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
1,806 Commits 743 Branches 38 Tags
342df944449ec84ec374fb8896fa754cb9664614
Commit Graph

1806 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Muhammed Emin Ozturk
342df94444 CkProfiler StreamK GemmUniversal Fix and Split Gemm_universal Test Redo PR #2044 (#2070) 
* fix and split gemm_universal test


* Update test_gemm_universal_streamk_ut_cases_fp8.inc

[ROCm/composable_kernel commit: 74fda2e796]
 2025-04-11 10:17:29 -07:00
jakpiase
addcd203eb [CK_TILE] Add 2:4 structured sparsity support for fp16 gemm (#1957) 
* add structured sparsity fp16 support for gemm

* added reviewer suggestions

* update changelog

* update changelog

* add reviewers suggestions

* Minor fix

* clang fix

* fix doxygen

[ROCm/composable_kernel commit: 6c61f4d237]
 2025-04-11 12:18:26 +02:00
slippedJim
cca9cca699 add fmha fwd splitkv receipt for aiter c++ api (#2068) 
* add s_randval for c++ api

* Fix bug of bias in splitkv

---------

Co-authored-by: rocking <ChunYu.Lai@amd.com>

[ROCm/composable_kernel commit: 5f885d2b7a]
 2025-04-10 23:21:13 +08:00
Juan Manuel Martinez Caamaño
c4e0659e51 Replace inline assembly with builtins in FHMA (#2067) 
* Replace inline assembly with builtins in FHMA

---------

Co-authored-by: illsilin <Illia.Silin@amd.com>

[ROCm/composable_kernel commit: f14e648e7c]
 2025-04-10 09:48:37 +02:00
Illia Silin
9cc561987f enable gfx115x support (#2065) 
[ROCm/composable_kernel commit: 3e6d21adeb]
 2025-04-09 10:06:42 -07:00
MHYang-gh
4e5b335b5e Make buffer coherence configurable in tensor view (#2041) 
* Make buffer coherence configurable in tensor view

* Fix clang-format for tensor_view.hpp

[ROCm/composable_kernel commit: 03ce8729fd]
 2025-04-08 15:34:11 -07:00
valarLip
aa6eeb7f66 add passthrough for int32->float32 (#2062) 
[ROCm/composable_kernel commit: 2c563fecf7]
 2025-04-08 15:16:30 -07:00
Khushbu Agarwal
8a65c29ffb New instances for gemm_multiply_multiply_weightpreshuffle operator (#2061) 
* Add new instances for weight_preshuffle for f8->bf16

* Add new instances for weight_preshuffle for f8->f16

* clang formatted

---------

Co-authored-by: Khushbu Agarwal <khuagar@amd.com>
Co-authored-by: Thomas Ning <Thomas.Ning@amd.com>

[ROCm/composable_kernel commit: 263ff689e0]
 2025-04-08 15:14:53 -07:00
spolifroni-amd
6491cd870a fixed broken github link (#2063) 
[ROCm/composable_kernel commit: 2c8132126c]
 2025-04-08 10:20:31 -07:00
dependabot[bot]
409f3dccc6 Bump rocm-docs-core from 1.18.1 to 1.18.2 in /docs/sphinx (#2047) 
Bumps [rocm-docs-core](https://github.com/ROCm/rocm-docs-core) from 1.18.1 to 1.18.2.
- [Release notes](https://github.com/ROCm/rocm-docs-core/releases)
- [Changelog](https://github.com/ROCm/rocm-docs-core/blob/develop/CHANGELOG.md)
- [Commits](https://github.com/ROCm/rocm-docs-core/compare/v1.18.1...v1.18.2)

---
updated-dependencies:
- dependency-name: rocm-docs-core
  dependency-version: 1.18.2
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

[ROCm/composable_kernel commit: b12cd6580b]
 2025-04-08 09:06:38 -07:00
Max Podkorytov
1eec5cc073 simplify generate_tuple (#2043) 
[ROCm/composable_kernel commit: 6ce0797dad]
 2025-04-08 09:00:51 -07:00
aledudek
a42081842c [CK_TILE] Fix GEMM Memory Pipeline (#2034) 
* [CK_TILE] Fix GEMM Memory Pipeline

* Fix transpose tile

* Add comments

[ROCm/composable_kernel commit: 80aae6119b]
 2025-04-08 12:40:04 +02:00
Illia Silin
09ace4ab2f Fix a couple of CI issues. (#2050) 
* fix jenkins jobs

* fix perf log name for gfx908

* only run gemm perf tests on gfx908

[ROCm/composable_kernel commit: 72c0261ef1]
 2025-04-07 12:48:34 -07:00
Illia Silin
ad65c947c9 fix codegen issues (#2052) 
[ROCm/composable_kernel commit: 1793228422]
 2025-04-07 07:08:39 -07:00
Illia Silin
6346304629 Revert "CkProfiler StreamK GemmUniversal Fix and Split Gemm_universal Test (…" (#2054) 
This reverts commit cddc580bae98ff61b4d429c0ba2fe510e060a6de.

[ROCm/composable_kernel commit: 29f7266216]
 2025-04-07 06:49:36 -07:00
slippedJim
e27f844849 Add new receipt (#2055) 
[ROCm/composable_kernel commit: 5a22b61de5]
 2025-04-07 14:18:01 +08:00
Khushbu Agarwal
09792fa112 file clang formatted (#2053) 
[ROCm/composable_kernel commit: 3bda57c204]
 2025-04-03 16:55:49 -07:00
Khushbu Agarwal
844730776f Documentation for newly added struct (#2051) 
[ROCm/composable_kernel commit: b443056a26]
 2025-04-03 16:24:34 -07:00
Illia Silin
6d90b2eb50 Split env.hpp header from the ck.hpp header. (#2049) 
* split env.hpp out of main headers

* fix namespace logic

[ROCm/composable_kernel commit: 572cd820ce]
 2025-04-03 15:30:21 -07:00
Muhammed Emin Ozturk
d451d52350 CkProfiler StreamK GemmUniversal Fix and Split Gemm_universal Test (#2044) 
* fix and split gemm_universal test

* clang

* Update test_gemm_universal_ut_cases_bf16.inc

* Update test_gemm_universal_xdl_bf16.cpp

* Update test_gemm_universal_ut_cases_fp16.inc

[ROCm/composable_kernel commit: 7142d8003c]
 2025-04-03 14:22:43 -07:00
Khushbu Agarwal
b85b103194 [New] Build up the feature of CK Tile GEMM CodeGen (#1994) 
* New branch for codegen changes

* Fix verify function for int4

* pk_int4 codegen

* Update to review comments

* Remove codegen directory and rename filenames

* Remove extra files; clean up CMake file

* New branch for codegen changes

* Fix verify function for int4

* pk_int4 codegen

* Update to review comments

* Remove codegen directory and rename filenames

* Remove extra files; clean up CMake file

* code changes for single instance

* config file rename, added few more combinations in json file

* Fix cmake file

* Addressing review comments

* Reverting files changed by merge to develop

---------

Co-authored-by: ThomasNing <thomas.ning@amd.com>

[ROCm/composable_kernel commit: fed0709121]
 2025-04-03 11:54:12 -07:00
Thomas Ning
05e13817ba Add the MI355 support for CK TILE GEMM (#2046) 
* Get the root cause of the ck tile gemm failing on mi355

* Fix the ck tile gemm on MI355

* delete the debug info

[ROCm/composable_kernel commit: 50d1f8ff90]
 2025-04-03 11:48:54 -07:00
Rostyslav Geyyer
1716380358 Add FP16/BF16<->FP8/BF8 conversions (#2035) 
* Move conversion functions and add missing conversions

* Add tests

* Add missing conversions

* Add missing conversions

* Add bf8 tests

* Update clipping for vectors

* Add missing conversions

* Add bf16 fp8 tests

* Add bf16 bf8 tests

* Fix device conversion

* Fix conversions

* Fix vector use

* Minor fix

* Add a workaround flag

* Add a workaround flag for bf16 conversion

* Add another workaround

* Add a workaround for fp16 to bf8 conversion

* Update type alias

* Add docstrings and missing wrappers

* Fix if defined macros

* Fix more if defined macros

* Add comments

* Remove __host__ specifier

* Add a gfx950 guard

* Update function naming

[ROCm/composable_kernel commit: 265af71a71]
 2025-04-03 12:42:03 -05:00
aledudek
7a78bc823a Post-merge changes for fully async args copy in ck grouped gemm (#1991) 
* Post-merge changes for fully async args copy in ck grouped gemm

* Post-merge documentation and naming changes

* Build fix and updated changelog

* Revised comments

[ROCm/composable_kernel commit: 9329432f6c]
 2025-04-03 13:35:43 +02:00
Bartłomiej Kocot
49565538fe Add support for GKCYX grouped conv weight (#2023) 
* Grouped conv bwd weight GKCYX support

* fix and changelog

* fix

* fix

* fixes

* comments

* fix

[ROCm/composable_kernel commit: 2ccf914888]
 2025-04-02 23:59:49 +02:00
Adam Osewski
0a607256cf Basic docs for universal gemm & ck-tile gemm. (#2014) 
* Basic docs for universal gemm & ck-tile gemm.

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com>

* Update include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com>

* Update include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp

Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com>

* Reviewers suggestions.

* Align tparam names in doc with class tparams.

* More reviewers fine tuning ;)

---------

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>
Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com>

[ROCm/composable_kernel commit: e5ad48a784]
 2025-04-02 11:03:40 +02:00
Bartłomiej Kocot
50fb390d6f Grouped conv backward data GKCYX support (#2029) 
* Grouped conv backward data GKCYX support

* profiler

* Converter

* split instances

[ROCm/composable_kernel commit: 8c0ab61ece]
 2025-04-01 13:24:38 -07:00
Bartłomiej Kocot
b417137276 Grouped conv fwd v3 fix for SplitN an G > 1 (#2038) 
* Grouped conv fwd v3 fix for SplitN an G > 1

* Remove int8 large test

* Retore int8 test

[ROCm/composable_kernel commit: ec742908bd]
 2025-04-01 13:19:35 -07:00
Seunghoon Lee
0ac4c37028 Fix Windows build. (#2012) 
* Remove duplicate using uint64_t.

* Cast before shift.

[ROCm/composable_kernel commit: df32020f93]
 2025-04-01 12:22:10 -07:00
Max Podkorytov
70ad8571c2 add a fast compilation path for static for (0..N) (#2005) 
* add a fast compilation path for static for (0..N)

* Update functional2.hpp

add comment and put range applier into detail namespace

* Update functional.hpp

ditto for ck-tile

* prettify

* prettify more

* add comment

* clang-format

[ROCm/composable_kernel commit: c59a8bb206]
 2025-04-01 12:06:25 -07:00
Bartłomiej Kocot
a9ec282dba Improve compilation time for grouped conv fwd (#2039) 
* Improve compilation time for grouped conv fwd

* Fix

[ROCm/composable_kernel commit: 6355ee7ca5]
 2025-04-01 07:11:42 -07:00
Muhammed Emin Ozturk
30e5c8cb49 f8/bf16 GEMM Stream-K (#1879) 
[ROCm/composable_kernel commit: dd4c12b155]
 2025-03-31 20:30:17 -06:00
jefyang1
92f6e02b96 Fix gemm universal and grouped_conv_fwd test failures on gfx950 (#2031) 
[ROCm/composable_kernel commit: 16b15e336a]
 2025-03-31 09:20:52 -07:00
Adel Johar
de769ac359 Docs: Add precision support reference page (#1973) 
* Docs: Add precision support reference page

* edit of the precision type content

* added more description on scalars

---------

Co-authored-by: spolifroni-amd <sandra.polifroni@amd.com>
Co-authored-by: Aviral Goel <aviral.goel@amd.com>

[ROCm/composable_kernel commit: fc073b483e]
 2025-03-28 08:12:27 -06:00
rocking
2b657d9a2c Reduce redundant space in bias tensor (#2024) 
Co-authored-by: Po Yen Chen <PoYen.Chen@amd.com>

[ROCm/composable_kernel commit: 8a20b62e91]
 2025-03-28 21:58:06 +08:00
felix
817752cdb4 hotfix fix sorting int64 (#2025) 
* fix sorting int64

* clang format

* fix example issue

* update WA issue #

---------

Co-authored-by: coderfeli <coderfeli@163.com>
Co-authored-by: carlushuang <carlus.huang@amd.com>

[ROCm/composable_kernel commit: a82f338fb9]
 2025-03-28 11:31:52 +08:00
Illia Silin
5bfbcb38ac add gfx950 to default targets for rocm6.4+ (#2032) 
[ROCm/composable_kernel commit: d142e15f5e]
 2025-03-27 18:48:47 -07:00
spolifroni-amd
98cc377db2 creation of install doc and refactor of doc in general (#1908) 
* creation of install doc and refactor of doc in general

* updates based on review comments

* updated based on review comments

* updated readme and contributors markdown

* added extra note to not use -j on its own

* added note about smoke tests and regression tests

* made changes as per Illia's feedback

---------

Co-authored-by: Aviral Goel <aviral.goel@amd.com>

[ROCm/composable_kernel commit: a426f67301]
 2025-03-27 15:13:18 -06:00
felix
00342e69a2 ckmoe: change cmake; use smaller shape for i4 (#2027) 
* change cmake; use smaller shape for i4

* fix pki4 run

* fix typo

* fix runtime arch logic for moe_gemm2 example

---------

Co-authored-by: coderfeli <coderfeli@163.com>
Co-authored-by: illsilin <Illia.Silin@amd.com>

[ROCm/composable_kernel commit: 36d50de50e]
 2025-03-27 09:04:31 -07:00
Rostyslav Geyyer
48fa126a9e Add MX FP4 device conversion tests (#1889) 
* Add conversion tests

* Fix ctor

* Fix nan logic

* Fix conversion logic

* Permute packed f4_t values

* Fix conversion to float, repack vector elements

* Fix device tests

* Permute elements in a vector

* Add a repro test

* Add a conversion for a repro test

* Update test vectors

* Update conversion

* Fix the test

* Update test vector generator

* Fix vector sr conversion

* Permute conversion args

* Update conversion

* Test

* Fix packing

* Simplify conversion function

* Pack conversion in a loop

* Pack conversion in a loop

* Pack another conversion in a loop

* Pack one more conversion in a loop

* Pack the last conversion in a loop

* Clean up

* Add printf to fix intrinsic

* Add a sw-based workaround

[ROCm/composable_kernel commit: 441343a23d]
 2025-03-26 19:23:01 -05:00
Illia Silin
d424bbe440 Disable all pk_i4 tests for all targets except gfx942/950. (#2022) 
* only build gemm_fp8_pk_i4 examples for gfx942/950

* fix cmake logic

* moved the architecture check to IsSupported function

* Revert "moved the architecture check to IsSupported function"

This reverts commit 056d2a08b3.

* disable all pk_i4 tests for targets other than gfx942/950

* fix cmake logic

[ROCm/composable_kernel commit: 23a949706c]
 2025-03-26 15:15:57 -07:00
Bartłomiej Kocot
f967fd7296 Add support for GKCYX grouped conv fwd (#2015) 
* Add support for GKCYX grouped conv fwd

* fixes

* fix

* changelog

* Fixes

[ROCm/composable_kernel commit: 54c81a1fcf]
 2025-03-26 21:13:38 +01:00
Illia Silin
ba16351a03 fix clang format (#2021) 
[ROCm/composable_kernel commit: fd915b83f7]
 2025-03-26 09:42:10 -07:00
Mirza Halilčević
97b32a1f18 Add default arguments for prologue and epilogue. (#2020) 
[ROCm/composable_kernel commit: 21e0ca197d]
 2025-03-26 09:28:40 -07:00
Illia Silin
9a88442eba Make sure gemm_fp8_pk_i4 examples only build and run on gfx942/950. (#2010) 
* only build gemm_fp8_pk_i4 examples for gfx942/950

* fix cmake logic

* moved the architecture check to IsSupported function

* Revert "moved the architecture check to IsSupported function"

This reverts commit 056d2a08b3.

[ROCm/composable_kernel commit: 99b2bbc1d6]
 2025-03-25 14:43:38 -07:00
Andriy Roshchenko
3f06d019ba MX GEMM examples with FP8, FP16, and E8M0 scales (#2016) 
* Add `scalar_type` specification for E8M0 exponent

* Specialize `nnvb_data_t_selector` for E8M0 exponent

* Remove partial specializations for `scalar_type` of `non_native_vector_base` template

* Reword command line helper string

* Create MX GEMM examples for different scales


[ROCm/composable_kernel commit: 72d888821c]
 2025-03-25 15:33:03 -06:00
Illia Silin
aaf6a0343d Enable ClangBuildAnalizer when doing ninja build traces. (#2009) 
* enable ClangBuildAnalizer when doing ninja traces

* add branch and date to clang build log name

* fix jenkins syntax

* fix jenkins syntax once more

* fix jenkins syntax once more

* simplify the clang_build log name

* simplify the clang_build log name further

[ROCm/composable_kernel commit: 44c093ba0c]
 2025-03-25 12:27:04 -07:00
Max Podkorytov
a70c642a78 use fast path for sequence generation in old CK (#1993) 
[ROCm/composable_kernel commit: 1a58522f01]
 2025-03-25 11:28:44 -07:00
ruanjm
fcbf9630fe [CK_TILE] Improve RMS/Layer Normalization 2 Pass Pipeline Performance (#1861) 
* 50ms -> 28ms

* Fix bug in non fuse_add_store cases

* Fine tuned setting for 2 pass pipeline

* adjust workload

* remove unnecessary change

* add layernorm

* Adding output quant and unquant results at the same time.

* fix test

* fix format

* tune for cases 128x640 and 128x1024

* bug ifx

[ROCm/composable_kernel commit: d49abdaa87]
 2025-03-25 20:09:45 +08:00
Illia Silin
9dcfd79dcb Split up data_type header. (#1996) 
* split fp64 vector data type

* add missing header

* move e8m0 structs

* split off numeric_utils header

* fix typo

* split off numeric limits header

* update data_type header

* fix clang format

* split off vector type header

* fix clang format

* fix typo for binary_inf

[ROCm/composable_kernel commit: d2eab23958]
 2025-03-24 15:08:54 -07:00