Ningxin Hu

Add helloCompute path

3

Inspired by https://developers.google.com/web/updates/2019/08/get-started-with-gpu-compute-on-the-web

Models: https://github.com/google/mediapipe/tree/master/mediapipe/models#hand-detection-and-tracking

Latest version of mandelbrot-asm.html crashes on both Firefox Nightly and Chromium

6

I've reported a bug to Firefox: https://bugzilla.mozilla.org/show_bug.cgi?id=1068331

pthreadpool cfi-icall check failure for XNNPACK function pointers casting

3

This issue was raised when integrating XNNPACK and pthreadpool into Chromium ([CL-4608520](https://chromium-review.googlesource.com/c/chromium/src/+/4608520/comments/80ec8200_7d0bd6d6)) where Chromium has cfi ([control-flow-integrity](https://www.chromium.org/developers/testing/control-flow-integrity/)) enabled build that also checks any cfi-icall (indirect call) issues. For example, the...

According to [UAV barriers and resource state barriers in DirectML example 2](https://docs.microsoft.com/en-us/windows/win32/direct3d12/dml-barriers#example-2), ops without data dependency can execute concurrently on the GPU, which may improve performance. In current DML backend,...

https://github.com/otcshare/chromium-src/blob/webml/services/ml/compilation_delegate_dml.cc#L646 It should be "CompilationDelegateDML" in the debug log.

Support loading extension

29

It includes loading CPU extensions by AddExtension and GPU extensions by SetConfig. The native sample code could be found at https://github.com/opencv/dldt/blob/2020/inference-engine/samples/classification_sample_async/main.cpp#L83 for CPU extension and https://github.com/opencv/dldt/blob/2020/inference-engine/samples/classification_sample_async/main.cpp#L89 for GPU extension. Please...