On Wed, Feb 17, 2016 at 04:18:26PM +0100, Clément Bœsch wrote: > From: Clément Bœsch <clem...@stupeflix.com> > > --- > This code is mirroring the ARM version. > > It's not tested on real hardware (qemu only), so I have no idea about > performance. This also means I didn't put any prefetch mechanism (which had a > noticeable impact on performance on ARM). > > Last but not least, the 32-bit precision functions are added (and tested) but > are not enabled at all, just like the ARM counterpart. > --- > libswscale/aarch64/Makefile | 3 + > libswscale/aarch64/swscale_unscaled.c | 142 +++++++++++++++ > libswscale/aarch64/yuv2rgb_neon.S | 332 > ++++++++++++++++++++++++++++++++++ > libswscale/swscale_internal.h | 1 + > libswscale/swscale_unscaled.c | 2 + > 5 files changed, 480 insertions(+) > create mode 100644 libswscale/aarch64/Makefile > create mode 100644 libswscale/aarch64/swscale_unscaled.c > create mode 100644 libswscale/aarch64/yuv2rgb_neon.S
LGTM but i cant really comment on aarch64 as i have no hw and you know the asm much better than i do thanks [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB I do not agree with what you have to say, but I'll defend to the death your right to say it. -- Voltaire
signature.asc
Description: Digital signature
_______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
