[FFmpeg-devel] [PATCH 11/17] x86: remove VLA in ac3_downmix_sse

Ronald S. Bultje rsbultje
Tue Jan 18 20:58:58 CET 2011


Hi,

On Tue, Jan 18, 2011 at 2:32 PM, Mans Rullgard <mans at mansr.com> wrote:
> ---
> ?libavcodec/x86/dsputil_mmx.c | ? ?3 ++-
> ?1 files changed, 2 insertions(+), 1 deletions(-)
>
> diff --git a/libavcodec/x86/dsputil_mmx.c b/libavcodec/x86/dsputil_mmx.c
> index 5d04672..d8edb0f 100644
> --- a/libavcodec/x86/dsputil_mmx.c
> +++ b/libavcodec/x86/dsputil_mmx.c
> @@ -28,6 +28,7 @@
> ?#include "libavcodec/h264dsp.h"
> ?#include "libavcodec/mpegvideo.h"
> ?#include "libavcodec/simple_idct.h"
> +#include "libavcodec/ac3dec.h"
> ?#include "dsputil_mmx.h"
> ?#include "idct_xvid.h"
>
> @@ -2049,7 +2050,7 @@ static void ac3_downmix_sse(float (*samples)[256], float (*matrix)[2], int out_c
> ? ? } else if(in_ch == 5 && out_ch == 1 && matrix_cmp[0][0]==matrix_cmp[2][0] && matrix_cmp[3][0]==matrix_cmp[4][0]) {
> ? ? ? ? MIX5(IF1,IF0);
> ? ? } else {
> - ? ? ? ?DECLARE_ALIGNED(16, float, matrix_simd)[in_ch][2][4];
> + ? ? ? ?DECLARE_ALIGNED(16, float, matrix_simd)[AC3_MAX_CHANNELS][2][4];
> ? ? ? ? j = 2*in_ch*sizeof(float);
> ? ? ? ? __asm__ volatile(
> ? ? ? ? ? ? "1: \n"
> --
> 1.7.3.5

OK.

Ronald



More information about the ffmpeg-devel mailing list