[FFmpeg-devel] [PATCH] Altivec version of-altivec h264_h-v_loop_filter_luma

[Michael Niedermayer]

Hi

On Fri, May 11, 2007 at 09:45:37PM +0200, Guillaume Poirier wrote:
> Hi,
> Attached patch implements $SUBJ, based on a patch by Graham Booker % 
> perian A cod3r P com%
> 
> Comments welcome about trivial things, I would like to have it merged  
> to SVN ASAP since it's been rotting on my repository for far too long  
> without much work done to it.
> 
> It's not perfect, but it does what it's supposed so, i.e. accelerate  
> h264 decoding on G4/5
> 
> Cheers
> 
> Guillaume
> 

> From 992d7e3a4ebd555acef7c18d0f83184fcf8ae2c4 Mon Sep 17 00:00:00 2001
> From: Guillaume Poirier <guillaumepoirier at G5.local>
> Date: Thu, 10 May 2007 22:35:15 +0200
> Subject: [PATCH] Altivec version of altivec h264_h/v_loop_filter_luma_ routines
> Based on a patch by Graham Booker %perian A cod3r P com%
> ---
>  libavcodec/ppc/h264_altivec.c |  514 ++++++++++++++++++++++++++++++++---------
>  1 files changed, 405 insertions(+), 109 deletions(-)
> 
> diff --git a/libavcodec/ppc/h264_altivec.c b/libavcodec/ppc/h264_altivec.c
> index bac620e..3b2c52a 100644
> --- a/libavcodec/ppc/h264_altivec.c
> +++ b/libavcodec/ppc/h264_altivec.c
> @@ -181,10 +181,10 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc32_ ## CODETYPE(uint8_t *dst, uint
>  /* this code assume that stride % 16 == 0 */
>  void put_no_rnd_h264_chroma_mc8_altivec(uint8_t * dst, uint8_t * src, int stride, int h, int x, int y) {
>      signed int ABCD[4] __attribute__((aligned(16))) =
> -                        {((8 - x) * (8 - y)),
> -                          ((x) * (8 - y)),
> -                          ((8 - x) * (y)),
> -                          ((x) * (y))};
> +			{((8 - x) * (8 - y)),
> +			  ((x) * (8 - y)),
> +			  ((8 - x) * (y)),
> +			  ((x) * (y))};
>      register int i;

adding tabs?

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

Concerning the gods, I have no means of knowing whether they exist or not
or of what sort they may be, because of the obscurity of the subject, and
the brevity of human life -- Protagoras
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20070511/9a817fd3/attachment.pgp>