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

FPU operator #101

Merged
merged 32 commits into from
Jun 8, 2023
Merged

FPU operator #101

merged 32 commits into from
Jun 8, 2023

Commits on May 10, 2023

  1. FPUOp relevant changes from dev

    @wardvermeulen
    wardvermeulen committed May 10, 2023
    Configuration menu
    Copy the full SHA
    79651bf View commit details
    Browse the repository at this point in the history

Commits on May 11, 2023

  1. attempt at letting RowMajor work with FPUOp

    @wardvermeulen
    wardvermeulen committed May 11, 2023
    Configuration menu
    Copy the full SHA
    4b70b54 View commit details
    Browse the repository at this point in the history

  2. operater shape + mixed types

    @wardvermeulen
    wardvermeulen committed May 11, 2023
    Configuration menu
    Copy the full SHA
    f77d9c2 View commit details
    Browse the repository at this point in the history

  3. add compute and data type test

    @wardvermeulen
    wardvermeulen committed May 11, 2023
    Configuration menu
    Copy the full SHA
    d27d7a3 View commit details
    Browse the repository at this point in the history

Commits on May 12, 2023

  1. more performant because less registers needed

    @wardvermeulen
    wardvermeulen committed May 12, 2023
    Configuration menu
    Copy the full SHA
    352ba15 View commit details
    Browse the repository at this point in the history

Commits on May 13, 2023

  1. finished benchmarking, still work on cutlass

    @wardvermeulen
    wardvermeulen committed May 13, 2023
    Configuration menu
    Copy the full SHA
    49f74a9 View commit details
    Browse the repository at this point in the history

Commits on May 14, 2023

  1. add plot

    @wardvermeulen
    wardvermeulen committed May 14, 2023
    Configuration menu
    Copy the full SHA
    f0412db View commit details
    Browse the repository at this point in the history

  2. mixed precision fix

    @wardvermeulen
    wardvermeulen committed May 14, 2023
    Configuration menu
    Copy the full SHA
    0c98756 View commit details
    Browse the repository at this point in the history

  3. add config file

    @wardvermeulen
    wardvermeulen committed May 14, 2023
    Configuration menu
    Copy the full SHA
    c56e8a7 View commit details
    Browse the repository at this point in the history

  4. preparation for pr

    @wardvermeulen
    wardvermeulen committed May 14, 2023
    Configuration menu
    Copy the full SHA
    117adde View commit details
    Browse the repository at this point in the history

  5. small fix to benchmarking

    @wardvermeulen
    wardvermeulen committed May 14, 2023
    Configuration menu
    Copy the full SHA
    cc918d0 View commit details
    Browse the repository at this point in the history

Commits on May 16, 2023

  1. Merge branch 'master' into FPUOp

    @thomasfaingnaert
    thomasfaingnaert committed May 16, 2023
    Configuration menu
    Copy the full SHA
    18e7fe5 View commit details
    Browse the repository at this point in the history

  2. resolve some issues

    @wardvermeulen
    wardvermeulen committed May 16, 2023
    Configuration menu
    Copy the full SHA
    ae2176b View commit details
    Browse the repository at this point in the history

Commits on Jun 2, 2023

  1. fix 2D indexing 

    In addition, I added support for other operations than fma, and included tropical mma as an example.
The elements in shared memory are now accessed in a column-major order, for consistency purposes (there was no difference in performance).
    @wardvermeulen
    wardvermeulen committed Jun 2, 2023
    Configuration menu
    Copy the full SHA
    a9dc939 View commit details
    Browse the repository at this point in the history

Commits on Jun 5, 2023

  1. FPUOp relevant changes from dev

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    ff43c0a View commit details
    Browse the repository at this point in the history

  2. attempt at letting RowMajor work with FPUOp

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    862369e View commit details
    Browse the repository at this point in the history

  3. operater shape + mixed types

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    07fe75e View commit details
    Browse the repository at this point in the history

  4. add compute and data type test

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    c35dcf5 View commit details
    Browse the repository at this point in the history

  5. more performant because less registers needed

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    d851dab View commit details
    Browse the repository at this point in the history

  6. finished benchmarking, still work on cutlass

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    1f49263 View commit details
    Browse the repository at this point in the history

  7. add plot

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    5efa8eb View commit details
    Browse the repository at this point in the history

  8. mixed precision fix

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    af20919 View commit details
    Browse the repository at this point in the history

  9. add config file

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    c3618a4 View commit details
    Browse the repository at this point in the history

  10. preparation for pr

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    20bfa9e View commit details
    Browse the repository at this point in the history

  11. small fix to benchmarking

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    6bf9e8b View commit details
    Browse the repository at this point in the history

  12. resolve some issues

    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    3bb2507 View commit details
    Browse the repository at this point in the history

  13. fix 2D indexing 

    In addition, I added support for other operations than fma, and included tropical mma as an example.
The elements in shared memory are now accessed in a column-major order, for consistency purposes (there was no difference in performance).
    @wardvermeulen @maleadt
    wardvermeulen authored and maleadt committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    0d6da7c View commit details
    Browse the repository at this point in the history

  14. Bump CI timeout

    @thomasfaingnaert
    thomasfaingnaert committed Jun 5, 2023
    Configuration menu
    Copy the full SHA
    3746767 View commit details
    Browse the repository at this point in the history

Commits on Jun 6, 2023

  1. Merge branch 'FPUOp' of https://github.com/wardvermeulen/GemmKernels.jl… 

    … into FPUOp
    @wardvermeulen
    wardvermeulen committed Jun 6, 2023
    Configuration menu
    Copy the full SHA
    ed9e0b3 View commit details
    Browse the repository at this point in the history

  2. clean up mma, add integer and tropical tests

    @wardvermeulen
    wardvermeulen committed Jun 6, 2023
    Configuration menu
    Copy the full SHA
    6fa7753 View commit details
    Browse the repository at this point in the history

Commits on Jun 7, 2023

  1. add cutlass kernels closest to fpu configuration

    @wardvermeulen
    wardvermeulen committed Jun 7, 2023
    Configuration menu
    Copy the full SHA
    4085d20 View commit details
    Browse the repository at this point in the history

  2. add information on benchmarking with nvprof

    @wardvermeulen
    wardvermeulen committed Jun 7, 2023
    Configuration menu
    Copy the full SHA
    148c57a View commit details
    Browse the repository at this point in the history