youmuliu/stgsharp
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
549 Commits 2 Branches 0 Tags
6b449a341eab998dc8d2b0e153a2d56c8f123c4e
Commit Graph
39 Commits
Author SHA1 Message Date
Youmu Liu ae664cb541 matrix parallel start to use two different policy 2026-04-27 20:08:10 +08:00
Youmu Liu ecb1f04770 matrix thread pool use ParaWaveFront and heavy sync 2026-04-26 17:38:10 +08:00
Youmu Liu 2e2d64a2ce HLSf use nuint rather than long, L4 now uses hlsf rather than array to alloc cache line. 2026-04-20 18:10:14 +08:00
Nitload 5fc8e62f28 solution to length unfixed cacheline in l4 2026-04-19 02:30:35 +08:00
Youmu Liu 8a66c1697a L4 completely add support for multi predictor, working on ver 5 mat para 2026-04-10 18:05:40 +08:00
Youmu Liu f7187ca7fd add multi predictor support to l4, working for ver 5 of mat parallel 2026-04-09 18:07:49 +08:00
Nitload 0c1887d8f8 Daily update on 20260312-004950 by Nitload Youmu. Additional Message : use one single index mux as replacement of two slab allocators for L4. 
Signed-off-by: Nitload <YoumuLiu_NSI@whv0t.onmicrosoft.com>
 2026-03-12 00:49:50 +08:00
Youmu Liu 9b5dfec9d5 some chore 2026-03-11 18:55:45 +08:00
Youmu Liu d97fb1dd90 l4 handle 2026-03-09 18:14:49 +08:00
Nitload 64360b63a1 Daily update on 20260307-020120 by Nitload Youmu. Additional Message : None 
Signed-off-by: Nitload <YoumuLiu_NSI@whv0t.onmicrosoft.com>
 2026-03-07 02:01:20 +08:00
Youmu Liu 2149bbeede Normalize all existing files to LF 2026-03-06 18:31:11 +08:00
Youmu Liu f6e897ed06 swiss ok, needs more benchmark 2026-03-06 18:22:30 +08:00
Nitload 7e970352ab Better module loading. 2026-03-04 00:51:29 +08:00
Youmu Liu 91e136f81f Working on L4, now on swiss. FMA_TILE is re arranged. 2026-03-03 18:02:31 +08:00
Youmu Liu 7e3ed8b0fd decide to use L4 to solve numa and pack problem for large compute. 2026-03-02 17:33:03 +08:00
Nitload d99c1e3aa6 add unicode figure 2026-02-23 19:54:21 +08:00
Nitload c155cab510 HLSF use long as aloloc size 2026-02-23 12:32:15 +08:00
Youmu Liu 807658111c fix errors in interpreter. add a enumerator for tree. working on tree optimizer 2026-02-09 18:08:49 +08:00
Youmu Liu e5539bcc6d keep working on PLN facotrize 2025-12-25 11:59:40 +08:00
Youmu Liu 28bdc42982 buffer op and parallel framework is now on full complete 2025-11-20 17:50:06 +08:00
Youmu Liu 04fc55d31e add scheduler method; fixing buffer parallel errors 2025-11-19 18:09:25 +08:00
Youmu Liu ece12f2ea7 new simplified timer. and little other ops 2025-11-18 20:39:03 +08:00
Youmu Liu a8c522ab23 small size buffer op ok 2025-11-14 18:28:15 +08:00
Youmu Liu 44bf78b937 use modular initing system rather than calling Init() method. 2025-11-11 18:22:28 +08:00
Youmu Liu 678a846e54 rewrite of f32 fma, and other optmization of simd kernel ops 2025-11-10 20:40:19 +08:00
Nitload 1dee467220 Daily update on 20251110-002339 by Nitload Youmu. Additional Message : Dynamic wrapping is ok, going on test. 
Signed-off-by: Nitload <YoumuLiu_NSI@whv0t.onmicrosoft.com>
 2025-11-10 00:23:39 +08:00
Youmu Liu f959339e18 working on dynamic wrapping, a good awaiting is on way 2025-11-07 18:22:50 +08:00
Youmu Liu ec1b8dbd51 working on dynamic wrapping 2025-11-06 18:12:04 +08:00
Youmu Liu ea0e6f3ff6 optimize para performance of buffered stack 2025-11-05 20:33:58 +08:00
Youmu Liu 9c4c71dc5b [rebuid]use dynamic wrapping instead of static wrapping 2025-11-03 20:03:50 +08:00
Youmu Liu e1ec761e83 [Internal] Seperate InternalIO into different static classes. [Matrix] Fix logic error of compute model. 2025-10-27 19:59:12 +08:00
Youmu Liu 3707e5bd74 Put thread pool to individual type define as thread resource manager. 2025-10-22 20:35:05 +08:00
Youmu Liu b26ec45433 Arrange code structure. Introduce dynamic scheduler and multi leader. 2025-10-21 16:39:34 +08:00
Youmu Liu 1927f4fdf3 code clean up 2025-10-17 10:49:00 +08:00
Youmu Liu d3e0b5d005 new manage of wraps 2025-10-15 20:29:02 +08:00
Youmu Liu 0cafe9fd4d mpl going on good process 2025-09-24 20:38:15 +08:00
Nitload b8a51c13dd Daily update on 20250921-173333 by Nitload Youmu. Additional Message : New design of matrix enumeration, capable for parallel task publication. 
Signed-off-by: Nitload <YoumuLiu_NSI@whv0t.onmicrosoft.com>
 2025-09-21 17:33:33 +08:00
Nitload 1359ff4ecc old data 2025-08-25 09:50:30 +08:00
Nitload Youmu 43bbc06bbb add from giga-device, new branch added 2025-08-21 16:51:25 +08:00