From 19fb234e4af1ff9f58ff2fdd604ac6f6bb87ad6b Mon Sep 17 00:00:00 2001 From: Jason Garrett-Glaser Date: Fri, 14 Jan 2011 21:34:25 +0000 Subject: H.264: split luma dc idct out and implement MMX/SSE2 versions About 2.5x the speed. NOTE: the way that the asm code handles large qmuls is a bit suboptimal. If x264-style dequant was used (separate shift and qmul values), it might be possible to get some extra speed. Originally committed as revision 26336 to svn://svn.ffmpeg.org/ffmpeg/trunk --- libavcodec/h264.h | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'libavcodec/h264.h') diff --git a/libavcodec/h264.h b/libavcodec/h264.h index 7158d9748e..189864bb38 100644 --- a/libavcodec/h264.h +++ b/libavcodec/h264.h @@ -406,6 +406,7 @@ typedef struct H264Context{ GetBitContext *inter_gb_ptr; DECLARE_ALIGNED(16, DCTELEM, mb)[16*24]; + DECLARE_ALIGNED(16, DCTELEM, mb_luma_dc)[16]; DCTELEM mb_padding[256]; ///< as mb is addressed by scantable[i] and scantable is uint8_t we can either check that i is not too large or ensure that there is some unused stuff after mb /** @@ -600,10 +601,6 @@ typedef struct H264Context{ extern const uint8_t ff_h264_chroma_qp[52]; -void ff_svq3_luma_dc_dequant_idct_c(DCTELEM *block, int qp); - -void ff_svq3_add_idct_c(uint8_t *dst, DCTELEM *block, int stride, int qp, int dc); - /** * Decode SEI */ -- cgit v1.2.3