c1995d119a
Daily update on 20251124-010227 by Nitload Youmu. Additional Message : error in hlsf
Nitload2025-11-24 01:02:27 +08:00
61fbce14a2
add cancellable time span await
Youmu Liu2025-11-21 18:11:52 +08:00
1f05a6a962
Daily update on 20251120-211401 by Nitload Youmu. Additional Message : buffer operation is ok, thread parallel model is ok, building operators
Nitload2025-11-20 21:14:01 +08:00
98f47aaa53
add sub and scalar_mul operation
Youmu Liu2025-11-20 19:53:54 +08:00
28bdc42982
buffer op and parallel framework is now on full complete
Youmu Liu2025-11-20 17:50:06 +08:00
c595076005
buffer op brief ok, but blocks on unkown reason
Youmu Liu2025-11-20 16:17:24 +08:00
44bf78b937
use modular initing system rather than calling Init() method.
Youmu Liu2025-11-11 18:22:28 +08:00
678a846e54
rewrite of f32 fma, and other optmization of simd kernel ops
Youmu Liu2025-11-10 20:40:19 +08:00
1dee467220
Daily update on 20251110-002339 by Nitload Youmu. Additional Message : Dynamic wrapping is ok, going on test.
Nitload2025-11-10 00:23:39 +08:00
1479cd7c9f
Daily update on 20251109-030228 by Nitload Youmu. Additional Message : Dynamic wrapping is briefly ok, a special sorting is required.
Nitload2025-11-09 03:02:28 +08:00
54e0690ebf
Merge branch 'giga' of ssh://ssh.github.com:443/Nitload-NSI/CCD_Github into local merge newest code to main branch
Nitload2025-11-08 21:46:17 +08:00
f959339e18
working on dynamic wrapping, a good awaiting is on way
Youmu Liu2025-11-07 18:22:50 +08:00
ea0e6f3ff6
optimize para performance of buffered stack
Youmu Liu2025-11-05 20:33:58 +08:00
133c2a063b
Daily update on 20251104-232717 by Nitload Youmu. Additional Message : New design, dynamic wrapping instead of former static wrapping.
Nitload2025-11-04 23:27:17 +08:00
3b23272f38
Daily update on 20251023-010006 by Nitload Youmu. Additional Message : Put matrix parallel pool and other supporting methods into individual classes.
Nitload2025-10-23 01:00:06 +08:00
affaf03845
Put thread pool to individual type define as thread resource manager.
Youmu Liu2025-10-22 20:36:15 +08:00
3707e5bd74
Put thread pool to individual type define as thread resource manager.
Youmu Liu2025-10-22 20:35:05 +08:00
3788db4cd0
Arrange code structure. Introduce dynamic scheduler and multi leader.
Youmu Liu2025-10-22 14:00:13 +08:00
91a0add1d1
Arrange code structure. Introduce dynamic scheduler and multi leader.
Youmu Liu2025-10-21 18:06:20 +08:00
b26ec45433
Arrange code structure. Introduce dynamic scheduler and multi leader.
Youmu Liu2025-10-21 16:39:34 +08:00
838ce73133
Brief solution for publishing, debuging
Youmu Liu2025-10-20 20:36:20 +08:00
547ba80a01
Daily update on 20251020-041041 by Nitload Youmu. Additional Message : Code cleanup and new design of parallel (maybe THE THENTH VERSION? Tough work!).
Nitload2025-10-20 04:10:41 +08:00
b8a51c13dd
Daily update on 20250921-173333 by Nitload Youmu. Additional Message : New design of matrix enumeration, capable for parallel task publication.
Nitload2025-09-21 17:33:33 +08:00
0404ecf044
Merge branch 'giga' of ssh://ssh.github.com:443/Nitload-NSI/CCD_Github into local merge back to main device
Nitload2025-09-20 20:53:30 +08:00
62cf2430f4
mat kernel turns to be generic
Youmu Liu2025-09-19 20:38:44 +08:00
36f8fb1107
Daily update on 20250915-010637 by Nitload Youmu. Additional Message : Matrix parallel pool and wrap is almost done, going on test and merge.
Nitload2025-09-15 01:06:37 +08:00
4769986c37
parallel task executor framework ok, working on scheduler
Youmu Liu2025-09-08 21:37:41 +08:00
a7f82c04ea
Daily update on 20250907-184649 by Nitload Youmu. Additional Message : Matrix parallel task packet is done, parallel task executer is done. Now is working on building parallel wrap and pool.
Nitload2025-09-07 18:46:49 +08:00
c9442d2b4f
Problem on fma data sync or thread load unbalance
Youmu Liu2025-09-03 21:29:45 +08:00
125040620b
Huge progress on Matrix Parallel, a parallel pool has its basic structure.
Youmu Liu2025-09-02 21:56:10 +08:00
2475a3b0ed
no progress in matrix parallel
Youmu Liu2025-09-01 22:17:52 +08:00
5e68a1e786
Daily update on 20250831-201511 by Nitload Youmu. Additional Message : HLSF allocator full functional available, working on matrix operations with native allocation and parallel computing.
Nitload2025-08-31 20:15:11 +08:00