#include "faandct.h"
#include "faanidct.h"
#include "mathops.h"
-#include "h263.h"
#include "snow.h"
-
-/* snow.c */
-void ff_spatial_dwt(int *buffer, int width, int height, int stride, int type, int decomposition_count);
-
-/* vorbis.c */
-void vorbis_inverse_coupling(float *mag, float *ang, int blocksize);
-
-/* ac3dec.c */
-void ff_ac3_downmix_c(float (*samples)[256], float (*matrix)[2], int out_ch, int in_ch, int len);
-
-/* flacenc.c */
-void ff_flac_compute_autocorr(const int32_t *data, int len, int lag, double *autoc);
-
-/* pngdec.c */
-void ff_add_png_paeth_prediction(uint8_t *dst, uint8_t *src, uint8_t *top, int w, int bpp);
-
-/* eaidct.c */
-void ff_ea_idct_put_c(uint8_t *dest, int linesize, DCTELEM *block);
+#include "mpegvideo.h"
+#include "config.h"
+#include "lpc.h"
+#include "ac3dec.h"
+#include "vorbis.h"
+#include "png.h"
uint8_t ff_cropTbl[256 + 2 * MAX_NEG_CROP] = {0, };
uint32_t ff_squareTbl[512] = {0, };
};
/* not permutated inverse zigzag_direct + 1 for MMX quantizer */
-DECLARE_ALIGNED_16(uint16_t, inv_zigzag_direct16[64]);
+DECLARE_ALIGNED(16, uint16_t, inv_zigzag_direct16)[64];
const uint8_t ff_alternate_horizontal_scan[64] = {
0, 1, 2, 3, 8, 9, 16, 17,
}
}
+static void put_pixels_nonclamped_c(const DCTELEM *block, uint8_t *restrict pixels,
+ int line_size)
+{
+ int i;
+
+ /* read the pixels */
+ for(i=0;i<8;i++) {
+ pixels[0] = block[0];
+ pixels[1] = block[1];
+ pixels[2] = block[2];
+ pixels[3] = block[3];
+ pixels[4] = block[4];
+ pixels[5] = block[5];
+ pixels[6] = block[6];
+ pixels[7] = block[7];
+
+ pixels += line_size;
+ block += 8;
+ }
+}
+
static void add_pixels_clamped_c(const DCTELEM *block, uint8_t *restrict pixels,
int line_size)
{
return sum;
}
+static void fill_block16_c(uint8_t *block, uint8_t value, int line_size, int h)
+{
+ int i;
+
+ for (i = 0; i < h; i++) {
+ memset(block, value, 16);
+ block += line_size;
+ }
+}
+
+static void fill_block8_c(uint8_t *block, uint8_t value, int line_size, int h)
+{
+ int i;
+
+ for (i = 0; i < h; i++) {
+ memset(block, value, 8);
+ block += line_size;
+ }
+}
+
+static void scale_block_c(const uint8_t src[64]/*align 8*/, uint8_t *dst/*align 8*/, int linesize)
+{
+ int i, j;
+ uint16_t *dst1 = dst;
+ uint16_t *dst2 = dst + linesize;
+
+ for (j = 0; j < 8; j++) {
+ for (i = 0; i < 8; i++) {
+ dst1[i] = dst2[i] = src[i] * 0x0101;
+ }
+ src += 8;
+ dst1 += linesize;
+ dst2 += linesize;
+ }
+}
+
#if 0
#define PIXOP2(OPNAME, OP) \
#if CONFIG_CAVS_DECODER
/* AVS specific */
-void ff_cavsdsp_init(DSPContext* c, AVCodecContext *avctx);
-
void ff_put_cavs_qpel8_mc00_c(uint8_t *dst, uint8_t *src, int stride) {
put_pixels8_c(dst, src, stride, 8);
}
}
#endif /* CONFIG_CAVS_DECODER */
-void ff_mlp_init(DSPContext* c, AVCodecContext *avctx);
-
#if CONFIG_VC1_DECODER
/* VC-1 specific */
-void ff_vc1dsp_init(DSPContext* c, AVCodecContext *avctx);
-
-void ff_put_vc1_mspel_mc00_c(uint8_t *dst, uint8_t *src, int stride, int rnd) {
+void ff_put_vc1_mspel_mc00_c(uint8_t *dst, const uint8_t *src, int stride, int rnd) {
put_pixels8_c(dst, src, stride, 8);
}
-void ff_avg_vc1_mspel_mc00_c(uint8_t *dst, uint8_t *src, int stride, int rnd) {
+void ff_avg_vc1_mspel_mc00_c(uint8_t *dst, const uint8_t *src, int stride, int rnd) {
avg_pixels8_c(dst, src, stride, 8);
}
#endif /* CONFIG_VC1_DECODER */
-void ff_intrax8dsp_init(DSPContext* c, AVCodecContext *avctx);
-
/* H264 specific */
void ff_h264dspenc_init(DSPContext* c, AVCodecContext *avctx);
-#if CONFIG_RV30_DECODER
-void ff_rv30dsp_init(DSPContext* c, AVCodecContext *avctx);
-#endif /* CONFIG_RV30_DECODER */
-
#if CONFIG_RV40_DECODER
static void put_rv40_qpel16_mc33_c(uint8_t *dst, uint8_t *src, int stride){
put_pixels16_xy2_c(dst, src, stride, 16);
static void avg_rv40_qpel8_mc33_c(uint8_t *dst, uint8_t *src, int stride){
avg_pixels8_xy2_c(dst, src, stride, 8);
}
-
-void ff_rv40dsp_init(DSPContext* c, AVCodecContext *avctx);
#endif /* CONFIG_RV40_DECODER */
static void wmv2_mspel8_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int w){
}
static void h263_v_loop_filter_c(uint8_t *src, int stride, int qscale){
- if(CONFIG_ANY_H263) {
+ if(CONFIG_H263_DECODER || CONFIG_H263_ENCODER) {
int x;
const int strength= ff_h263_loop_filter_strength[qscale];
}
static void h263_h_loop_filter_c(uint8_t *src, int stride, int qscale){
- if(CONFIG_ANY_H263) {
+ if(CONFIG_H263_DECODER || CONFIG_H263_ENCODER) {
int y;
const int strength= ff_h263_loop_filter_strength[qscale];
}
}
-static inline void h264_loop_filter_luma_c(uint8_t *pix, int xstride, int ystride, int alpha, int beta, int8_t *tc0)
+static av_always_inline av_flatten void h264_loop_filter_luma_c(uint8_t *pix, int xstride, int ystride, int alpha, int beta, int8_t *tc0)
{
int i, d;
for( i = 0; i < 4; i++ ) {
int i_delta;
if( FFABS( p2 - p0 ) < beta ) {
+ if(tc0[i])
pix[-2*xstride] = p1 + av_clip( (( p2 + ( ( p0 + q0 + 1 ) >> 1 ) ) >> 1) - p1, -tc0[i], tc0[i] );
tc++;
}
if( FFABS( q2 - q0 ) < beta ) {
+ if(tc0[i])
pix[ xstride] = q1 + av_clip( (( q2 + ( ( p0 + q0 + 1 ) >> 1 ) ) >> 1) - q1, -tc0[i], tc0[i] );
tc++;
}
h264_loop_filter_luma_c(pix, 1, stride, alpha, beta, tc0);
}
-static inline void h264_loop_filter_luma_intra_c(uint8_t *pix, int xstride, int ystride, int alpha, int beta)
+static av_always_inline av_flatten void h264_loop_filter_luma_intra_c(uint8_t *pix, int xstride, int ystride, int alpha, int beta)
{
int d;
for( d = 0; d < 16; d++ ) {
h264_loop_filter_luma_intra_c(pix, 1, stride, alpha, beta);
}
-static inline void h264_loop_filter_chroma_c(uint8_t *pix, int xstride, int ystride, int alpha, int beta, int8_t *tc0)
+static av_always_inline av_flatten void h264_loop_filter_chroma_c(uint8_t *pix, int xstride, int ystride, int alpha, int beta, int8_t *tc0)
{
int i, d;
for( i = 0; i < 4; i++ ) {
h264_loop_filter_chroma_c(pix, 1, stride, alpha, beta, tc0);
}
-static inline void h264_loop_filter_chroma_intra_c(uint8_t *pix, int xstride, int ystride, int alpha, int beta)
+static av_always_inline av_flatten void h264_loop_filter_chroma_intra_c(uint8_t *pix, int xstride, int ystride, int alpha, int beta)
{
int d;
for( d = 0; d < 8; d++ ) {
dst[i+0] = src1[i+0]-src2[i+0];
}
-static void add_hfyu_median_prediction_c(uint8_t *dst, uint8_t *src1, uint8_t *diff, int w, int *left, int *left_top){
+static void add_hfyu_median_prediction_c(uint8_t *dst, const uint8_t *src1, const uint8_t *diff, int w, int *left, int *left_top){
int i;
uint8_t l, lt;
*left_top= lt;
}
-static void sub_hfyu_median_prediction_c(uint8_t *dst, uint8_t *src1, uint8_t *src2, int w, int *left, int *left_top){
+static void sub_hfyu_median_prediction_c(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int w, int *left, int *left_top){
int i;
uint8_t l, lt;
*left_top= lt;
}
-static int add_hfyu_left_prediction_c(uint8_t *dst, uint8_t *src, int w, int acc){
+static int add_hfyu_left_prediction_c(uint8_t *dst, const uint8_t *src, int w, int acc){
int i;
for(i=0; i<w-1; i++){
#define B 3
#define G 2
#define R 1
+#define A 0
#else
#define B 0
#define G 1
#define R 2
+#define A 3
#endif
-static inline void add_hfyu_left_prediction_bgr32_c(uint8_t *dst, uint8_t *src, int w, int *red, int *green, int *blue){
+static void add_hfyu_left_prediction_bgr32_c(uint8_t *dst, const uint8_t *src, int w, int *red, int *green, int *blue, int *alpha){
int i;
- int r,g,b;
+ int r,g,b,a;
r= *red;
g= *green;
b= *blue;
+ a= *alpha;
for(i=0; i<w; i++){
b+= src[4*i+B];
g+= src[4*i+G];
r+= src[4*i+R];
+ a+= src[4*i+A];
dst[4*i+B]= b;
dst[4*i+G]= g;
dst[4*i+R]= r;
+ dst[4*i+A]= a;
}
*red= r;
*green= g;
*blue= b;
+ *alpha= a;
}
#undef B
#undef G
#undef R
+#undef A
#define BUTTERFLY2(o1,o2,i1,i2) \
o1= (i1)+(i2);\
static int dct_sad8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int stride, int h){
MpegEncContext * const s= (MpegEncContext *)c;
- DECLARE_ALIGNED_16(uint64_t, aligned_temp[sizeof(DCTELEM)*64/8]);
- DCTELEM * const temp= (DCTELEM*)aligned_temp;
+ LOCAL_ALIGNED_16(DCTELEM, temp, [64]);
assert(h==8);
static int dct_max8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int stride, int h){
MpegEncContext * const s= (MpegEncContext *)c;
- DECLARE_ALIGNED_16(uint64_t, aligned_temp[sizeof(DCTELEM)*64/8]);
- DCTELEM * const temp= (DCTELEM*)aligned_temp;
+ LOCAL_ALIGNED_16(DCTELEM, temp, [64]);
int sum=0, i;
assert(h==8);
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_16(uint64_t, aligned_temp[sizeof(DCTELEM)*64*2/8]);
- DCTELEM * const temp= (DCTELEM*)aligned_temp;
- DCTELEM * const bak = ((DCTELEM*)aligned_temp)+64;
+ LOCAL_ALIGNED_16(DCTELEM, temp, [64*2]);
+ DCTELEM * const bak = temp+64;
int sum=0, i;
assert(h==8);
static int rd8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int stride, int h){
MpegEncContext * const s= (MpegEncContext *)c;
const uint8_t *scantable= s->intra_scantable.permutated;
- DECLARE_ALIGNED_16(uint64_t, aligned_temp[sizeof(DCTELEM)*64/8]);
- DECLARE_ALIGNED_16(uint64_t, aligned_src1[8]);
- DECLARE_ALIGNED_16(uint64_t, aligned_src2[8]);
- DCTELEM * const temp= (DCTELEM*)aligned_temp;
- uint8_t * const lsrc1 = (uint8_t*)aligned_src1;
- uint8_t * const lsrc2 = (uint8_t*)aligned_src2;
+ LOCAL_ALIGNED_16(DCTELEM, temp, [64]);
+ LOCAL_ALIGNED_16(uint8_t, lsrc1, [64]);
+ LOCAL_ALIGNED_16(uint8_t, lsrc2, [64]);
int i, last, run, bits, level, distortion, start_i;
const int esc_length= s->ac_esc_length;
uint8_t * length;
static int bit8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int stride, int h){
MpegEncContext * const s= (MpegEncContext *)c;
const uint8_t *scantable= s->intra_scantable.permutated;
- DECLARE_ALIGNED_16(uint64_t, aligned_temp[sizeof(DCTELEM)*64/8]);
- DCTELEM * const temp= (DCTELEM*)aligned_temp;
+ LOCAL_ALIGNED_16(DCTELEM, temp, [64]);
int i, last, run, bits, level, start_i;
const int esc_length= s->ac_esc_length;
uint8_t * length;
}
}
-static void add_int16_c(int16_t * v1, int16_t * v2, int order)
-{
- while (order--)
- *v1++ += *v2++;
-}
-
-static void sub_int16_c(int16_t * v1, int16_t * v2, int order)
-{
- while (order--)
- *v1++ -= *v2++;
-}
-
static int32_t scalarproduct_int16_c(int16_t * v1, int16_t * v2, int order, int shift)
{
int res = 0;
return res;
}
+static int32_t scalarproduct_and_madd_int16_c(int16_t *v1, int16_t *v2, int16_t *v3, int order, int mul)
+{
+ int res = 0;
+ while (order--) {
+ res += *v1 * *v2++;
+ *v1++ += mul * *v3++;
+ }
+ return res;
+}
+
#define W0 2048
#define W1 2841 /* 2048*sqrt (2)*cos (1*pi/16) */
#define W2 2676 /* 2048*sqrt (2)*cos (2*pi/16) */
static void just_return(void *mem av_unused, int stride av_unused, int h av_unused) { return; }
/* init static data */
-void dsputil_static_init(void)
+av_cold void dsputil_static_init(void)
{
int i;
int ff_check_alignment(void){
static int did_fail=0;
- DECLARE_ALIGNED_16(int, aligned);
+ DECLARE_ALIGNED(16, int, aligned);
if((intptr_t)&aligned & 15){
if(!did_fail){
return 0;
}
-void dsputil_init(DSPContext* c, AVCodecContext *avctx)
+av_cold void dsputil_init(DSPContext* c, AVCodecContext *avctx)
{
int i;
}else if(CONFIG_EATGQ_DECODER && avctx->idct_algo==FF_IDCT_EA) {
c->idct_put= ff_ea_idct_put_c;
c->idct_permutation_type= FF_NO_IDCT_PERM;
+ }else if(CONFIG_BINK_DECODER && avctx->idct_algo==FF_IDCT_BINK) {
+ c->idct = ff_bink_idct_c;
+ c->idct_add = ff_bink_idct_add_c;
+ c->idct_put = ff_bink_idct_put_c;
+ 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->diff_pixels = diff_pixels_c;
c->put_pixels_clamped = put_pixels_clamped_c;
c->put_signed_pixels_clamped = put_signed_pixels_clamped_c;
+ c->put_pixels_nonclamped = put_pixels_nonclamped_c;
c->add_pixels_clamped = add_pixels_clamped_c;
c->add_pixels8 = add_pixels8_c;
c->add_pixels4 = add_pixels4_c;
c->pix_sum = pix_sum_c;
c->pix_norm1 = pix_norm1_c;
+ c->fill_block_tab[0] = fill_block16_c;
+ c->fill_block_tab[1] = fill_block8_c;
+ c->scale_block = scale_block_c;
+
/* TODO [0] 16 [1] 8 */
c->pix_abs[0][0] = pix_abs16_c;
c->pix_abs[0][1] = pix_abs16_x2_c;
c->h264_h_loop_filter_chroma_intra= h264_h_loop_filter_chroma_intra_c;
c->h264_loop_filter_strength= NULL;
- if (CONFIG_ANY_H263) {
+ if (CONFIG_H263_DECODER || CONFIG_H263_ENCODER) {
c->h263_h_loop_filter= h263_h_loop_filter_c;
c->h263_v_loop_filter= h263_v_loop_filter_c;
}
#if CONFIG_AC3_DECODER
c->ac3_downmix = ff_ac3_downmix_c;
#endif
-#if CONFIG_FLAC_ENCODER
- c->flac_compute_autocorr = ff_flac_compute_autocorr;
+#if CONFIG_LPC
+ c->lpc_compute_autocorr = ff_lpc_compute_autocorr;
#endif
c->vector_fmul = vector_fmul_c;
c->vector_fmul_reverse = vector_fmul_reverse_c;
c->vector_clipf = vector_clipf_c;
c->float_to_int16 = ff_float_to_int16_c;
c->float_to_int16_interleave = ff_float_to_int16_interleave_c;
- c->add_int16 = add_int16_c;
- c->sub_int16 = sub_int16_c;
c->scalarproduct_int16 = scalarproduct_int16_c;
+ c->scalarproduct_and_madd_int16 = scalarproduct_and_madd_int16_c;
c->scalarproduct_float = scalarproduct_float_c;
c->butterflies_float = butterflies_float_c;
c->vector_fmul_scalar = vector_fmul_scalar_c;
projects / ffmpeg / blobdiff