88 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
Chao Liu	79e6abbda8	update cuda cmake config	2019-02-15 02:14:26 -06:00
Chao Liu	b2888adfbe	change file extension to hip.hpp and hip.cpp	2019-02-15 02:13:21 -06:00
Chao Liu	a414e3fdf8	update build	2019-02-15 02:06:34 -06:00
Chao Liu	67c6f73ffe	hip build	2019-02-15 00:54:30 -06:00
Chao Liu	121693b3d3	update cmake config	2019-02-14 15:12:29 -06:00
Chao Liu	e80fbbdd71	refactor build, clean up	2019-02-14 15:10:16 -06:00
Chao Liu	28354a0fa3	make LDS double buffer works, 1x1 conv now hits 80% of peak	2019-02-12 00:57:08 -06:00
Chao Liu	61ac08661d	tune for 1x1	2019-02-11 22:36:17 -06:00
Chao Liu	abf75ac039	refactor	2019-02-11 17:45:22 -06:00
Chao Liu	120ab94aa1	update with new copy op	2019-02-07 01:31:00 -06:00
Chao Liu	07f16673c9	add lds double buffer for cnhw implicit gemm	2019-02-07 00:56:53 -06:00
Chao Liu	c866773642	unroll some loop, register double buffer gemm	2019-02-06 23:44:21 -06:00
Chao Liu	1b323316a8	add another blockwise gemm	2019-02-06 23:10:08 -06:00
Chao Liu	5e77650415	fixed LDS alignment bug	2019-02-06 01:54:13 -06:00
Chao Liu	079d63a788	bug fixes	2019-02-05 23:19:57 -06:00
Chao Liu	42f4c7fd56	refactor	2019-02-05 18:04:23 -06:00
Chao Liu	6614729a68	add another version of blockwise 2d copy, refactor	2019-02-05 17:06:53 -06:00
Chao Liu	4b616aad52	refactor	2019-02-05 00:51:37 -06:00
Chao Liu	0741a8ab88	working on reducing index calculation...	2019-02-04 17:16:28 -06:00
Chao Liu	9bbe9073ab	refactor	2019-02-04 15:40:34 -06:00
Chao Liu	3439e4b5b7	padding works (sort of), but code looks ugly. Tuned some resnet configs	2019-01-25 02:50:28 -06:00
Chao Liu	8bd6ea1a97	improve implicit gemm NCHW, SRCK, NKHW, and tuned	2019-01-24 22:24:30 -06:00
Chao Liu	1de6fd0753	fixed a bug, and refactored	2019-01-24 21:20:29 -06:00
Chao Liu	1410850ecc	add another implicit gemm: CHWN, CSRK, KHWN	2019-01-24 21:03:21 -06:00
Chao Liu	bd811e2c20	refactor	2019-01-24 16:15:51 -06:00
Chao Liu	c39c573eb8	refactor	2019-01-24 16:02:24 -06:00
Chao Liu	c9af4dece0	implicit gemm: LDS double buffer	2019-01-24 14:28:46 -06:00
Chao Liu	1f3870ca19	another version of blockwise 2d tensor copy	2019-01-23 16:42:57 -06:00
Chao Liu	e9ac4855f8	tune	2019-01-21 16:38:13 -06:00
Chao Liu	b5b4fd28ed	refactor	2019-01-21 15:33:34 -06:00
Chao Liu	c64f63d5ec	refactor	2019-01-21 11:36:31 -06:00
Chao Liu	2096847297	add 2nd variation of implicit gemm	2019-01-20 21:14:35 -06:00
Chao Liu	3bd51021ab	tune implicit_gemm	2019-01-17 00:01:21 -06:00
Chao Liu	216e3da609	bug fix and tune implicit gemm	2019-01-16 23:24:49 -06:00
Chao Liu	caf4d7e6f5	refactor	2019-01-16 16:11:08 -06:00
Chao Liu	5872b710df	refactor	2019-01-16 15:45:02 -06:00
Chao Liu	2b52fbd24a	bug fix	2019-01-16 12:58:44 -06:00
Chao Liu	ff7a62198d	refactor	2019-01-16 11:58:12 -06:00
Chao Liu	89ee259752	adding implicit gemm	2019-01-16 02:44:10 -06:00
Chao Liu	913afaeb5d	adding implicit gemm	2019-01-16 02:01:56 -06:00
Chao Liu	e7b8705b91	adding implicit gemm	2019-01-15 18:11:41 -06:00
Chao Liu	84d9802d30	adding implicit gemm	2019-01-15 00:11:30 -06:00
Chao Liu	aa0199a31c	adding implicit gemm	2019-01-14 11:13:36 -06:00
Chao Liu	dc60d16962	adding implicit gemm	2019-01-09 19:12:22 -06:00
Chao Liu	0597116330	refactor	2019-01-09 19:11:45 -06:00
Chao Liu	df228b3cf5	refactor	2019-01-08 16:56:46 -06:00
Chao Liu	0b8e67ef08	refactor	2019-01-08 14:05:03 -06:00
Chao Liu	ac1f62be3f	refactor	2019-01-07 23:01:41 -06:00
Chao Liu	3dbd47252c	added threadwise tensor reorder operation	2019-01-04 15:34:13 -06:00
Chao Liu	21c918162e	added blockwise tensor reorder operation	2019-01-04 14:48:57 -06:00