ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-14 18:17:44 +00:00
Code Issues Packages Projects Releases Wiki Activity
94 Commits 743 Branches 38 Tags
1c962a13ee8ce6dcdd9275af3c3d0701b261cc36
Commit Graph

94 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Chao Liu
1c962a13ee device_implicit_gemm_convolution_1_chwn_csrk_khwn: use tensor copy (instead of pointwise) for writing output, 3x3 increased from 78% to 84%, 5x5 from 80% to 84% 
[ROCm/composable_kernel commit: a65ef90308]
 2019-02-19 11:47:46 -06:00
Chao Liu
f6f3f52faa gridwise_implicit_gemm_convolution_1_chwn_csrk_khwn use khwn for thread C data now 
[ROCm/composable_kernel commit: 50b96745c6]
 2019-02-17 02:28:20 -06:00
Chao Liu
13388cef47 add anther verision of batch gemm 
[ROCm/composable_kernel commit: 1cb9885058]
 2019-02-17 01:50:57 -06:00
Chao Liu
8bde1393d4 2-type implicit gemm using chwn 
[ROCm/composable_kernel commit: 9f2e8f8bb4]
 2019-02-15 22:51:51 -06:00
Chao Liu
64620dfcd4 delete useless code 
[ROCm/composable_kernel commit: d7c84daf66]
 2019-02-15 22:24:18 -06:00
Chao Liu
180f7266cf updated build 
[ROCm/composable_kernel commit: fc60454e3a]
 2019-02-15 02:17:00 -06:00
Chao Liu
69cb94cf82 update cuda cmake config 
[ROCm/composable_kernel commit: 79e6abbda8]
 2019-02-15 02:14:26 -06:00
Chao Liu
c0baa18a3f change file extension to hip.hpp and hip.cpp 
[ROCm/composable_kernel commit: b2888adfbe]
 2019-02-15 02:13:21 -06:00
Chao Liu
153629655f update build 
[ROCm/composable_kernel commit: a414e3fdf8]
 2019-02-15 02:06:34 -06:00
Chao Liu
e7f6b820cd hip build 
[ROCm/composable_kernel commit: 67c6f73ffe]
 2019-02-15 00:54:30 -06:00
Chao Liu
ca9b55417e update cmake config 
[ROCm/composable_kernel commit: 121693b3d3]
 2019-02-14 15:12:29 -06:00
Chao Liu
662ba16dc6 refactor build, clean up 
[ROCm/composable_kernel commit: e80fbbdd71]
 2019-02-14 15:10:16 -06:00
Chao Liu
485d762f55 make LDS double buffer works, 1x1 conv now hits 80% of peak 
[ROCm/composable_kernel commit: 28354a0fa3]
 2019-02-12 00:57:08 -06:00
Chao Liu
f4ca4c6710 tune for 1x1 
[ROCm/composable_kernel commit: 61ac08661d]
 2019-02-11 22:36:17 -06:00
Chao Liu
4e28dc5d98 refactor 
[ROCm/composable_kernel commit: abf75ac039]
 2019-02-11 17:45:22 -06:00
Chao Liu
12ac44917f update with new copy op 
[ROCm/composable_kernel commit: 120ab94aa1]
 2019-02-07 01:31:00 -06:00
Chao Liu
590a9ff0d3 add lds double buffer for cnhw implicit gemm 
[ROCm/composable_kernel commit: 07f16673c9]
 2019-02-07 00:56:53 -06:00
Chao Liu
f1bb74f29a unroll some loop, register double buffer gemm 
[ROCm/composable_kernel commit: c866773642]
 2019-02-06 23:44:21 -06:00
Chao Liu
0d9cae1613 add another blockwise gemm 
[ROCm/composable_kernel commit: 1b323316a8]
 2019-02-06 23:10:08 -06:00
Chao Liu
7967bb4f94 fixed LDS alignment bug 
[ROCm/composable_kernel commit: 5e77650415]
 2019-02-06 01:54:13 -06:00
Chao Liu
663703f1e9 bug fixes 
[ROCm/composable_kernel commit: 079d63a788]
 2019-02-05 23:19:57 -06:00
Chao Liu
008027fef6 refactor 
[ROCm/composable_kernel commit: 42f4c7fd56]
 2019-02-05 18:04:23 -06:00
Chao Liu
ebf3d27947 add another version of blockwise 2d copy, refactor 
[ROCm/composable_kernel commit: 6614729a68]
 2019-02-05 17:06:53 -06:00
Chao Liu
4627565374 refactor 
[ROCm/composable_kernel commit: 4b616aad52]
 2019-02-05 00:51:37 -06:00
Chao Liu
38cb63f129 working on reducing index calculation... 
[ROCm/composable_kernel commit: 0741a8ab88]
 2019-02-04 17:16:28 -06:00
Chao Liu
2e0ff7d413 refactor 
[ROCm/composable_kernel commit: 9bbe9073ab]
 2019-02-04 15:40:34 -06:00
Chao Liu
6a96ec4a50 padding works (sort of), but code looks ugly. Tuned some resnet configs 
[ROCm/composable_kernel commit: 3439e4b5b7]
 2019-01-25 02:50:28 -06:00
Chao Liu
7544bd482b improve implicit gemm NCHW, SRCK, NKHW, and tuned 
[ROCm/composable_kernel commit: 8bd6ea1a97]
 2019-01-24 22:24:30 -06:00
Chao Liu
41a0b0b4bb fixed a bug, and refactored 
[ROCm/composable_kernel commit: 1de6fd0753]
 2019-01-24 21:20:29 -06:00
Chao Liu
29e74d62c3 add another implicit gemm: CHWN, CSRK, KHWN 
[ROCm/composable_kernel commit: 1410850ecc]
 2019-01-24 21:03:21 -06:00
Chao Liu
89b6be4ffd refactor 
[ROCm/composable_kernel commit: bd811e2c20]
 2019-01-24 16:15:51 -06:00
Chao Liu
1de6e5f711 refactor 
[ROCm/composable_kernel commit: c39c573eb8]
 2019-01-24 16:02:24 -06:00
Chao Liu
f155149751 implicit gemm: LDS double buffer 
[ROCm/composable_kernel commit: c9af4dece0]
 2019-01-24 14:28:46 -06:00
Chao Liu
b30d00c0d8 another version of blockwise 2d tensor copy 
[ROCm/composable_kernel commit: 1f3870ca19]
 2019-01-23 16:42:57 -06:00
Chao Liu
a8ef877996 tune 
[ROCm/composable_kernel commit: e9ac4855f8]
 2019-01-21 16:38:13 -06:00
Chao Liu
41c65ce2db refactor 
[ROCm/composable_kernel commit: b5b4fd28ed]
 2019-01-21 15:33:34 -06:00
Chao Liu
4ac54f0f31 refactor 
[ROCm/composable_kernel commit: c64f63d5ec]
 2019-01-21 11:36:31 -06:00
Chao Liu
fdd85c6161 add 2nd variation of implicit gemm 
[ROCm/composable_kernel commit: 2096847297]
 2019-01-20 21:14:35 -06:00
Chao Liu
7e4029f05e tune implicit_gemm 
[ROCm/composable_kernel commit: 3bd51021ab]
 2019-01-17 00:01:21 -06:00
Chao Liu
1257ff442d bug fix and tune implicit gemm 
[ROCm/composable_kernel commit: 216e3da609]
 2019-01-16 23:24:49 -06:00
Chao Liu
b869c589a2 refactor 
[ROCm/composable_kernel commit: caf4d7e6f5]
 2019-01-16 16:11:08 -06:00
Chao Liu
6e58457d19 refactor 
[ROCm/composable_kernel commit: 5872b710df]
 2019-01-16 15:45:02 -06:00
Chao Liu
148676b07c bug fix 
[ROCm/composable_kernel commit: 2b52fbd24a]
 2019-01-16 12:58:44 -06:00
Chao Liu
77fddd04bd refactor 
[ROCm/composable_kernel commit: ff7a62198d]
 2019-01-16 11:58:12 -06:00
Chao Liu
07817cbaae adding implicit gemm 
[ROCm/composable_kernel commit: 89ee259752]
 2019-01-16 02:44:10 -06:00
Chao Liu
ba06d1ee1a adding implicit gemm 
[ROCm/composable_kernel commit: 913afaeb5d]
 2019-01-16 02:01:56 -06:00
Chao Liu
8b3c613be1 adding implicit gemm 
[ROCm/composable_kernel commit: e7b8705b91]
 2019-01-15 18:11:41 -06:00
Chao Liu
aa885b185d adding implicit gemm 
[ROCm/composable_kernel commit: 84d9802d30]
 2019-01-15 00:11:30 -06:00
Chao Liu
50256bbcfe adding implicit gemm 
[ROCm/composable_kernel commit: aa0199a31c]
 2019-01-14 11:13:36 -06:00
Chao Liu
61e180de4a adding implicit gemm 
[ROCm/composable_kernel commit: dc60d16962]
 2019-01-09 19:12:22 -06:00