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

intelblas_gemm clean patch #19

Merged
merged 9 commits into from
Feb 16, 2017
Merged

intelblas_gemm clean patch #19

merged 9 commits into from
Feb 16, 2017

Commits on Feb 10, 2017

  1. integrate intelblas_gemm. 

    Change-Id: Ic9edf18a3ae0f41b21c2ac374d50000fc5d4e6f3
    @wujunkai166
    wujunkai166 authored and Lin, Lixiang committed Feb 10, 2017
    Configuration menu
    Copy the full SHA
    ce1c65d View commit details
    Browse the repository at this point in the history

  2. Added log printing of external kernels 

    Change-Id: I89f632e2598594805e24b6aa2d084dcfa1c4f218
    Lin, Lixiang committed Feb 10, 2017
    Configuration menu
    Copy the full SHA
    bf28d43 View commit details
    Browse the repository at this point in the history

  3. Modified external template bench print for more than one template for… 

    … the same operatin.
    Lin, Lixiang committed Feb 10, 2017
    Configuration menu
    Copy the full SHA
    f04f866 View commit details
    Browse the repository at this point in the history

  4. Add image support for gemm kernel. 

    v2: by zhigang, fix some warnings and remove half relative code.
v3: by lixiang, modify json file
v4: by junkai, optimize gemm image kernel and force isaac to run gemm image kernel.
v5: by junkai, change json file to force issac to run gemm image kernel.
Change-Id: Ieab41924476bfc001f7026fbea3b5ea5e56eb00b
    @wujunkai166
    wujunkai166 authored and Lin, Lixiang committed Feb 10, 2017
    Configuration menu
    Copy the full SHA
    8702578 View commit details
    Browse the repository at this point in the history

  5. Fix some minor issues and update json files for Intel platforms.

    Lin, Lixiang committed Feb 10, 2017
    Configuration menu
    Copy the full SHA
    4b5cf48 View commit details
    Browse the repository at this point in the history

  6. Introduce some new dispatch APIs for OCL. 

    We need to use image related API for Intel's image based kernels.
Also fixed a memory leak issue in the image based kernel.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
    Zhigang Gong authored and Lin, Lixiang committed Feb 10, 2017
    Configuration menu
    Copy the full SHA
    b882fe9 View commit details
    Browse the repository at this point in the history

  7. Add image block version of gemm kernel. No need to fall back to 

    buffer kernel when input sizes are too large.

v2:
fix minor issues when release images. And also avoid use OpenCL
library directly.

Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
    @wujunkai166
    wujunkai166 authored and Lin, Lixiang committed Feb 10, 2017
    Configuration menu
    Copy the full SHA
    1bc3be6 View commit details
    Browse the repository at this point in the history

Commits on Feb 13, 2017

  1. Added intelblas python expternal scripts and updated json files

    Lin, Lixiang committed Feb 13, 2017
    Configuration menu
    Copy the full SHA
    bc9121d View commit details
    Browse the repository at this point in the history

Commits on Feb 16, 2017

  1. Ocl building warnings removed

    Lin, Lixiang committed Feb 16, 2017
    Configuration menu
    Copy the full SHA
    461b83d View commit details
    Browse the repository at this point in the history