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

Add the first implememtation of fusion_group op #19621 #3

Merged
merged 23 commits into from
Nov 18, 2019
Merged

Add the first implememtation of fusion_group op #19621 #3

merged 23 commits into from
Nov 18, 2019

Commits on Aug 26, 2019

  1. Add the dynamic load of nvrtc, and support runtime compiling of CUDA … 

    …kernel using nvrtc.

test=develop
    @Xreki
    Xreki committed Aug 26, 2019
    Configuration menu
    Copy the full SHA
    6f26abb View commit details
    Browse the repository at this point in the history

  2. Call CUDA driver api to launch the kernel compiled by nvrtc. 

    test=develop
    @Xreki
    Xreki committed Aug 26, 2019
    Configuration menu
    Copy the full SHA
    8c04f5b View commit details
    Browse the repository at this point in the history

  3. Merge branch 'develop' into fuse_jit_compile 

    test=develop
    @Xreki
    Xreki committed Aug 26, 2019
    Configuration menu
    Copy the full SHA
    1dad4e6 View commit details
    Browse the repository at this point in the history

Commits on Aug 27, 2019

  1. Disable for mac and windows. 

    test=develop
    @Xreki
    Xreki committed Aug 27, 2019
    Configuration menu
    Copy the full SHA
    fc953b6 View commit details
    Browse the repository at this point in the history

  2. Refine the codes to support manually specified num_threads and worklo… 

    …ad_per_thread.

test=develop
    @Xreki
    Xreki committed Aug 27, 2019
    Configuration menu
    Copy the full SHA
    b5fc76c View commit details
    Browse the repository at this point in the history

Commits on Aug 29, 2019

  1. Merge branch 'develop' into fuse_jit_compile

    @Xreki
    Xreki committed Aug 29, 2019
    Configuration menu
    Copy the full SHA
    a32fc9a View commit details
    Browse the repository at this point in the history

  2. Refine the CUDA kernel to support large dims. 

    test=develop
    @Xreki
    Xreki committed Aug 29, 2019
    Configuration menu
    Copy the full SHA
    3c879d8 View commit details
    Browse the repository at this point in the history

Commits on Sep 3, 2019

  1. Merge branch 'develop' into fuse_fusion_group

    @Xreki
    Xreki committed Sep 3, 2019
    Configuration menu
    Copy the full SHA
    e0ac413 View commit details
    Browse the repository at this point in the history

  2. Add DeviceCodePool to manage all device codes.

    @Xreki
    Xreki committed Sep 3, 2019
    Configuration menu
    Copy the full SHA
    dc9947f View commit details
    Browse the repository at this point in the history

Commits on Sep 4, 2019

  1. Add the first implementation fusion_group op.

    @Xreki
    Xreki committed Sep 4, 2019
    Configuration menu
    Copy the full SHA
    9c51937 View commit details
    Browse the repository at this point in the history

  2. Add unit-test for fusion_group op.

    @Xreki
    Xreki committed Sep 4, 2019
    Configuration menu
    Copy the full SHA
    7becebd View commit details
    Browse the repository at this point in the history

  3. Add the check of result.

    @Xreki
    Xreki committed Sep 4, 2019
    Configuration menu
    Copy the full SHA
    9f0c215 View commit details
    Browse the repository at this point in the history

  4. Merge branch 'develop' into fuse_fusion_group 

    test=develop
    @Xreki
    Xreki committed Sep 4, 2019
    Configuration menu
    Copy the full SHA
    cfc91e1 View commit details
    Browse the repository at this point in the history

  5. Add the check of nvrtc in unit-test. 

    test=develop
    @Xreki
    Xreki committed Sep 4, 2019
    Configuration menu
    Copy the full SHA
    07aeb1a View commit details
    Browse the repository at this point in the history

  6. Add comment to explain the inputs, outputs and features of fusion_gro… 

    …up op.

test=develop
    @Xreki
    Xreki committed Sep 4, 2019
    Configuration menu
    Copy the full SHA
    583a254 View commit details
    Browse the repository at this point in the history

  7. Disable fusion_group op for mac and windows. 

    test=develop
    @Xreki
    Xreki committed Sep 4, 2019
    Configuration menu
    Copy the full SHA
    56bb1f1 View commit details
    Browse the repository at this point in the history

Commits on Sep 5, 2019

  1. Merge branch 'develop' into fuse_fusion_group 

    test=develop
    @Xreki
    Xreki committed Sep 5, 2019
    Configuration menu
    Copy the full SHA
    6b3a03e View commit details
    Browse the repository at this point in the history

Commits on Sep 6, 2019

  1. Make the compiling of device code return status instead of hanging up. 

    test=develop
    @Xreki
    Xreki committed Sep 6, 2019
    Configuration menu
    Copy the full SHA
    a1fd12e View commit details
    Browse the repository at this point in the history

Commits on Oct 29, 2019

  1. Merge branch 'develop' into fuse_fusion_group

    @Xreki
    Xreki committed Oct 29, 2019
    Configuration menu
    Copy the full SHA
    121a919 View commit details
    Browse the repository at this point in the history

  2. Add the check of whether there is CUDA driver library, and do not cor… 

    …e dump when failing to call the CUDA driver API.
    @Xreki
    Xreki committed Oct 29, 2019
    Configuration menu
    Copy the full SHA
    b907704 View commit details
    Browse the repository at this point in the history

  3. Unify fusion_group_op's input and output names. 

    test=develop
    @Xreki
    Xreki committed Oct 29, 2019
    Configuration menu
    Copy the full SHA
    fb8ddc9 View commit details
    Browse the repository at this point in the history

Commits on Oct 30, 2019

  1. Merge branch 'develop' into fuse_fusion_group

    @Xreki
    Xreki committed Oct 30, 2019
    Configuration menu
    Copy the full SHA
    b1fb85a View commit details
    Browse the repository at this point in the history

  2. Add the check of CUDA driver library in unittest. 

    test=develop
    @Xreki
    Xreki committed Oct 30, 2019
    Configuration menu
    Copy the full SHA
    3e73678 View commit details
    Browse the repository at this point in the history