search

awdseq1357 / libyuv

Automatically exported from code.google.com/p/libyuv
0 stars 0 forks source link

NV12ToARGB is slow on NEON #115

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
NV12ToARGB is fast on x86 by making it weave in assembly.  But its C on NEON.  
Do it like X86 with an NV12 fetch followed by YUV TO RGB

export LIBYUV_REPEAT=1000
./libyuv_unittest --gtest_filter=*NV12*

NV12ToRGB565 is also slow and will benefit from NV12ToARGB

NV12ToARGB_OptVsC (28699 ms)
NV12ToRGB565_OptVsC (32048 ms)

Original issue reported on code.google.com by fbarch...@google.com on 8 Oct 2012 at 9:19