[Webkit-unassigned] [Bug 73545] [GStreamer] FFTFrame implementation

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Wed Jan 25 12:22:29 PST 2012


https://bugs.webkit.org/show_bug.cgi?id=73545





--- Comment #24 from Chris Rogers <crogers at google.com>  2012-01-25 12:22:29 PST ---
(From update of attachment 123962)
View in context: https://bugs.webkit.org/attachment.cgi?id=123962&action=review

> Source/WebCore/platform/audio/gstreamer/FFTFrameGStreamer.cpp:47
> +    m_complexData = WTF::fastNewArray<GstFFTF32Complex>(unpackedFFTDataSize(m_FFTSize));

It seems odd to use fastNewArray for the allocation.  I don't know if it's "wrong" or not, but FFTFrameFFMPEG and FFTFrameMKL simply use an AudioFloatArray

> Source/WebCore/platform/audio/gstreamer/FFTFrameGStreamer.cpp:123
> +    for (size_t i = 0; i < unpackedFFTDataSize(m_FFTSize); ++i) {

Thanks to Intel, you should be able to optimize this loop using the new VectorMath::zvmul()  and VectorMath::vsmul() functions.  You can look at FFTFrameMac and FFTFrameFFMPEG as an example

> Source/WebCore/platform/audio/gstreamer/FFTFrameGStreamer.cpp:145
> +    for (size_t i = 0; i < unpackedFFTDataSize(m_FFTSize); ++i) {

I think you could use VectorMath::vsmul() here

> Source/WebCore/platform/audio/gstreamer/FFTFrameGStreamer.cpp:166
> +    for (unsigned i = 0; i < m_FFTSize; ++i)

could use VectorMath::vsmul() here

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.



More information about the webkit-unassigned mailing list