moved scaling from dequantization to inverse transform for T8x8

Previously, the scaling related to extended quantize range happens in
dequantization stage, which implies the coefficients form forward
transform are in different scale(4x) from dequantization coefficients
This worked fine when there was not distortion computation done based
on 8x8 transform, but it completely wracked the distortion estimation
based on transform coefficients and dequantized transform coefficients
introduced in commit f64725a00 for macroblocks using 8x8 transform.
This commit fixed the issue by moving the scaling into the stage of
inverse 8x8 transform.

TODO: Test&Verify the transform/quantization pipeline accuracy.

Change-Id: Iff77b36a965c2a6b247e59b9c59df93eba5d60e2

6 files changed, 11 insertions, 16 deletions

diff --git a/vp8/common/idctllm.c b/vp8/common/idctllm.c
index c11386898..22c9baa46 100644
--- a/vp8/common/idctllm.c
+++ b/vp8/common/idctllm.c
@@ -211,7 +211,8 @@ void vp8_short_idct8x8_1_c(short *input, short *output, int pitch)
     short *op = output;
     short *orig_op = output;
     int shortpitch = pitch >> 1;
-    a1 = ((input[0] + 4) >> 3);
+    //a1 = ((input[0] + 4) >> 3);
+    a1 = ((input[0] + 16) >> 5);
     for (b = 0; b < 4; b++)
     {
         for (i = 0; i < 4; i++)
@@ -228,7 +229,8 @@ void vp8_short_idct8x8_1_c(short *input, short *output, int pitch)
 
 void vp8_dc_only_idct_add_8x8_c(short input_dc, unsigned char *pred_ptr, unsigned char *dst_ptr, int pitch, int stride)
 {
-    int a1 = ((input_dc + 4) >> 3);
+    //int a1 = ((input_dc + 4) >> 3);
+    int a1 = ((input_dc + 16) >> 5);
     int r, c, b;
     unsigned char *orig_pred = pred_ptr;
     unsigned char *orig_dst = dst_ptr;
@@ -394,7 +396,7 @@ void vp8_short_idct8x8_c(short *coefs, short *block, int pitch)
     {
         for (j = 0; j < TX_DIM; j++)
         {
-             X[i * TX_DIM + j] = (int)coefs[i * TX_DIM + j];
+             X[i * TX_DIM + j] = (int)(coefs[i * TX_DIM + j]+2)>>2;
         }
     }
   for (i = 0; i < 8; i++)
diff --git a/vp8/decoder/dequantize.c b/vp8/decoder/dequantize.c
index 83bf8a76c..c48f5c23f 100644
--- a/vp8/decoder/dequantize.c
+++ b/vp8/decoder/dequantize.c
@@ -131,7 +131,7 @@ void vp8_dequantize_b_2x2_c(BLOCKD *d)
 
     for (i = 0; i < 16; i++)
     {
-        DQ[i] = (short)((Q[i] * DQC[i]+2)>>2);
+        DQ[i] = (short)((Q[i] * DQC[i]));
     }
 #ifdef DEC_DEBUG
     if (dec_debug) {
@@ -164,12 +164,12 @@ void vp8_dequant_idct_add_8x8_c(short *input, short *dq, unsigned char *pred,
     }
 #endif
 
-    input[0]= (input[0] * dq[0]+2)>>2;
+    input[0]= input[0] * dq[0];
 
     // recover quantizer for 4 4x4 blocks
     for (i = 1; i < 64; i++)
     {
-      input[i]=(input[i] * dq[1]+2)>>2;
+      input[i]=input[i] * dq[1];
     }
 #ifdef DEC_DEBUG
     if (dec_debug) {
@@ -262,7 +262,7 @@ void vp8_dequant_dc_idct_add_8x8_c(short *input, short *dq, unsigned char *pred,
 #endif
     for (i = 1; i < 64; i++)
     {
-        input[i]=(input[i] * dq[1]+2)>>2;
+        input[i]=input[i] * dq[1];
     }
 
 #ifdef DEC_DEBUG
diff --git a/vp8/encoder/encodemb.c b/vp8/encoder/encodemb.c
index 270bdf585..b8f92c970 100644
--- a/vp8/encoder/encodemb.c
+++ b/vp8/encoder/encodemb.c
@@ -1035,7 +1035,7 @@ void optimize_b_8x8(MACROBLOCK *mb, int i, int type,
             final_eob = i;
         rc = vp8_default_zig_zag1d_8x8[i];
         qcoeff_ptr[rc] = x;
-        dqcoeff_ptr[rc] = (x * dequant_ptr[rc!=0]+2)>>2;
+        dqcoeff_ptr[rc] = (x * dequant_ptr[rc!=0]);
 
         next = tokens[i][best].next;
         best = (best_mask[best] >> i) & 1;
diff --git a/vp8/encoder/onyx_if.c b/vp8/encoder/onyx_if.c
index 8d170b18a..c2f7e92e7 100644
--- a/vp8/encoder/onyx_if.c
+++ b/vp8/encoder/onyx_if.c
@@ -3411,7 +3411,6 @@ static void Pass1Encode(VP8_COMP *cpi, unsigned long *size, unsigned char *dest,
     scale_and_extend_source(cpi->un_scaled_source, cpi);
     vp8_first_pass(cpi);
 }
-
 //#define WRITE_RECON_BUFFER 1
 #if WRITE_RECON_BUFFER
 void write_cx_frame_to_file(YV12_BUFFER_CONFIG *frame, int this_frame)
diff --git a/vp8/encoder/quantize.c b/vp8/encoder/quantize.c
index 300c407e7..3b22e57b2 100644
--- a/vp8/encoder/quantize.c
+++ b/vp8/encoder/quantize.c
@@ -566,7 +566,6 @@ void vp8_regular_quantize_b_2x2(BLOCK *b, BLOCKD *d)
       qcoeff_ptr[rc]  = x;                         // write to destination
       //dqcoeff_ptr[rc] = x * dequant_ptr[rc]/q2nd;        // dequantized value
       dqcoeff_ptr[rc] = x * dequant_ptr[rc];        // dequantized value
-      dqcoeff_ptr[rc] = (dqcoeff_ptr[rc]+2)>>2;
 
 
       if (y)
@@ -627,7 +626,6 @@ void vp8_regular_quantize_b_8x8(BLOCK *b, BLOCKD *d)
       qcoeff_ptr[rc]  = x;                         // write to destination
       //dqcoeff_ptr[rc] = x * dequant_ptr[rc!=0] / q1st;        // dequantized value
       dqcoeff_ptr[rc] = x * dequant_ptr[rc!=0];        // dequantized value
-      dqcoeff_ptr[rc] = (dqcoeff_ptr[rc]+2)>>2;
 
       if (y)
       {
diff --git a/vp8/encoder/rdopt.c b/vp8/encoder/rdopt.c
index 1d593fbee..c7bd8fc99 100644
--- a/vp8/encoder/rdopt.c
+++ b/vp8/encoder/rdopt.c
@@ -777,11 +777,7 @@ static void macro_block_yrd_8x8( MACROBLOCK *mb,
     mb->e_mbd.dqcoeff[192] = 0;
     d = ENCODEMB_INVOKE(&rtcd->encodemb, mberr)(mb, 0) << 2;
 
-#if CONFIG_EXTEND_QRANGE
-    d += ENCODEMB_INVOKE(rtcd, berr)(mb_y2->coeff, x_y2->dqcoeff)<<2;
-#else
-    d += ENCODEMB_INVOKE(&rtcd->encodemb, berr)(mb_y2->coeff, x_y2->dqcoeff);
-#endif
+    d += ENCODEMB_INVOKE(&rtcd->encodemb, berr)(mb_y2->coeff, x_y2->dqcoeff)<<2;
 
     *Distortion = (d >> 4);
     // rate