chenhongmin.will
|
c7143a7bda
|
Merge branch 'main' into will_fp8_mr
|
2025-02-28 22:15:46 +08:00 |
|
chenhongmin.will
|
8b939854d8
|
enable scale
|
2025-02-28 20:07:32 +08:00 |
|
chenhongmin.will
|
bfe38ab106
|
fix combine
|
2025-02-28 18:45:09 +08:00 |
|
chenhongmin.will
|
fd1e662deb
|
fix mma0
|
2025-02-28 16:52:30 +08:00 |
|
chenhongmin.will
|
061af5fc56
|
use fa'3 transv
|
2025-02-28 14:54:44 +08:00 |
|
chenhongmin.will
|
0337732dc1
|
reorg
|
2025-02-28 08:09:02 +08:00 |
|
chenhongmin.will
|
1df91aff33
|
fix compile
|
2025-02-27 23:53:23 +08:00 |
|
chenhongmin.will
|
855c985b00
|
use 64x64 transpose_v
|
2025-02-27 22:45:00 +08:00 |
|
chenhongmin.will
|
d1689ab64f
|
use mm1's Aregs instead of mma0's Cregs
|
2025-02-27 11:59:17 +08:00 |
|
chenhongmin.will
|
1757a6db07
|
try fix
|
2025-02-27 09:11:17 +08:00 |
|
chenhongmin.will
|
dbd8c307eb
|
fix sV
|
2025-02-27 01:42:58 +08:00 |
|
chenhongmin.will
|
6dcea4952c
|
add TransV
|
2025-02-26 18:48:24 +08:00 |
|
chenhongmin.will
|
6a4eb631e2
|
add transv barrier
|
2025-02-26 17:57:00 +08:00 |
|
chenhongmin.will
|
59f691763e
|
fix Vt illegal
|
2025-02-26 17:39:29 +08:00 |
|
chenhongmin.will
|
f6fab1b915
|
change to use per_tensor
|
2025-02-26 10:21:09 +08:00 |
|
chenhongmin.will
|
4b314cd655
|
update fp8 api
|
2025-02-26 08:33:25 +08:00 |
|
chenhongmin.will
|
870418802a
|
add fp8 ut
|
2025-02-26 07:57:51 +08:00 |
|
chenhongmin.will
|
dfe8ffc75a
|
enable fp8 api
|
2025-02-25 23:02:57 +08:00 |
|
chenhongmin.will
|
c50d29d170
|
fix compile
|
2025-02-25 21:52:11 +08:00 |
|
chenhongmin.will
|
7409203f44
|
enable fp8 compile
|
2025-02-25 21:12:40 +08:00 |
|
chenhongmin.will
|
fed0499301
|
fp8 shared mem
|
2025-02-25 11:26:50 +08:00 |
|
chenhongmin.will
|
b67a18f850
|
update gmem
|
2025-02-25 09:45:19 +08:00 |
|
chenhongmin.will
|
d833dbd711
|
enable fp8
|
2025-02-25 09:03:02 +08:00 |
|
Sijia Chen
|
a3b74b8574
|
add flag to disable FP16 compile
|
2025-02-24 10:01:59 -08:00 |
|
chenhongmin.will
|
dae0690055
|
init fp8
|
2025-02-24 21:12:36 +08:00 |
|
Sijia Chen
|
65fb7732fc
|
support fp16
|
2025-02-24 01:58:53 -08:00 |
|
Sijia Chen
|
15a82b81b8
|
replace c10 optional with std optional
|
2025-02-24 00:25:40 -08:00 |
|
Jiashi Li
|
414a2f3eed
|
Initial commit
i
|
2025-02-24 09:20:23 +08:00 |
|