Adds 8-, 9- and 10-bit versions of some of the functions used by the h264 decoder.

This patch lets e.g. dsputil_init chose dsp functions with respect to
the bit depth to decode. The naming scheme of bit depth dependent
functions is <base name>_<bit depth>[_<prefix>] (i.e. the old
clear_blocks_c is now named clear_blocks_8_c).

Note: Some of the functions for high bit depth is not dependent on the
bit depth, but only on the pixel size. This leaves some room for
optimizing binary size.

Preparatory patch for high bit depth h264 decoding support.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>

2 files changed, 10 insertions, 0 deletions

diff --git a/libavcodec/sh4/dsputil_align.c b/libavcodec/sh4/dsputil_align.c
index 7a8d60d73e..db40ece670 100644
--- a/libavcodec/sh4/dsputil_align.c
+++ b/libavcodec/sh4/dsputil_align.c
@@ -333,6 +333,9 @@ DEFFUNC(avg,no_rnd,xy,16,OP_XY,PACK)
 
 void dsputil_init_align(DSPContext* c, AVCodecContext *avctx)
 {
+        const int high_bit_depth = avctx->codec_id == CODEC_ID_H264 && avctx->bits_per_raw_sample > 8;
+
+        if (!high_bit_depth) {
         c->put_pixels_tab[0][0] = put_rnd_pixels16_o;
         c->put_pixels_tab[0][1] = put_rnd_pixels16_x;
         c->put_pixels_tab[0][2] = put_rnd_pixels16_y;
@@ -368,6 +371,7 @@ void dsputil_init_align(DSPContext* c, AVCodecContext *avctx)
         c->avg_no_rnd_pixels_tab[1][1] = avg_no_rnd_pixels8_x;
         c->avg_no_rnd_pixels_tab[1][2] = avg_no_rnd_pixels8_y;
         c->avg_no_rnd_pixels_tab[1][3] = avg_no_rnd_pixels8_xy;
+        }
 
 #ifdef QPEL
 
@@ -401,20 +405,24 @@ void dsputil_init_align(DSPContext* c, AVCodecContext *avctx)
     dspfunc(avg_qpel, 1, 8);
     /* dspfunc(avg_no_rnd_qpel, 1, 8); */
 
+    if (!high_bit_depth) {
     dspfunc(put_h264_qpel, 0, 16);
     dspfunc(put_h264_qpel, 1, 8);
     dspfunc(put_h264_qpel, 2, 4);
     dspfunc(avg_h264_qpel, 0, 16);
     dspfunc(avg_h264_qpel, 1, 8);
     dspfunc(avg_h264_qpel, 2, 4);
+    }
 
 #undef dspfunc
+    if (!high_bit_depth) {
     c->put_h264_chroma_pixels_tab[0]= put_h264_chroma_mc8_sh4;
     c->put_h264_chroma_pixels_tab[1]= put_h264_chroma_mc4_sh4;
     c->put_h264_chroma_pixels_tab[2]= put_h264_chroma_mc2_sh4;
     c->avg_h264_chroma_pixels_tab[0]= avg_h264_chroma_mc8_sh4;
     c->avg_h264_chroma_pixels_tab[1]= avg_h264_chroma_mc4_sh4;
     c->avg_h264_chroma_pixels_tab[2]= avg_h264_chroma_mc2_sh4;
+    }
 
     c->put_mspel_pixels_tab[0]= put_mspel8_mc00_sh4;
     c->put_mspel_pixels_tab[1]= put_mspel8_mc10_sh4;
diff --git a/libavcodec/sh4/dsputil_sh4.c b/libavcodec/sh4/dsputil_sh4.c
index 0c724c346d..9ea48ad4a1 100644
--- a/libavcodec/sh4/dsputil_sh4.c
+++ b/libavcodec/sh4/dsputil_sh4.c
@@ -92,8 +92,10 @@ static void idct_add(uint8_t *dest, int line_size, DCTELEM *block)
 void dsputil_init_sh4(DSPContext* c, AVCodecContext *avctx)
 {
         const int idct_algo= avctx->idct_algo;
+        const int high_bit_depth = avctx->codec_id == CODEC_ID_H264 && avctx->bits_per_raw_sample > 8;
         dsputil_init_align(c,avctx);
 
+        if (!high_bit_depth)
         c->clear_blocks = clear_blocks_sh4;
         if(idct_algo==FF_IDCT_AUTO || idct_algo==FF_IDCT_SH4){
                 c->idct_put = idct_put;