[FFmpeg-cvslog] lavfi/super2xsai: fix fate test on bigendian
Stefano Sabatini
stefasab at gmail.com
Sun Jun 17 12:26:47 CEST 2012
On date Saturday 2012-06-16 16:34:26 +0200, Paul B Mahol wrote:
> ffmpeg | branch: master | Paul B Mahol <onemda at gmail.com> | Sat Jun 16 09:35:05 2012 +0000| [fd8def92488921d475a7ceff341eaf4cfb009f9b] | committer: Paul B Mahol
>
> lavfi/super2xsai: fix fate test on bigendian
>
> Signed-off-by: Paul B Mahol <onemda at gmail.com>
>
> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=fd8def92488921d475a7ceff341eaf4cfb009f9b
> ---
>
> libavfilter/vf_super2xsai.c | 29 +++++++++++++++++++++--------
> tests/ref/lavfi/pixfmts_super2xsai | 4 ++++
> 2 files changed, 25 insertions(+), 8 deletions(-)
>
> diff --git a/libavfilter/vf_super2xsai.c b/libavfilter/vf_super2xsai.c
> index e342078..1b0b6a0 100644
> --- a/libavfilter/vf_super2xsai.c
> +++ b/libavfilter/vf_super2xsai.c
> @@ -40,6 +40,7 @@ typedef struct {
> uint32_t q_lo_pixel_mask;
>
> int bpp; ///< bytes per pixel, pixel stride for each (packed) pixel
> + int is_be;
> } Super2xSaIContext;
>
> #define GET_RESULT(A, B, C, D) ((A != C || A != D) - (B != C || B != D))
> @@ -72,7 +73,7 @@ static void super2xsai(AVFilterContext *ctx,
>
> #define READ_COLOR4(dst, src_line, off) dst = *((const uint32_t *)src_line + off)
> #define READ_COLOR3(dst, src_line, off) dst = AV_RL24 (src_line + 3*off)
> -#define READ_COLOR2(dst, src_line, off) dst = *((const uint16_t *)src_line + off)
> +#define READ_COLOR2(dst, src_line, off) dst = sai->is_be ? AV_RB16(src_line + 2 * off) : AV_RL16(src_line + 2 * off)
>
> for (y = 0; y < height; y++) {
> uint8_t *dst_line[2];
> @@ -175,8 +176,13 @@ static void super2xsai(AVFilterContext *ctx,
> AV_WL24(dst_line[1] + x * 6 + 3, product2b);
> break;
> default: // bpp = 2
> - AV_WN32A(dst_line[0] + x * 4, product1a | (product1b << 16));
> - AV_WN32A(dst_line[1] + x * 4, product2a | (product2b << 16));
> + if (sai->is_be) {
> + AV_WB32(dst_line[0] + x * 4, product1a | (product1b << 16));
> + AV_WB32(dst_line[1] + x * 4, product2a | (product2b << 16));
> + } else {
> + AV_WL32(dst_line[0] + x * 4, product1a | (product1b << 16));
> + AV_WL32(dst_line[1] + x * 4, product2a | (product2b << 16));
> + }
This may have a performance hit, but maybe the difference is not very
significant.
The alternative was to only support native formats, and create
specific tests for the BE/LE cases.
[...]
Thanks.
--
FFmpeg = Furious and Fantastic Most Portentous Earthshaking Gymnast
More information about the ffmpeg-cvslog
mailing list