#include "avcodec.h"
#include "dsputil.h"
-#include "mpegvideo.h"
#include "simple_idct.h"
#include "faandct.h"
+#include "faanidct.h"
#include "h263.h"
#include "snow.h"
uint8_t ff_cropTbl[256 + 2 * MAX_NEG_CROP] = {0, };
uint32_t ff_squareTbl[512] = {0, };
+// 0x7f7f7f7f or 0x7f7f7f7f7f7f7f7f or whatever, depending on the cpu's native arithmetic size
+#define pb_7f (~0UL/255 * 0x7f)
+#define pb_80 (~0UL/255 * 0x80)
+
const uint8_t ff_zigzag_direct[64] = {
0, 1, 8, 16, 9, 2, 3, 10,
17, 24, 32, 25, 18, 11, 4, 5,
0x32, 0x3A, 0x36, 0x3B, 0x33, 0x3E, 0x37, 0x3F,
};
+static const uint8_t idct_sse2_row_perm[8] = {0, 4, 1, 5, 2, 6, 3, 7};
+
+void ff_init_scantable(uint8_t *permutation, ScanTable *st, const uint8_t *src_scantable){
+ int i;
+ int end;
+
+ st->scantable= src_scantable;
+
+ for(i=0; i<64; i++){
+ int j;
+ j = src_scantable[i];
+ st->permutated[i] = permutation[j];
+#ifdef ARCH_POWERPC
+ st->inverse[j] = i;
+#endif
+ }
+
+ end=-1;
+ for(i=0; i<64; i++){
+ int j;
+ j = st->permutated[i];
+ if(j>end) end=j;
+ st->raster_end[i]= end;
+ }
+}
+
static int pix_sum_c(uint8_t * pix, int line_size)
{
int s, i, j;
}
#endif
+/* draw the edges of width 'w' of an image of size width, height */
+//FIXME check that this is ok for mpeg4 interlaced
+static void draw_edges_c(uint8_t *buf, int wrap, int width, int height, int w)
+{
+ uint8_t *ptr, *last_line;
+ int i;
+
+ last_line = buf + (height - 1) * wrap;
+ for(i=0;i<w;i++) {
+ /* top and bottom */
+ memcpy(buf - (i + 1) * wrap, buf, width);
+ memcpy(last_line + (i + 1) * wrap, last_line, width);
+ }
+ /* left and right */
+ ptr = buf;
+ for(i=0;i<height;i++) {
+ memset(ptr - w, ptr[0], w);
+ memset(ptr + width, ptr[width-1], w);
+ ptr += wrap;
+ }
+ /* corners */
+ for(i=0;i<w;i++) {
+ memset(buf - (i + 1) * wrap - w, buf[0], w); /* top left */
+ memset(buf - (i + 1) * wrap + width, buf[width-1], w); /* top right */
+ memset(last_line + (i + 1) * wrap - w, last_line[0], w); /* top left */
+ memset(last_line + (i + 1) * wrap + width, last_line[width-1], w); /* top right */
+ }
+}
+
+/**
+ * Copies a rectangular area of samples to a temporary buffer and replicates the boarder samples.
+ * @param buf destination buffer
+ * @param src source buffer
+ * @param linesize number of bytes between 2 vertically adjacent samples in both the source and destination buffers
+ * @param block_w width of block
+ * @param block_h height of block
+ * @param src_x x coordinate of the top left sample of the block in the source buffer
+ * @param src_y y coordinate of the top left sample of the block in the source buffer
+ * @param w width of the source buffer
+ * @param h height of the source buffer
+ */
+void ff_emulated_edge_mc(uint8_t *buf, uint8_t *src, int linesize, int block_w, int block_h,
+ int src_x, int src_y, int w, int h){
+ int x, y;
+ int start_y, start_x, end_y, end_x;
+
+ if(src_y>= h){
+ src+= (h-1-src_y)*linesize;
+ src_y=h-1;
+ }else if(src_y<=-block_h){
+ src+= (1-block_h-src_y)*linesize;
+ src_y=1-block_h;
+ }
+ if(src_x>= w){
+ src+= (w-1-src_x);
+ src_x=w-1;
+ }else if(src_x<=-block_w){
+ src+= (1-block_w-src_x);
+ src_x=1-block_w;
+ }
+
+ start_y= FFMAX(0, -src_y);
+ start_x= FFMAX(0, -src_x);
+ end_y= FFMIN(block_h, h-src_y);
+ end_x= FFMIN(block_w, w-src_x);
+
+ // copy existing part
+ for(y=start_y; y<end_y; y++){
+ for(x=start_x; x<end_x; x++){
+ buf[x + y*linesize]= src[x + y*linesize];
+ }
+ }
+
+ //top
+ for(y=0; y<start_y; y++){
+ for(x=start_x; x<end_x; x++){
+ buf[x + y*linesize]= buf[x + start_y*linesize];
+ }
+ }
+
+ //bottom
+ for(y=end_y; y<block_h; y++){
+ for(x=start_x; x<end_x; x++){
+ buf[x + y*linesize]= buf[x + (end_y-1)*linesize];
+ }
+ }
+
+ for(y=0; y<block_h; y++){
+ //left
+ for(x=0; x<start_x; x++){
+ buf[x + y*linesize]= buf[start_x + y*linesize];
+ }
+
+ //right
+ for(x=end_x; x<block_w; x++){
+ buf[x + y*linesize]= buf[end_x - 1 + y*linesize];
+ }
+ }
+}
+
static void get_pixels_c(DCTELEM *restrict block, const uint8_t *pixels, int line_size)
{
int i;
}
static void add_bytes_c(uint8_t *dst, uint8_t *src, int w){
- int i;
- for(i=0; i+7<w; i+=8){
- dst[i+0] += src[i+0];
- dst[i+1] += src[i+1];
- dst[i+2] += src[i+2];
- dst[i+3] += src[i+3];
- dst[i+4] += src[i+4];
- dst[i+5] += src[i+5];
- dst[i+6] += src[i+6];
- dst[i+7] += src[i+7];
+ long i;
+ for(i=0; i<=w-sizeof(long); i+=sizeof(long)){
+ long a = *(long*)(src+i);
+ long b = *(long*)(dst+i);
+ *(long*)(dst+i) = ((a&pb_7f) + (b&pb_7f)) ^ ((a^b)&pb_80);
}
for(; i<w; i++)
dst[i+0] += src[i+0];
}
static void add_bytes_l2_c(uint8_t *dst, uint8_t *src1, uint8_t *src2, int w){
- int i;
+ long i;
for(i=0; i<=w-sizeof(long); i+=sizeof(long)){
long a = *(long*)(src1+i);
long b = *(long*)(src2+i);
- *(long*)(dst+i) = ((a&0x7f7f7f7f7f7f7f7fL) + (b&0x7f7f7f7f7f7f7f7fL)) ^ ((a^b)&0x8080808080808080L);
+ *(long*)(dst+i) = ((a&pb_7f) + (b&pb_7f)) ^ ((a^b)&pb_80);
}
for(; i<w; i++)
dst[i] = src1[i]+src2[i];
}
static void diff_bytes_c(uint8_t *dst, uint8_t *src1, uint8_t *src2, int w){
- int i;
- for(i=0; i+7<w; i+=8){
- dst[i+0] = src1[i+0]-src2[i+0];
- dst[i+1] = src1[i+1]-src2[i+1];
- dst[i+2] = src1[i+2]-src2[i+2];
- dst[i+3] = src1[i+3]-src2[i+3];
- dst[i+4] = src1[i+4]-src2[i+4];
- dst[i+5] = src1[i+5]-src2[i+5];
- dst[i+6] = src1[i+6]-src2[i+6];
- dst[i+7] = src1[i+7]-src2[i+7];
+ long i;
+#ifndef HAVE_FAST_UNALIGNED
+ if((long)src2 & (sizeof(long)-1)){
+ for(i=0; i+7<w; i+=8){
+ dst[i+0] = src1[i+0]-src2[i+0];
+ dst[i+1] = src1[i+1]-src2[i+1];
+ dst[i+2] = src1[i+2]-src2[i+2];
+ dst[i+3] = src1[i+3]-src2[i+3];
+ dst[i+4] = src1[i+4]-src2[i+4];
+ dst[i+5] = src1[i+5]-src2[i+5];
+ dst[i+6] = src1[i+6]-src2[i+6];
+ dst[i+7] = src1[i+7]-src2[i+7];
+ }
+ }else
+#endif
+ for(i=0; i<=w-sizeof(long); i+=sizeof(long)){
+ long a = *(long*)(src1+i);
+ long b = *(long*)(src2+i);
+ *(long*)(dst+i) = ((a|pb_80) - (b&pb_7f)) ^ ((a^b^pb_80)&pb_80);
}
for(; i<w; i++)
dst[i+0] = src1[i+0]-src2[i+0];
DECLARE_ALIGNED_8 (uint64_t, aligned_bak[stride]);
DCTELEM * const temp= (DCTELEM*)aligned_temp;
uint8_t * const bak= (uint8_t*)aligned_bak;
- int i, last, run, bits, level, distoration, start_i;
+ int i, last, run, bits, level, distortion, start_i;
const int esc_length= s->ac_esc_length;
uint8_t * length;
uint8_t * last_length;
s->dsp.idct_add(bak, stride, temp);
- distoration= s->dsp.sse[1](NULL, bak, src1, stride, 8);
+ distortion= s->dsp.sse[1](NULL, bak, src1, stride, 8);
- return distoration + ((bits*s->qscale*s->qscale*109 + 64)>>7);
+ return distortion + ((bits*s->qscale*s->qscale*109 + 64)>>7);
}
static int bit8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int stride, int h){
c->idct_add= ff_wmv2_idct_add_c;
c->idct = ff_wmv2_idct_c;
c->idct_permutation_type= FF_NO_IDCT_PERM;
+ }else if(avctx->idct_algo==FF_IDCT_FAAN){
+ c->idct_put= ff_faanidct_put;
+ c->idct_add= ff_faanidct_add;
+ c->idct = ff_faanidct;
+ c->idct_permutation_type= FF_NO_IDCT_PERM;
}else{ //accurate/default
c->idct_put= ff_simple_idct_put;
c->idct_add= ff_simple_idct_add;
c->biweight_h264_pixels_tab[8]= biweight_h264_pixels2x4_c;
c->biweight_h264_pixels_tab[9]= biweight_h264_pixels2x2_c;
+ c->draw_edges = draw_edges_c;
+
#ifdef CONFIG_CAVS_DECODER
ff_cavsdsp_init(c,avctx);
#endif
for(i=0; i<64; i++)
c->idct_permutation[i]= (i&0x24) | ((i&3)<<3) | ((i>>3)&3);
break;
+ case FF_SSE2_IDCT_PERM:
+ for(i=0; i<64; i++)
+ c->idct_permutation[i]= (i&0x38) | idct_sse2_row_perm[i&7];
+ break;
default:
av_log(avctx, AV_LOG_ERROR, "Internal error, IDCT permutation not set\n");
}