It just does that part in scalar form, I doubt using a vector store over 2 array would speed it up particularly. The function should be written to not use a scratch buffer. |
14 years ago | |
|---|---|---|
| .. | ||
| swscale_altivec.c | swscale: add dithering to yuv2yuvX_altivec_real | 14 years ago |
| yuv2rgb_altivec.c | swscale: split yuv2packedX_altivec in smaller functions. | 14 years ago |
| yuv2rgb_altivec.h | swscale: split yuv2packedX_altivec in smaller functions. | 14 years ago |
| yuv2yuv_altivec.c | swscale: split out unscaled altivec YUV converters in their own file. | 14 years ago |