[FFmpeg-devel] [PATCH] x86/dsputilenc: implement XOP version of pix_sum16

Michael Niedermayer michaelni at gmx.at
Thu May 29 18:54:03 CEST 2014


On Thu, May 29, 2014 at 12:57:39AM -0300, James Almer wrote:
> SSE2: 137 cycles
> XOP:   87 cycles
> Signed-off-by: James Almer <jamrial at gmail.com>
> ---
> The differences aren't as many as i originally thought after i realized 
> that paddw can be used inside the loop.
> The resulting macro is still a bit ugly, so commit whichever version you 
> like the most.
> 
>  libavcodec/x86/dsputilenc.asm   | 29 ++++++++++++++++++++++++-----
>  libavcodec/x86/dsputilenc_mmx.c |  5 +++++
>  2 files changed, 29 insertions(+), 5 deletions(-)

applied

thx

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

When you are offended at any man's fault, turn to yourself and study your
own failings. Then you will forget your anger. -- Epictetus
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 181 bytes
Desc: Digital signature
URL: <https://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20140529/ce35f9a0/attachment.asc>


More information about the ffmpeg-devel mailing list