Skip to content

GitLab

Switch branch/tag
  1. 25 Apr, 2024 1 commit
    • ltqin's avatar
      Universal gemm flush cache (#1251) · f448d179
      ltqin authored 
      

* add flush cache to device op

* add flush cache parameter to ckProfiler

* change calculate size a and b method

* chang evaluation time method foro AVERAGE to MEDIAN

* format code

* adjust some code

* fix core dumped

* remove loop call flush icache in kernel

* remove loop(outer) call flush icache

---------
Co-authored-by: default avatarletaoqin <letaoqin@amd.com>
      f448d179
  3. 23 Apr, 2024 1 commit
  5. 22 Apr, 2024 1 commit
  7. 19 Apr, 2024 2 commits
  9. 18 Apr, 2024 3 commits
  11. 16 Apr, 2024 3 commits
  13. 15 Apr, 2024 1 commit
  15. 14 Apr, 2024 1 commit
    • Haocong WANG's avatar
      [GEMM] Gemm universal device operation (#1154) · f83e9701
      Haocong WANG authored 
      

* Optimize GEMM on MI200/300:
1. Add new blockwise gemm pipeline
2. Add irregular splitk intances

* clang format + typo fix

* Fix a bug

* initial commit

* Add more instances to irregular splitk

* blkgemm pipeline v1~4 prototype

* Sanity Checked. Known issue:
1. Poor performance of splitk
2. Register spill on blkgemmpipeline v3

* Sanity and Performance fix:
1. fix a bug related to sanity in grouped b2c mapping
2. fix a bug related to sanity and performance in splitk offset

* Sanity and API update:
1. Remove prefetch stage
2. Fix valid check bug
3, Add first gemm_universal instance into ckProfiler

* Add NN instances for gemm universal

* 1. Add NT instances for gemm_universal
2. Fix a bug about Kpadding in gemm_universal

* Fix a bug regarding padding Odd K number

* remove kernel print

* Fix KPadding bug...

* Update safety check

* another try to fix kpadding..

* Sanity checked

* new instances..

* clang format+typo fix

* remove clang format script's change

* Add non-hotloop compile option

* 1. Add fp16xfp8 example
2. pull packed convert f8 from pr1150

* Some miscs.. opt and fix

* Add pipeline description docs

* Split universal gemm instance library to cut profiler compiling time

* uncomment cmakefile

* Fix a bug caused by blockwise_gemm_pipe_v2

* reduce default splitk to 1

* Add 224x256x64 tile size

* update, including:
1. Experiment pipeline 5~7
2. Optimization for pipeline 4
3. Organized instance library

* temp save

* temp save

* Permuted lds layout, sanity and function checked

* clang format

* Move OOB check from RunRead to RunWrite, for better software pipeline.
TODO: agpr spill when NN layout

* clangformat

* A/B splitpipe scheduler for v3

* Fix two bugs

* bug fix

* fix a bug in oob check

* Example for mixed fp16_fp8 gemm

* Clean experimental code blocks

* Add mixed precision gemm into profiler

* tempsave

* optimize m/n major lds layout

* Add RRR GEMM  mixed precision instances

* Optimize f8 matrix transpose

* Add test_gemm_universal

* A/B spilt schedule for blkpip v5

* Take ds_read2 into iglp scheduling scheme

* format

* fixed cmake

* Add llvm-option into CI cmake flag

---------
Co-authored-by: default avatarJing Zhang <jizhan@amd.com>
      f83e9701
  17. 12 Apr, 2024 1 commit
  19. 11 Apr, 2024 3 commits
  21. 10 Apr, 2024 1 commit
  23. 09 Apr, 2024 3 commits
  25. 04 Apr, 2024 2 commits
  27. 03 Apr, 2024 1 commit
  29. 02 Apr, 2024 3 commits
    • Bartłomiej Kocot's avatar
      Introduce combined elementwise ops (#1217) · 9a194837
      Bartłomiej Kocot authored 
      * Introduce combined elementwise ops

* Introduce refrence elementwise
      9a194837
    • Illia Silin's avatar
      Split the instances by architecture. (#1223) · ae57e593
      Illia Silin authored 
      * parse examples inside the add_example_executable function

* fix the example 64 cmake file

* add xdl flag to the gemm_bias_softmax_gemm_permute example

* add filtering of tests based on architecture type

* enable test_grouped_gemm for gfx9 only

* enable test_transpose only for gfx9

* only linnk test_transpose if it gets built

* split the gemm instances by architectures

* split gemm_bilinear,grouped_conv_bwd_weight instances by targets

* split instances by architecture

* split grouped_conv instances by architecture

* fix clang format

* fix the if-else logic in group_conv headers

* small fix for grouped convolution instances

* fix the grouped conv bwd weight dl instances

* fix client examples

* only enable client examples 3 and 4 on gfx9

* set the gfx9 macro

* make sure the architecture macros are set by cmake

* use separate set of xdl/wmma flags for host code

* sinmplify the main cmake file

* add conv_fwd_bf8 instance declaration
      ae57e593
    • zjing14's avatar
      improved zeroing (#1221) · 303d4594
      zjing14 authored
      303d4594
  31. 27 Mar, 2024 1 commit
  33. 22 Mar, 2024 3 commits
  35. 21 Mar, 2024 1 commit
  37. 20 Mar, 2024 1 commit
  39. 19 Mar, 2024 1 commit
  41. 18 Mar, 2024 2 commits
    • Illia Silin's avatar
      update the changelog for ROCm6.1 release (#1205) · 9e011bcd
      Illia Silin authored 
      * update the changelog for ROCm6.1 release

* modifty the order of items in changelog, capitalize GEMMs
      9e011bcd
    • Illia Silin's avatar
      Re-enable the performance tracking in CI. (#1203) · bdcd0374
      Illia Silin authored 
      * test CK with rocm6.1 RC2

* add docker credentials for pull

* update the performance db name

* use environment variable for db name

* add rocm-llvm-dev package to ck docker

* turn off verification for daily performance runs

* do not stash ckProfiler on MI300 node

* add processing of mixed gemms to qa, fix parsing of splitk gemm logs

* fix the splitk gemm log file name

* turn the timing on for splitk gemm performance
      bdcd0374
  43. 15 Mar, 2024 1 commit
  45. 13 Mar, 2024 2 commits
  47. 12 Mar, 2024 1 commit