ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-06-29 11:16:59 +00:00
Code Issues Packages Projects Releases Wiki Activity
1,865 Commits 776 Branches 40 Tags
mi355_transpose_load_dev
Commit Graph

1865 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
joye
b406b584d7 fix clang format check 2025-04-28 08:56:00 +08:00
joye
292f89c860 fix a transpose index issue 2025-04-27 17:39:52 +08:00
joye
bd091d9a88 fix a transpose index issue 2025-04-27 17:33:32 +08:00
joye
de9407ed93 fix 16x16 related dimension transpose 2025-04-27 02:44:49 -05:00
joye
57e8e34705 exchange the iteration order 2025-04-26 21:37:59 -05:00
joye
944f36efb8 update transpose load example 2025-04-27 10:13:46 +08:00
joye
25ab38f913 fix transpose related codes 2025-04-27 09:55:29 +08:00
joye
a823e00dc0 delete unused variables 2025-04-25 14:47:49 +08:00
joye
a6564da629 update output tensor distribution 2025-04-25 14:45:45 +08:00
joye
119a8e0e16 hack for transpose 16x16 2025-04-25 00:56:59 -05:00
joye
e2f3c95d24 miss output tile distribution mapping 2025-04-24 21:55:03 -05:00
joye
6beb585dad update tile transpose 2025-04-24 20:36:09 -05:00
joye
efa7243ee5 transpose load enable 2025-04-24 19:19:44 -05:00
joye
df9769afba can pass; but no logic 2025-04-24 19:05:47 -05:00
joye
90a4501869 add some fix 2025-04-24 18:29:38 +08:00
joye
34040f43b6 update some codes 2025-04-24 17:53:19 +08:00
joye
8d75983536 fix a distribution issue 2025-04-24 02:24:31 -05:00
joye
6893165818 add some fixes 2025-04-24 01:20:35 -05:00
joye
afb1cec9c4 update transpose load logic 2025-04-24 11:14:07 +08:00
joye
3918a35870 Merge branch 'mi355_transpose_load_dev' of https://github.com/ROCm/composable_kernel into mi355_transpose_load_dev 2025-04-24 08:28:14 +08:00
joye
aaab4aacc5 Merge branch 'develop' of https://github.com/ROCm/composable_kernel into mi355_transpose_load_dev 2025-04-24 08:27:53 +08:00
joye
c862437f05 fix some issues 2025-04-23 19:27:29 -05:00
carlushuang
5487289fc4 [CK_TILE] support gfx950 matrix core in 01_fmha fwd (#2110) 
* gfx950 01_fmha fwd

* fix comment

---------

Co-authored-by: Thomas Ning <Thomas.Ning@amd.com>
 2025-04-23 12:40:18 -07:00
John Afaganis
854159fd00 Update CODEOWNERS (#2119) 2025-04-23 10:25:41 -07:00
joye
acce2df3bf fix some compile errors 2025-04-23 05:27:52 -05:00
joye
e14a16359f add transpose load; no real logic 2025-04-23 16:55:13 +08:00
lalala-sh
39ba03f25d Moe gemm activation (#2026) 
* fix useless code and remove usless oob

* clang format

* fix coredump in e2e test

* fix2

* fix clang format

* fix output oob

* impl int64 but result not correct

* int64 index ok now

* input output all ok

* fix uint32

* revert v1 test

* use uint32

* mork to support 13w tokens

* moe sorting fix moebuf

* fix merge

* update moe api fix aiter build

* fix buid

* fuse silu

* silu ok

* acale ok

* add silu

* change code

* gemm2 ok

* gufusion compatible ok, fix warnings

* gu fusion for m32 m64 ok

* support bf16 cshuffle

* i4 gemm2 ok

* i4 gemm2 ok and i4 gemm1 build

* 16x16 run ok

* change flops; change cshuffle dtype

* fuse gelu silu act in moe gemm1

* fp8 with act ready

* int4 act ready

* remove useless changes

* remove useless code change

* fix clang format

* add the arch limit of int4 moe gemm

* fuse moe activation

* fix fp8 16x16

* fix no quant case

* fix bugs

* fix fp8 gufusion bug

* remove useless comments

* refine activation code & complete moe example

* fix int8 bugs

* merge tkw1

---------

Co-authored-by: coderfeli <coderfeli@163.com>
Co-authored-by: feli <felix.li@amd.com>
Co-authored-by: illsilin <Illia.Silin@amd.com>
Co-authored-by: root <root@hjbog-srdc-51.amd.com>
Co-authored-by: Illia Silin <98187287+illsilin@users.noreply.github.com>
 2025-04-23 10:35:34 +08:00
Khushbu Agarwal
94662b02d0 Adding include directory in tile_engine (#2116) 2025-04-22 15:55:19 -07:00
Gino Lu
504f563f78 [CK-Tile] warp-gemm support for using V_MFMA_F32_16x16x32_BF16 (#2073) 
* draft v_mfma_f32_16x16x32_bf16

* fix error config and add debug code.

* Solve the CShuffle Problem

* draft v_mfma_f32_16x16x32_bf16

* fix error config and add debug code.

* Solve the CShuffle Problem

* fix error while testing new command

* Finished the feature of new mfma 16*16*32

* Addressed the comment

---------

Co-authored-by: ThomasNing <thomas.ning@amd.com>
 2025-04-22 15:52:36 -07:00
Rostyslav Geyyer
416e851584 Temporarily disable MX FP4 device tests (#2112) 2025-04-22 16:08:48 -05:00
Thomas Ning
0cca8fa28f GEMM Multiply Multiply Fix (#2102) 
* fix the type convert and increase the BF16 conversion + the profile comment

* fix the CI
 2025-04-22 01:13:22 -07:00
Thomas Ning
4bef60aa57 update code owner (#2113) 2025-04-21 13:53:03 -07:00
Muhammed Emin Ozturk
b092c18da7 MI308 fix for streamk 1-Tile floating point exception (#2101) 2025-04-21 11:44:07 -07:00
Thomas Ning
a738e43445 MFMA 16x16x32fp8 (#2103) 
* add mfma_16x16x32_fp8

* clang format code

* Finished the fix for gemm basic

* clang foramt

* rebuild CI

* recover gemm.hpp

* add MFMA 16*16*32bf8

---------

Co-authored-by: solin <bingzhou@amd.com>
 2025-04-21 10:21:35 -07:00
Illia Silin
ce61759538 fix daily gfx942 build (#2106) 2025-04-21 08:48:22 -07:00
Khushbu Agarwal
7cadf187e2 multi instance generation for CkTileEngine (#2080) 
* Add support for multi-instance verification, print detail for each instance, documentation fix

* clang formatted

* Added Readme file

* updated readme

* Addressing review comments

* clang formatted

* Updated ReadMe and GPU reference code

* simplified dispatch kernel code

* indentation
 2025-04-21 08:39:45 -07:00
solin
c318ec0778 fix CI build fail 2025-04-21 16:00:12 +08:00
lalala-sh
bcf5bb41be enable do top k weights in moe stage1 gemm (#2094) 
* add switch for mul topk weights

* fix bf16/f16 bugs

* complete
 2025-04-18 10:45:49 +08:00
Andriy Roshchenko
213b203a3c MX GEMM - Parameterized Test Template (#2088) 
* Tests for MX FP8 GEMM

* Improve documentation
 2025-04-16 19:56:00 -06:00
Andriy Roshchenko
da54464cce MX GEMM - Add MX BF8 example (#2071) 
* Add MX GEMM example for MX BF8

* Verified MX FP8 with 16x16x128 scale builtin

* Verify MX BF8 GEMM with BF16 output
 2025-04-16 15:25:02 -06:00
Illia Silin
3bb62f16cd Upgrade default docker to Ubuntu24.04 (#2090) 
* upgrade docker to Ubuntu24.04

* add break-system-packages flag to pip install

* fix dockerfile
 2025-04-16 12:10:15 -07:00
aledudek
7c32652e03 Add grouped conv fwd 3d GKCYX instances for f32, f16, bf16 (#2069) 
* Part1

* Add grouped conv fwd 3d GKCYX instances for f32, f16, bf16

* Add missing coma

* Add missing cpp instance files

* Fix 3d layout

* Add missing closing bracket

* Add missing comp x2 and part2 instances

* Fix typo in instance name

* fix

* Fix

---------

Co-authored-by: Bartlomiej Kocot <barkocot@amd.com>
 2025-04-16 11:00:55 +02:00
BingYuan.Zhou
eaf1f0bf3b [flatmm] implement basic fp16 flatmm (#2089) 
* [flatmm] implement basic fp16 flatmm

* fix CI build fail

---------

Co-authored-by: root <root@hjbog-srdc-50.amd.com>
Co-authored-by: solin <bingzhou@amd.com>
 2025-04-16 16:51:17 +08:00
felix
c5975529bb add preshuffle gemm fp16 (#2036) 
* add preshuffle gemm fp16

* clang format and test ok

* Update gemm_multiply_multiply_xdl_fp16_bpreshuffle.cpp

remove useless comments in example

* Update gemm_multiply_multiply_xdl_fp16_bpreshuffle.cpp

remove 2

---------

Co-authored-by: coderfeli <coderfeli@163.com>
 2025-04-16 10:53:21 +08:00
joyeamd
94d47b1680 fmha hdim256 vectorize improve (#2086) 
For hdim 256, will not have vectorized buffer load when seqlen % 256 != 0 and hdim % 256 = 0; this commit tries to solve this condition.
 2025-04-16 09:21:04 +08:00
Andriy Roshchenko
7106976a72 MX GEMM - New GEMM pipeline for MX data types (#2059) 
* Allow selection of mfma_scale instructions

* Read B tensor from LDS to VGPR in chunks of 16 in MFMA order

* Add constexpr and synchronize return type for `get_exponent_value`

* Pass scales by reference and add comments to `mfma_scale_f32_32x32x64`

* Add support for microscaling instructions in `XdlopsGemm`

* Fix `mfma_scale_f32_16x16x128f8f6f4` wrapper

* Remove software implementation of MX GEMM

* Make interface of `intrin_mfma_scale_f32_16x16x128f8f6f4<16, 16>` consistent with the other scale instruction

* Update README

* Updated CHANGELOG

* Remove unused static methods
 2025-04-15 17:17:07 -06:00
Illia Silin
d55c9cb313 Upgrade default docker image to ROCm6.4 release. (#2082) 
* upgrade to rocm6.4

* fix gfx10 generic target syntax

* use gfx1101 target for unit tests

* use gfx1201 target for unit tests

* do not use generic targets until 6.4.1 release

* update target list and dockerfile.compiler
 2025-04-14 16:41:47 -07:00
Mingtao Gu
56378f810f CK pk_i4_t test failures fix (SWDEV-518629) (#2075) 
* fix pk_i4_v3 tests failures in Unbuntu env.

* fix pk_i4_t tests failure on Unbuntu issues.

* some fixed.

---------

Co-authored-by: mtgu0705 <mtgu@amd.com>
 2025-04-14 16:58:57 +08:00
Thomas Ning
269f4f6af5 Solve the Static Encoding Pattern compile error when the tile size is too small (#2079) 2025-04-13 20:09:30 -07:00
Illia Silin
0d4f145078 Fix build issues for multiple targets. (#2077) 
* build for multiple targets on gfx942

* add missing ignore statements
 2025-04-11 12:12:53 -07:00