From 76741b0e56bfbc74cfa32ff59e15cf420463569b Mon Sep 17 00:00:00 2001
From: Baptiste Coudurier <baptiste.coudurier@gmail.com>
Date: Tue, 16 Aug 2011 17:05:44 +0200
Subject: h264: 4:2:2 intra decoding support

Signed-off-by: Diego Biurrun <diego@biurrun.de>
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
---
 libavcodec/h264idct_template.c | 50 ++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 50 insertions(+)

(limited to 'libavcodec/h264idct_template.c')

diff --git a/libavcodec/h264idct_template.c b/libavcodec/h264idct_template.c
index ba5571576d..eba850ac6f 100644
--- a/libavcodec/h264idct_template.c
+++ b/libavcodec/h264idct_template.c
@@ -224,6 +224,29 @@ void FUNCC(ff_h264_idct_add8)(uint8_t **dest, const int *block_offset, DCTELEM *
         }
     }
 }
+
+void FUNCC(ff_h264_idct_add8_422)(uint8_t **dest, const int *block_offset, DCTELEM *block, int stride, const uint8_t nnzc[15*8]){
+    int i, j;
+
+    for(j=1; j<3; j++){
+        for(i=j*16; i<j*16+4; i++){
+            if(nnzc[ scan8[i] ])
+                FUNCC(ff_h264_idct_add   )(dest[j-1] + block_offset[i], block + i*16*sizeof(pixel), stride);
+            else if(((dctcoef*)block)[i*16])
+                FUNCC(ff_h264_idct_dc_add)(dest[j-1] + block_offset[i], block + i*16*sizeof(pixel), stride);
+        }
+    }
+
+    for(j=1; j<3; j++){
+        for(i=j*16+4; i<j*16+8; i++){
+            if(nnzc[ scan8[i+4] ])
+                FUNCC(ff_h264_idct_add   )(dest[j-1] + block_offset[i+4], block + i*16*sizeof(pixel), stride);
+            else if(((dctcoef*)block)[i*16])
+                FUNCC(ff_h264_idct_dc_add)(dest[j-1] + block_offset[i+4], block + i*16*sizeof(pixel), stride);
+        }
+    }
+}
+
 /**
  * IDCT transforms the 16 dc values and dequantizes them.
  * @param qmul quantization parameter
@@ -263,6 +286,33 @@ void FUNCC(ff_h264_luma_dc_dequant_idct)(DCTELEM *_output, DCTELEM *_input, int
 #undef stride
 }
 
+void FUNCC(ff_h264_chroma422_dc_dequant_idct)(DCTELEM *_block, int qmul){
+    const int stride= 16*2;
+    const int xStride= 16;
+    int i;
+    int temp[8];
+    static const uint8_t x_offset[2]={0, 16};
+    dctcoef *block = (dctcoef*)_block;
+
+    for(i=0; i<4; i++){
+        temp[2*i+0] = block[stride*i + xStride*0] + block[stride*i + xStride*1];
+        temp[2*i+1] = block[stride*i + xStride*0] - block[stride*i + xStride*1];
+    }
+
+    for(i=0; i<2; i++){
+        const int offset= x_offset[i];
+        const int z0= temp[2*0+i] + temp[2*2+i];
+        const int z1= temp[2*0+i] - temp[2*2+i];
+        const int z2= temp[2*1+i] - temp[2*3+i];
+        const int z3= temp[2*1+i] + temp[2*3+i];
+
+        block[stride*0+offset]= ((z0 + z3)*qmul + 128) >> 8;
+        block[stride*1+offset]= ((z1 + z2)*qmul + 128) >> 8;
+        block[stride*2+offset]= ((z1 - z2)*qmul + 128) >> 8;
+        block[stride*3+offset]= ((z0 - z3)*qmul + 128) >> 8;
+    }
+}
+
 void FUNCC(ff_h264_chroma_dc_dequant_idct)(DCTELEM *_block, int qmul){
     const int stride= 16*2;
     const int xStride= 16;
-- 
cgit v1.2.3