10:01:15 +08:00
Chenggang Zhao
ce65d5e33c Remove unused x256 WGMMA 2025-04-09 09:32:46 +08:00
sazc
97575bf1c6 Performance: BlockTile 256x128 optimizations enable 1500+ TFLOPS FP8 performance on the H800-SXM platform 2025-04-08 17:42:23 +08:00
Chenggang Zhao
b4ecf9c3ff Fix TMA multicast bugs 2025-04-07 14:34:42 +08:00
Chenggang Zhao
bff5724ded Code format 2025-04-07 09:32:43 +08:00
Chenggang Zhao
3ea3cb203c
Merge pull request #80 from abcdabcd987/fix-link-error 
Fix linking error from ODR violation
 2025-04-07 09:31:58 +08:00
Chenggang Zhao
b0868c9014
Merge pull request #79 from yizhang2077/lru-cache-opt 
Add lru-cache for get_best_configs to avoid repeated calculation
 2025-04-07 09:31:30 +08:00
Lequn Chen
611e3f659d Fix linking error from ODR violation 2025-04-05 17:35:23 +00:00
Yi Zhang
776bd0cccc
add lru-cache to avoid repeated calculation 2025-04-04 12:44:26 +08:00
Chenggang Zhao
c187c23ba8
Merge pull request #78 from deepseek-ai/tma-3d-padding 
Solving bank conflict via padding and TMA 3D store
 2025-04-03 16:06:10 +08:00