Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Cleanup and AVX10 Support #22

Draft
wants to merge 65 commits into
base: master
Choose a base branch
from
Draft

Cleanup and AVX10 Support #22

wants to merge 65 commits into from

Commits on Jul 23, 2024

  1. Add half precision

    @davschneller
    davschneller committed Jul 23, 2024
    Configuration menu
    Copy the full SHA
    0ce03c9 View commit details
    Browse the repository at this point in the history

  2. Merge remote-tracking branch 'origin/master' into davschneller/half

    @davschneller
    davschneller committed Jul 23, 2024
    Configuration menu
    Copy the full SHA
    0716375 View commit details
    Browse the repository at this point in the history

  3. Fix half precision, add full AVX10 support

    @davschneller
    davschneller committed Jul 23, 2024
    Configuration menu
    Copy the full SHA
    7dc330e View commit details
    Browse the repository at this point in the history

  4. Forward half precision, refactor a bit

    @davschneller
    davschneller committed Jul 23, 2024
    Configuration menu
    Copy the full SHA
    bfb3677 View commit details
    Browse the repository at this point in the history

  5. Forward new vector lengths

    @davschneller
    davschneller committed Jul 23, 2024
    Configuration menu
    Copy the full SHA
    56287e7 View commit details
    Browse the repository at this point in the history

Commits on Aug 6, 2024

  1. Prepare for BFLOAT16, AVX512 masks, single-precision NEON

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    aa29c9b View commit details
    Browse the repository at this point in the history

  2. Fix build

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    9265e89 View commit details
    Browse the repository at this point in the history

  3. Autoselect bk

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    0b51b2f View commit details
    Browse the repository at this point in the history

  4. Fix register counting

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    3dbc276 View commit details
    Browse the repository at this point in the history

  5. Add extra strategy for hsw

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    3ad7718 View commit details
    Browse the repository at this point in the history

  6. Adjust block condition

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    8e604f2 View commit details
    Browse the repository at this point in the history

  7. Fix register offset

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    c51fa7c View commit details
    Browse the repository at this point in the history

  8. Compact dense matrix multiplication

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    7152b9c View commit details
    Browse the repository at this point in the history

  9. Bugfixing

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    e8fb559 View commit details
    Browse the repository at this point in the history

  10. Fix pointers

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    1692187 View commit details
    Browse the repository at this point in the history

  11. Bugfixes

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    cefc13a View commit details
    Browse the repository at this point in the history

  12. Rollback condition extension

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    3f67d54 View commit details
    Browse the repository at this point in the history

  13. Begin adjusting tests

    @davschneller
    davschneller committed Aug 6, 2024
    Configuration menu
    Copy the full SHA
    f19ea1b View commit details
    Browse the repository at this point in the history

Commits on Aug 15, 2024

  1. Prepare for better sparsity loading

    @davschneller
    davschneller committed Aug 15, 2024
    Configuration menu
    Copy the full SHA
    23cbc2c View commit details
    Browse the repository at this point in the history

  2. Unroll again

    @davschneller
    davschneller committed Aug 15, 2024
    Configuration menu
    Copy the full SHA
    27e934d View commit details
    Browse the repository at this point in the history

Commits on Aug 19, 2024

  1. Update and extend tests

    @davschneller
    davschneller committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    927e4f6 View commit details
    Browse the repository at this point in the history

Commits on Oct 1, 2024

  1. Test cleanup

    @davschneller
    davschneller committed Oct 1, 2024
    Configuration menu
    Copy the full SHA
    56f9d3d View commit details
    Browse the repository at this point in the history

Commits on Oct 10, 2024

  1. Fix test names

    @davschneller
    davschneller committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    dae3e37 View commit details
    Browse the repository at this point in the history

Commits on Oct 11, 2024

  1. Fix bugs in generators

    @davschneller
    davschneller committed Oct 11, 2024
    Configuration menu
    Copy the full SHA
    0438eef View commit details
    Browse the repository at this point in the history

  2. Fix codegen bugs, prepare for AVX512 mask codegen

    @davschneller
    davschneller committed Oct 11, 2024
    Configuration menu
    Copy the full SHA
    2a64dfd View commit details
    Browse the repository at this point in the history

Commits on Oct 12, 2024

  1. Merge test scripts

    @davschneller
    davschneller committed Oct 12, 2024
    Configuration menu
    Copy the full SHA
    17f23f9 View commit details
    Browse the repository at this point in the history

Commits on Oct 18, 2024

  1. Adjust tests a bit more, and correct instructions

    @davschneller
    davschneller committed Oct 18, 2024
    Configuration menu
    Copy the full SHA
    1d2656a View commit details
    Browse the repository at this point in the history

  2. Output more error infos

    @davschneller
    davschneller committed Oct 18, 2024
    Configuration menu
    Copy the full SHA
    1edd02c View commit details
    Browse the repository at this point in the history

Commits on Oct 19, 2024

  1. Generalize broadcasting interface a tiny bit

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    ad887eb View commit details
    Browse the repository at this point in the history

  2. Avoid 0-adds

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    b98e8f0 View commit details
    Browse the repository at this point in the history

  3. Unique register clobbering

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    386266b View commit details
    Browse the repository at this point in the history

  4. Update blocksize args

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    fb70f32 View commit details
    Browse the repository at this point in the history

  5. Clean up block sizes

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    535c44e View commit details
    Browse the repository at this point in the history

  6. Clean up module finder

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    97fe82c View commit details
    Browse the repository at this point in the history

  7. Clean up unit tests

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    5d92250 View commit details
    Browse the repository at this point in the history

  8. Prepare for larger-k ARM kernels

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    57d1b47 View commit details
    Browse the repository at this point in the history

  9. Fix ARM kernel register block loading

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    a321480 View commit details
    Browse the repository at this point in the history

  10. Add some more comments

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    b2ceb79 View commit details
    Browse the repository at this point in the history

  11. Fix unit tests

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    2fc28a9 View commit details
    Browse the repository at this point in the history

  12. Rework test output

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    73988b2 View commit details
    Browse the repository at this point in the history

  13. Add missing classmethod modifiers

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    5b14a9d View commit details
    Browse the repository at this point in the history

  14. Remove old scripts references

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    985634c View commit details
    Browse the repository at this point in the history

  15. Fix more class-move bugs

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    9f99416 View commit details
    Browse the repository at this point in the history

  16. Fix ARM_SVE codegen

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    e3a4fc2 View commit details
    Browse the repository at this point in the history

  17. Yet even more class-move fixes

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    07a487b View commit details
    Browse the repository at this point in the history

  18. Add large matrix support and tests

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    207a373 View commit details
    Browse the repository at this point in the history

  19. Update readme

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    fddef2e View commit details
    Browse the repository at this point in the history

  20. Enable AVX512 mask support

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    1dab328 View commit details
    Browse the repository at this point in the history

  21. Fix tests once more

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    9bc9831 View commit details
    Browse the repository at this point in the history

  22. Bump version number

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    8d4f51f View commit details
    Browse the repository at this point in the history

  23. Fix large matrix offsetting

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    88a47fb View commit details
    Browse the repository at this point in the history

  24. Fix memory load masking

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    af3fbda View commit details
    Browse the repository at this point in the history

  25. Working larger B registers

    @davschneller
    davschneller committed Oct 19, 2024
    Configuration menu
    Copy the full SHA
    d87a5af View commit details
    Browse the repository at this point in the history

Commits on Oct 20, 2024

  1. Extend inline ARM_SVE broadcast fma

    @davschneller
    davschneller committed Oct 20, 2024
    Configuration menu
    Copy the full SHA
    c480249 View commit details
    Browse the repository at this point in the history

  2. Add Cube and MaxK scripts

    @davschneller
    davschneller committed Oct 20, 2024
    Configuration menu
    Copy the full SHA
    8ee2367 View commit details
    Browse the repository at this point in the history

  3. Towards masking for k

    @davschneller
    davschneller committed Oct 20, 2024
    Configuration menu
    Copy the full SHA
    a8a1a62 View commit details
    Browse the repository at this point in the history

Commits on Oct 21, 2024

  1. Fix ARM_SVE blocksizes

    @davschneller
    davschneller committed Oct 21, 2024
    Configuration menu
    Copy the full SHA
    096f190 View commit details
    Browse the repository at this point in the history

  2. Change load order for the B matrix

    @davschneller
    davschneller committed Oct 21, 2024
    Configuration menu
    Copy the full SHA
    e17d138 View commit details
    Browse the repository at this point in the history

Commits on Oct 25, 2024

  1. Set MaxK to default

    @davschneller
    davschneller committed Oct 25, 2024
    Configuration menu
    Copy the full SHA
    aa95d58 View commit details
    Browse the repository at this point in the history

Commits on Oct 27, 2024

  1. Bugfixes

    @davschneller
    davschneller committed Oct 27, 2024
    Configuration menu
    Copy the full SHA
    6193c09 View commit details
    Browse the repository at this point in the history

  2. Remove old gitignores

    @davschneller
    davschneller committed Oct 27, 2024
    Configuration menu
    Copy the full SHA
    5fcf5fd View commit details
    Browse the repository at this point in the history

Commits on Oct 30, 2024

  1. Fix k predicates

    @davschneller
    davschneller committed Oct 30, 2024
    Configuration menu
    Copy the full SHA
    13bd622 View commit details
    Browse the repository at this point in the history

  2. Fix vm overhead

    @davschneller
    davschneller committed Oct 30, 2024
    Configuration menu
    Copy the full SHA
    9c430fc View commit details
    Browse the repository at this point in the history

  3. Fix AVX512 masking

    @davschneller
    davschneller committed Oct 30, 2024
    Configuration menu
    Copy the full SHA
    885c9d4 View commit details
    Browse the repository at this point in the history

  4. Hotfix overlarge bm

    @davschneller
    davschneller committed Oct 30, 2024
    Configuration menu
    Copy the full SHA
    a0ce34a View commit details
    Browse the repository at this point in the history