summaryrefslogtreecommitdiff
path: root/src/libffmpeg/libavcodec/dsputil.c
diff options
context:
space:
mode:
Diffstat (limited to 'src/libffmpeg/libavcodec/dsputil.c')
-rw-r--r--src/libffmpeg/libavcodec/dsputil.c284
1 files changed, 145 insertions, 139 deletions
diff --git a/src/libffmpeg/libavcodec/dsputil.c b/src/libffmpeg/libavcodec/dsputil.c
index 9b79b8659..51eddbc60 100644
--- a/src/libffmpeg/libavcodec/dsputil.c
+++ b/src/libffmpeg/libavcodec/dsputil.c
@@ -3,18 +3,20 @@
* Copyright (c) 2000, 2001 Fabrice Bellard.
* Copyright (c) 2002-2004 Michael Niedermayer <michaelni@gmx.at>
*
- * This library is free software; you can redistribute it and/or
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
- * version 2 of the License, or (at your option) any later version.
+ * version 2.1 of the License, or (at your option) any later version.
*
- * This library is distributed in the hope that it will be useful,
+ * FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this library; if not, write to the Free Software
+ * License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*
* gmc & q-pel & 32/64 bit based MC by Michael Niedermayer <michaelni@gmx.at>
@@ -35,8 +37,11 @@
/* snow.c */
void ff_spatial_dwt(int *buffer, int width, int height, int stride, int type, int decomposition_count);
-uint8_t cropTbl[256 + 2 * MAX_NEG_CROP] = {0, };
-uint32_t squareTbl[512] = {0, };
+/* vorbis.c */
+void vorbis_inverse_coupling(float *mag, float *ang, int blocksize);
+
+uint8_t ff_cropTbl[256 + 2 * MAX_NEG_CROP] = {0, };
+uint32_t ff_squareTbl[512] = {0, };
const uint8_t ff_zigzag_direct[64] = {
0, 1, 8, 16, 9, 2, 3, 10,
@@ -88,7 +93,7 @@ const uint8_t ff_alternate_vertical_scan[64] = {
};
/* a*inverse[b]>>32 == a/b for all 0<=a<=65536 && 2<=b<=255 */
-const uint32_t inverse[256]={
+const uint32_t ff_inverse[256]={
0, 4294967295U,2147483648U,1431655766, 1073741824, 858993460, 715827883, 613566757,
536870912, 477218589, 429496730, 390451573, 357913942, 330382100, 306783379, 286331154,
268435456, 252645136, 238609295, 226050911, 214748365, 204522253, 195225787, 186737709,
@@ -160,7 +165,7 @@ static int pix_sum_c(uint8_t * pix, int line_size)
static int pix_norm1_c(uint8_t * pix, int line_size)
{
int s, i, j;
- uint32_t *sq = squareTbl + 256;
+ uint32_t *sq = ff_squareTbl + 256;
s = 0;
for (i = 0; i < 16; i++) {
@@ -226,7 +231,7 @@ static void bswap_buf(uint32_t *dst, uint32_t *src, int w){
static int sse4_c(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, int h)
{
int s, i;
- uint32_t *sq = squareTbl + 256;
+ uint32_t *sq = ff_squareTbl + 256;
s = 0;
for (i = 0; i < h; i++) {
@@ -243,7 +248,7 @@ static int sse4_c(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, int h)
static int sse8_c(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, int h)
{
int s, i;
- uint32_t *sq = squareTbl + 256;
+ uint32_t *sq = ff_squareTbl + 256;
s = 0;
for (i = 0; i < h; i++) {
@@ -264,7 +269,7 @@ static int sse8_c(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, int h)
static int sse16_c(void *v, uint8_t *pix1, uint8_t *pix2, int line_size, int h)
{
int s, i;
- uint32_t *sq = squareTbl + 256;
+ uint32_t *sq = ff_squareTbl + 256;
s = 0;
for (i = 0; i < h; i++) {
@@ -353,7 +358,7 @@ static inline int w_c(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, in
for(i=0; i<size; i++){
for(j=0; j<size; j++){
int v= tmp[sx + sy + i*stride + j] * scale[type][dec_count-3][level][ori];
- s += ABS(v);
+ s += FFABS(v);
}
}
}
@@ -431,7 +436,7 @@ static void put_pixels_clamped_c(const DCTELEM *block, uint8_t *restrict pixels,
int line_size)
{
int i;
- uint8_t *cm = cropTbl + MAX_NEG_CROP;
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
/* read the pixels */
for(i=0;i<8;i++) {
@@ -453,7 +458,7 @@ static void put_pixels_clamped4_c(const DCTELEM *block, uint8_t *restrict pixels
int line_size)
{
int i;
- uint8_t *cm = cropTbl + MAX_NEG_CROP;
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
/* read the pixels */
for(i=0;i<4;i++) {
@@ -471,7 +476,7 @@ static void put_pixels_clamped2_c(const DCTELEM *block, uint8_t *restrict pixels
int line_size)
{
int i;
- uint8_t *cm = cropTbl + MAX_NEG_CROP;
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
/* read the pixels */
for(i=0;i<2;i++) {
@@ -508,7 +513,7 @@ static void add_pixels_clamped_c(const DCTELEM *block, uint8_t *restrict pixels,
int line_size)
{
int i;
- uint8_t *cm = cropTbl + MAX_NEG_CROP;
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
/* read the pixels */
for(i=0;i<8;i++) {
@@ -529,7 +534,7 @@ static void add_pixels_clamped4_c(const DCTELEM *block, uint8_t *restrict pixels
int line_size)
{
int i;
- uint8_t *cm = cropTbl + MAX_NEG_CROP;
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
/* read the pixels */
for(i=0;i<4;i++) {
@@ -546,7 +551,7 @@ static void add_pixels_clamped2_c(const DCTELEM *block, uint8_t *restrict pixels
int line_size)
{
int i;
- uint8_t *cm = cropTbl + MAX_NEG_CROP;
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
/* read the pixels */
for(i=0;i<2;i++) {
@@ -1484,86 +1489,33 @@ H264_CHROMA_MC(avg_ , op_avg)
#undef op_avg
#undef op_put
-static inline void copy_block2(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
-{
- int i;
- for(i=0; i<h; i++)
- {
- ST16(dst , LD16(src ));
- dst+=dstStride;
- src+=srcStride;
- }
-}
-
-static inline void copy_block4(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
-{
- int i;
- for(i=0; i<h; i++)
- {
- ST32(dst , LD32(src ));
- dst+=dstStride;
- src+=srcStride;
- }
-}
-
-static inline void copy_block8(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
-{
- int i;
- for(i=0; i<h; i++)
- {
- ST32(dst , LD32(src ));
- ST32(dst+4 , LD32(src+4 ));
- dst+=dstStride;
- src+=srcStride;
- }
-}
-
-static inline void copy_block16(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
-{
+static void put_no_rnd_h264_chroma_mc8_c(uint8_t *dst/*align 8*/, uint8_t *src/*align 1*/, int stride, int h, int x, int y){
+ const int A=(8-x)*(8-y);
+ const int B=( x)*(8-y);
+ const int C=(8-x)*( y);
+ const int D=( x)*( y);
int i;
- for(i=0; i<h; i++)
- {
- ST32(dst , LD32(src ));
- ST32(dst+4 , LD32(src+4 ));
- ST32(dst+8 , LD32(src+8 ));
- ST32(dst+12, LD32(src+12));
- dst+=dstStride;
- src+=srcStride;
- }
-}
-static inline void copy_block17(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
-{
- int i;
- for(i=0; i<h; i++)
- {
- ST32(dst , LD32(src ));
- ST32(dst+4 , LD32(src+4 ));
- ST32(dst+8 , LD32(src+8 ));
- ST32(dst+12, LD32(src+12));
- dst[16]= src[16];
- dst+=dstStride;
- src+=srcStride;
- }
-}
+ assert(x<8 && y<8 && x>=0 && y>=0);
-static inline void copy_block9(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
-{
- int i;
for(i=0; i<h; i++)
{
- ST32(dst , LD32(src ));
- ST32(dst+4 , LD32(src+4 ));
- dst[8]= src[8];
- dst+=dstStride;
- src+=srcStride;
+ dst[0] = (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1] + 32 - 4) >> 6;
+ dst[1] = (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2] + 32 - 4) >> 6;
+ dst[2] = (A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3] + 32 - 4) >> 6;
+ dst[3] = (A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4] + 32 - 4) >> 6;
+ dst[4] = (A*src[4] + B*src[5] + C*src[stride+4] + D*src[stride+5] + 32 - 4) >> 6;
+ dst[5] = (A*src[5] + B*src[6] + C*src[stride+5] + D*src[stride+6] + 32 - 4) >> 6;
+ dst[6] = (A*src[6] + B*src[7] + C*src[stride+6] + D*src[stride+7] + 32 - 4) >> 6;
+ dst[7] = (A*src[7] + B*src[8] + C*src[stride+7] + D*src[stride+8] + 32 - 4) >> 6;
+ dst+= stride;
+ src+= stride;
}
}
-
#define QPEL_MC(r, OPNAME, RND, OP) \
static void OPNAME ## mpeg4_qpel8_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h){\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
for(i=0; i<h; i++)\
{\
@@ -1582,7 +1534,7 @@ static void OPNAME ## mpeg4_qpel8_h_lowpass(uint8_t *dst, uint8_t *src, int dstS
\
static void OPNAME ## mpeg4_qpel8_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
const int w=8;\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
for(i=0; i<w; i++)\
{\
@@ -1609,7 +1561,7 @@ static void OPNAME ## mpeg4_qpel8_v_lowpass(uint8_t *dst, uint8_t *src, int dstS
}\
\
static void OPNAME ## mpeg4_qpel16_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h){\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
\
for(i=0; i<h; i++)\
@@ -1636,7 +1588,7 @@ static void OPNAME ## mpeg4_qpel16_h_lowpass(uint8_t *dst, uint8_t *src, int dst
}\
\
static void OPNAME ## mpeg4_qpel16_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
const int w=16;\
for(i=0; i<w; i++)\
@@ -2062,7 +2014,7 @@ QPEL_MC(0, avg_ , _ , op_avg)
#define H264_LOWPASS(OPNAME, OP, OP2) \
static void OPNAME ## h264_qpel2_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
const int h=2;\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
for(i=0; i<h; i++)\
{\
@@ -2075,7 +2027,7 @@ static void OPNAME ## h264_qpel2_h_lowpass(uint8_t *dst, uint8_t *src, int dstSt
\
static void OPNAME ## h264_qpel2_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
const int w=2;\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
for(i=0; i<w; i++)\
{\
@@ -2096,7 +2048,7 @@ static void OPNAME ## h264_qpel2_v_lowpass(uint8_t *dst, uint8_t *src, int dstSt
static void OPNAME ## h264_qpel2_hv_lowpass(uint8_t *dst, int16_t *tmp, uint8_t *src, int dstStride, int tmpStride, int srcStride){\
const int h=2;\
const int w=2;\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
src -= 2*srcStride;\
for(i=0; i<h+5; i++)\
@@ -2124,7 +2076,7 @@ static void OPNAME ## h264_qpel2_hv_lowpass(uint8_t *dst, int16_t *tmp, uint8_t
}\
static void OPNAME ## h264_qpel4_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
const int h=4;\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
for(i=0; i<h; i++)\
{\
@@ -2139,7 +2091,7 @@ static void OPNAME ## h264_qpel4_h_lowpass(uint8_t *dst, uint8_t *src, int dstSt
\
static void OPNAME ## h264_qpel4_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
const int w=4;\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
for(i=0; i<w; i++)\
{\
@@ -2164,7 +2116,7 @@ static void OPNAME ## h264_qpel4_v_lowpass(uint8_t *dst, uint8_t *src, int dstSt
static void OPNAME ## h264_qpel4_hv_lowpass(uint8_t *dst, int16_t *tmp, uint8_t *src, int dstStride, int tmpStride, int srcStride){\
const int h=4;\
const int w=4;\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
src -= 2*srcStride;\
for(i=0; i<h+5; i++)\
@@ -2199,7 +2151,7 @@ static void OPNAME ## h264_qpel4_hv_lowpass(uint8_t *dst, int16_t *tmp, uint8_t
\
static void OPNAME ## h264_qpel8_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
const int h=8;\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
for(i=0; i<h; i++)\
{\
@@ -2218,7 +2170,7 @@ static void OPNAME ## h264_qpel8_h_lowpass(uint8_t *dst, uint8_t *src, int dstSt
\
static void OPNAME ## h264_qpel8_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
const int w=8;\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
for(i=0; i<w; i++)\
{\
@@ -2251,7 +2203,7 @@ static void OPNAME ## h264_qpel8_v_lowpass(uint8_t *dst, uint8_t *src, int dstSt
static void OPNAME ## h264_qpel8_hv_lowpass(uint8_t *dst, int16_t *tmp, uint8_t *src, int dstStride, int tmpStride, int srcStride){\
const int h=8;\
const int w=8;\
- uint8_t *cm = cropTbl + MAX_NEG_CROP;\
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
int i;\
src -= 2*srcStride;\
for(i=0; i<h+5; i++)\
@@ -2553,7 +2505,7 @@ H264_WEIGHT(2,2)
#undef H264_WEIGHT
static void wmv2_mspel8_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h){
- uint8_t *cm = cropTbl + MAX_NEG_CROP;
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
int i;
for(i=0; i<h; i++){
@@ -2598,7 +2550,7 @@ void ff_put_vc1_mspel_mc00_c(uint8_t *dst, uint8_t *src, int stride, int rnd) {
#endif /* CONFIG_VC1_DECODER||CONFIG_WMV3_DECODER */
static void wmv2_mspel8_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int w){
- uint8_t *cm = cropTbl + MAX_NEG_CROP;
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
int i;
for(i=0; i<w; i++){
@@ -2700,7 +2652,7 @@ static void h263_v_loop_filter_c(uint8_t *src, int stride, int qscale){
src[x-1*stride] = p1;
src[x+0*stride] = p2;
- ad1= ABS(d1)>>1;
+ ad1= FFABS(d1)>>1;
d2= clip((p0-p3)/4, -ad1, ad1);
@@ -2735,7 +2687,7 @@ static void h263_h_loop_filter_c(uint8_t *src, int stride, int qscale){
src[y*stride-1] = p1;
src[y*stride+0] = p2;
- ad1= ABS(d1)>>1;
+ ad1= FFABS(d1)>>1;
d2= clip((p0-p3)/4, -ad1, ad1);
@@ -2787,18 +2739,18 @@ static inline void h264_loop_filter_luma_c(uint8_t *pix, int xstride, int ystrid
const int q1 = pix[1*xstride];
const int q2 = pix[2*xstride];
- if( ABS( p0 - q0 ) < alpha &&
- ABS( p1 - p0 ) < beta &&
- ABS( q1 - q0 ) < beta ) {
+ if( FFABS( p0 - q0 ) < alpha &&
+ FFABS( p1 - p0 ) < beta &&
+ FFABS( q1 - q0 ) < beta ) {
int tc = tc0[i];
int i_delta;
- if( ABS( p2 - p0 ) < beta ) {
+ if( FFABS( p2 - p0 ) < beta ) {
pix[-2*xstride] = p1 + clip( (( p2 + ( ( p0 + q0 + 1 ) >> 1 ) ) >> 1) - p1, -tc0[i], tc0[i] );
tc++;
}
- if( ABS( q2 - q0 ) < beta ) {
+ if( FFABS( q2 - q0 ) < beta ) {
pix[ xstride] = q1 + clip( (( q2 + ( ( p0 + q0 + 1 ) >> 1 ) ) >> 1) - q1, -tc0[i], tc0[i] );
tc++;
}
@@ -2835,9 +2787,9 @@ static inline void h264_loop_filter_chroma_c(uint8_t *pix, int xstride, int ystr
const int q0 = pix[0];
const int q1 = pix[1*xstride];
- if( ABS( p0 - q0 ) < alpha &&
- ABS( p1 - p0 ) < beta &&
- ABS( q1 - q0 ) < beta ) {
+ if( FFABS( p0 - q0 ) < alpha &&
+ FFABS( p1 - p0 ) < beta &&
+ FFABS( q1 - q0 ) < beta ) {
int delta = clip( (((q0 - p0 ) << 2) + (p1 - q1) + 4) >> 3, -tc, tc );
@@ -2866,9 +2818,9 @@ static inline void h264_loop_filter_chroma_intra_c(uint8_t *pix, int xstride, in
const int q0 = pix[0];
const int q1 = pix[1*xstride];
- if( ABS( p0 - q0 ) < alpha &&
- ABS( p1 - p0 ) < beta &&
- ABS( q1 - q0 ) < beta ) {
+ if( FFABS( p0 - q0 ) < alpha &&
+ FFABS( p1 - p0 ) < beta &&
+ FFABS( q1 - q0 ) < beta ) {
pix[-xstride] = ( 2*p1 + p0 + q1 + 2 ) >> 2; /* p0' */
pix[0] = ( 2*q1 + q0 + p1 + 2 ) >> 2; /* q0' */
@@ -3097,9 +3049,9 @@ static int nsse16_c(void *v, uint8_t *s1, uint8_t *s2, int stride, int h){
}
if(y+1<h){
for(x=0; x<15; x++){
- score2+= ABS( s1[x ] - s1[x +stride]
+ score2+= FFABS( s1[x ] - s1[x +stride]
- s1[x+1] + s1[x+1+stride])
- -ABS( s2[x ] - s2[x +stride]
+ -FFABS( s2[x ] - s2[x +stride]
- s2[x+1] + s2[x+1+stride]);
}
}
@@ -3107,8 +3059,8 @@ static int nsse16_c(void *v, uint8_t *s1, uint8_t *s2, int stride, int h){
s2+= stride;
}
- if(c) return score1 + ABS(score2)*c->avctx->nsse_weight;
- else return score1 + ABS(score2)*8;
+ if(c) return score1 + FFABS(score2)*c->avctx->nsse_weight;
+ else return score1 + FFABS(score2)*8;
}
static int nsse8_c(void *v, uint8_t *s1, uint8_t *s2, int stride, int h){
@@ -3123,9 +3075,9 @@ static int nsse8_c(void *v, uint8_t *s1, uint8_t *s2, int stride, int h){
}
if(y+1<h){
for(x=0; x<7; x++){
- score2+= ABS( s1[x ] - s1[x +stride]
+ score2+= FFABS( s1[x ] - s1[x +stride]
- s1[x+1] + s1[x+1+stride])
- -ABS( s2[x ] - s2[x +stride]
+ -FFABS( s2[x ] - s2[x +stride]
- s2[x+1] + s2[x+1+stride]);
}
}
@@ -3133,8 +3085,8 @@ static int nsse8_c(void *v, uint8_t *s1, uint8_t *s2, int stride, int h){
s2+= stride;
}
- if(c) return score1 + ABS(score2)*c->avctx->nsse_weight;
- else return score1 + ABS(score2)*8;
+ if(c) return score1 + FFABS(score2)*c->avctx->nsse_weight;
+ else return score1 + FFABS(score2)*8;
}
static int try_8x8basis_c(int16_t rem[64], int16_t weight[64], int16_t basis[64], int scale){
@@ -3324,7 +3276,7 @@ o2= (i1)-(i2);
y= a-b;\
}
-#define BUTTERFLYA(x,y) (ABS((x)+(y)) + ABS((x)-(y)))
+#define BUTTERFLYA(x,y) (FFABS((x)+(y)) + FFABS((x)-(y)))
static int hadamard8_diff8x8_c(/*MpegEncContext*/ void *s, uint8_t *dst, uint8_t *src, int stride, int h){
int i;
@@ -3421,7 +3373,7 @@ static int hadamard8_intra8x8_c(/*MpegEncContext*/ void *s, uint8_t *src, uint8_
+BUTTERFLYA(temp[8*3+i], temp[8*7+i]);
}
- sum -= ABS(temp[8*0] + temp[8*4]); // -mean
+ sum -= FFABS(temp[8*0] + temp[8*4]); // -mean
return sum;
}
@@ -3438,7 +3390,7 @@ static int dct_sad8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2
s->dsp.fdct(temp);
for(i=0; i<64; i++)
- sum+= ABS(temp[i]);
+ sum+= FFABS(temp[i]);
return sum;
}
@@ -3487,7 +3439,7 @@ static int dct264_sad8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *s
#undef DST
#define SRC(x) dct[x][i]
-#define DST(x,v) sum += ABS(v)
+#define DST(x,v) sum += FFABS(v)
for( i = 0; i < 8; i++ )
DCT8_1D
#undef SRC
@@ -3508,13 +3460,11 @@ static int dct_max8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2
s->dsp.fdct(temp);
for(i=0; i<64; i++)
- sum= FFMAX(sum, ABS(temp[i]));
+ sum= FFMAX(sum, FFABS(temp[i]));
return sum;
}
-void simple_idct(DCTELEM *block); //FIXME
-
static int quant_psnr8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int stride, int h){
MpegEncContext * const s= (MpegEncContext *)c;
DECLARE_ALIGNED_8 (uint64_t, aligned_temp[sizeof(DCTELEM)*64*2/8]);
@@ -3684,8 +3634,8 @@ static int vsad_intra16_c(/*MpegEncContext*/ void *c, uint8_t *s, uint8_t *dummy
for(y=1; y<h; y++){
for(x=0; x<16; x+=4){
- score+= ABS(s[x ] - s[x +stride]) + ABS(s[x+1] - s[x+1+stride])
- +ABS(s[x+2] - s[x+2+stride]) + ABS(s[x+3] - s[x+3+stride]);
+ score+= FFABS(s[x ] - s[x +stride]) + FFABS(s[x+1] - s[x+1+stride])
+ +FFABS(s[x+2] - s[x+2+stride]) + FFABS(s[x+3] - s[x+3+stride]);
}
s+= stride;
}
@@ -3699,7 +3649,7 @@ static int vsad16_c(/*MpegEncContext*/ void *c, uint8_t *s1, uint8_t *s2, int st
for(y=1; y<h; y++){
for(x=0; x<16; x++){
- score+= ABS(s1[x ] - s2[x ] - s1[x +stride] + s2[x +stride]);
+ score+= FFABS(s1[x ] - s2[x ] - s1[x +stride] + s2[x +stride]);
}
s1+= stride;
s2+= stride;
@@ -3750,6 +3700,39 @@ WARPER8_16_SQ(quant_psnr8x8_c, quant_psnr16_c)
WARPER8_16_SQ(rd8x8_c, rd16_c)
WARPER8_16_SQ(bit8x8_c, bit16_c)
+static void vector_fmul_c(float *dst, const float *src, int len){
+ int i;
+ for(i=0; i<len; i++)
+ dst[i] *= src[i];
+}
+
+static void vector_fmul_reverse_c(float *dst, const float *src0, const float *src1, int len){
+ int i;
+ src1 += len-1;
+ for(i=0; i<len; i++)
+ dst[i] = src0[i] * src1[-i];
+}
+
+void ff_vector_fmul_add_add_c(float *dst, const float *src0, const float *src1, const float *src2, int src3, int len, int step){
+ int i;
+ for(i=0; i<len; i++)
+ dst[i*step] = src0[i] * src1[i] + src2[i] + src3;
+}
+
+void ff_float_to_int16_c(int16_t *dst, const float *src, int len){
+ int i;
+ for(i=0; i<len; i++) {
+ int_fast32_t tmp = ((int32_t*)src)[i];
+ if(tmp & 0xf0000){
+ tmp = (0x43c0ffff - tmp)>>31;
+ // is this faster on some gcc/cpu combinations?
+// if(tmp > 0x43c0ffff) tmp = 0xFFFF;
+// else tmp = 0;
+ }
+ dst[i] = tmp - 0x8000;
+ }
+}
+
/* XXX: those functions should be suppressed ASAP when all IDCTs are
converted */
static void ff_jref_idct_put(uint8_t *dest, int line_size, DCTELEM *block)
@@ -3787,13 +3770,13 @@ static void ff_jref_idct2_add(uint8_t *dest, int line_size, DCTELEM *block)
static void ff_jref_idct1_put(uint8_t *dest, int line_size, DCTELEM *block)
{
- uint8_t *cm = cropTbl + MAX_NEG_CROP;
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
dest[0] = cm[(block[0] + 4)>>3];
}
static void ff_jref_idct1_add(uint8_t *dest, int line_size, DCTELEM *block)
{
- uint8_t *cm = cropTbl + MAX_NEG_CROP;
+ uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
dest[0] = cm[dest[0] + ((block[0] + 4)>>3)];
}
@@ -3805,14 +3788,14 @@ void dsputil_static_init(void)
{
int i;
- for(i=0;i<256;i++) cropTbl[i + MAX_NEG_CROP] = i;
+ for(i=0;i<256;i++) ff_cropTbl[i + MAX_NEG_CROP] = i;
for(i=0;i<MAX_NEG_CROP;i++) {
- cropTbl[i] = 0;
- cropTbl[i + MAX_NEG_CROP + 256] = 255;
+ ff_cropTbl[i] = 0;
+ ff_cropTbl[i + MAX_NEG_CROP + 256] = 255;
}
for(i=0;i<512;i++) {
- squareTbl[i] = (i - 256) * (i - 256);
+ ff_squareTbl[i] = (i - 256) * (i - 256);
}
for(i=0; i<64; i++) inv_zigzag_direct16[ff_zigzag_direct[i]]= i+1;
@@ -3994,6 +3977,7 @@ void dsputil_init(DSPContext* c, AVCodecContext *avctx)
c->avg_h264_chroma_pixels_tab[0]= avg_h264_chroma_mc8_c;
c->avg_h264_chroma_pixels_tab[1]= avg_h264_chroma_mc4_c;
c->avg_h264_chroma_pixels_tab[2]= avg_h264_chroma_mc2_c;
+ c->put_no_rnd_h264_chroma_pixels_tab[0]= put_no_rnd_h264_chroma_mc8_c;
c->weight_h264_pixels_tab[0]= weight_h264_pixels16x16_c;
c->weight_h264_pixels_tab[1]= weight_h264_pixels16x8_c;
@@ -4075,6 +4059,7 @@ void dsputil_init(DSPContext* c, AVCodecContext *avctx)
c->h264_h_loop_filter_chroma= h264_h_loop_filter_chroma_c;
c->h264_v_loop_filter_chroma_intra= h264_v_loop_filter_chroma_intra_c;
c->h264_h_loop_filter_chroma_intra= h264_h_loop_filter_chroma_intra_c;
+ c->h264_loop_filter_strength= NULL;
c->h263_h_loop_filter= h263_h_loop_filter_c;
c->h263_v_loop_filter= h263_v_loop_filter_c;
@@ -4090,6 +4075,14 @@ void dsputil_init(DSPContext* c, AVCodecContext *avctx)
c->inner_add_yblock = ff_snow_inner_add_yblock;
#endif
+#ifdef CONFIG_VORBIS_DECODER
+ c->vorbis_inverse_coupling = vorbis_inverse_coupling;
+#endif
+ c->vector_fmul = vector_fmul_c;
+ c->vector_fmul_reverse = vector_fmul_reverse_c;
+ c->vector_fmul_add_add = ff_vector_fmul_add_add_c;
+ c->float_to_int16 = ff_float_to_int16_c;
+
c->shrink[0]= ff_img_copy_plane;
c->shrink[1]= ff_shrink22;
c->shrink[2]= ff_shrink44;
@@ -4097,6 +4090,9 @@ void dsputil_init(DSPContext* c, AVCodecContext *avctx)
c->prefetch= just_return;
+ memset(c->put_2tap_qpel_pixels_tab, 0, sizeof(c->put_2tap_qpel_pixels_tab));
+ memset(c->avg_2tap_qpel_pixels_tab, 0, sizeof(c->avg_2tap_qpel_pixels_tab));
+
#ifdef HAVE_MMX
dsputil_init_mmx(c, avctx);
#endif
@@ -4121,6 +4117,16 @@ void dsputil_init(DSPContext* c, AVCodecContext *avctx)
#ifdef ARCH_SH4
dsputil_init_sh4(c,avctx);
#endif
+#ifdef ARCH_BFIN
+ dsputil_init_bfin(c,avctx);
+#endif
+
+ for(i=0; i<64; i++){
+ if(!c->put_2tap_qpel_pixels_tab[0][i])
+ c->put_2tap_qpel_pixels_tab[0][i]= c->put_h264_qpel_pixels_tab[0][i];
+ if(!c->avg_2tap_qpel_pixels_tab[0][i])
+ c->avg_2tap_qpel_pixels_tab[0][i]= c->avg_h264_qpel_pixels_tab[0][i];
+ }
switch(c->idct_permutation_type){
case FF_NO_IDCT_PERM:
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-25 15:53:52 +0000