summaryrefslogtreecommitdiff
path: root/libavcodec/dsputil.c
diff options
context:
space:
mode:
Diffstat (limited to 'libavcodec/dsputil.c')
-rw-r--r--libavcodec/dsputil.c187
1 files changed, 174 insertions, 13 deletions
diff --git a/libavcodec/dsputil.c b/libavcodec/dsputil.c
index 942f606ea8..81521ea376 100644
--- a/libavcodec/dsputil.c
+++ b/libavcodec/dsputil.c
@@ -5,20 +5,20 @@
*
* gmc & q-pel & 32/64 bit based MC by Michael Niedermayer <michaelni@gmx.at>
*
- * This file is part of Libav.
+ * This file is part of FFmpeg.
*
- * Libav is free software; you can redistribute it and/or
+ * 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 2.1 of the License, or (at your option) any later version.
*
- * Libav 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 Libav; 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
*/
@@ -38,6 +38,7 @@
#include "config.h"
#include "ac3dec.h"
#include "vorbis.h"
+#include "diracdsp.h"
uint8_t ff_cropTbl[256 + 2 * MAX_NEG_CROP] = {0, };
uint32_t ff_squareTbl[512] = {0, };
@@ -380,6 +381,38 @@ void ff_put_pixels_clamped_c(const DCTELEM *block, uint8_t *restrict pixels,
}
}
+static void put_pixels_clamped4_c(const DCTELEM *block, uint8_t *restrict pixels,
+ int line_size)
+{
+ int i;
+
+ /* read the pixels */
+ for(i=0;i<4;i++) {
+ pixels[0] = av_clip_uint8(block[0]);
+ pixels[1] = av_clip_uint8(block[1]);
+ pixels[2] = av_clip_uint8(block[2]);
+ pixels[3] = av_clip_uint8(block[3]);
+
+ pixels += line_size;
+ block += 8;
+ }
+}
+
+static void put_pixels_clamped2_c(const DCTELEM *block, uint8_t *restrict pixels,
+ int line_size)
+{
+ int i;
+
+ /* read the pixels */
+ for(i=0;i<2;i++) {
+ pixels[0] = av_clip_uint8(block[0]);
+ pixels[1] = av_clip_uint8(block[1]);
+
+ pixels += line_size;
+ block += 8;
+ }
+}
+
void ff_put_signed_pixels_clamped_c(const DCTELEM *block,
uint8_t *restrict pixels,
int line_size)
@@ -421,6 +454,36 @@ void ff_add_pixels_clamped_c(const DCTELEM *block, uint8_t *restrict pixels,
}
}
+static void add_pixels_clamped4_c(const DCTELEM *block, uint8_t *restrict pixels,
+ int line_size)
+{
+ int i;
+
+ /* read the pixels */
+ for(i=0;i<4;i++) {
+ pixels[0] = av_clip_uint8(pixels[0] + block[0]);
+ pixels[1] = av_clip_uint8(pixels[1] + block[1]);
+ pixels[2] = av_clip_uint8(pixels[2] + block[2]);
+ pixels[3] = av_clip_uint8(pixels[3] + block[3]);
+ pixels += line_size;
+ block += 8;
+ }
+}
+
+static void add_pixels_clamped2_c(const DCTELEM *block, uint8_t *restrict pixels,
+ int line_size)
+{
+ int i;
+
+ /* read the pixels */
+ for(i=0;i<2;i++) {
+ pixels[0] = av_clip_uint8(pixels[0] + block[0]);
+ pixels[1] = av_clip_uint8(pixels[1] + block[1]);
+ pixels += line_size;
+ block += 8;
+ }
+}
+
static int sum_abs_dctelem_c(DCTELEM *block)
{
int sum=0, i;
@@ -1257,6 +1320,51 @@ void ff_avg_rv40_qpel8_mc33_c(uint8_t *dst, uint8_t *src, int stride){
}
#endif /* CONFIG_RV40_DECODER */
+#if CONFIG_DIRAC_DECODER
+#define DIRAC_MC(OPNAME)\
+void ff_ ## OPNAME ## _dirac_pixels8_c(uint8_t *dst, const uint8_t *src[5], int stride, int h)\
+{\
+ OPNAME ## _pixels8_8_c(dst, src[0], stride, h);\
+}\
+void ff_ ## OPNAME ## _dirac_pixels16_c(uint8_t *dst, const uint8_t *src[5], int stride, int h)\
+{\
+ OPNAME ## _pixels16_8_c(dst, src[0], stride, h);\
+}\
+void ff_ ## OPNAME ## _dirac_pixels32_c(uint8_t *dst, const uint8_t *src[5], int stride, int h)\
+{\
+ OPNAME ## _pixels16_8_c(dst , src[0] , stride, h);\
+ OPNAME ## _pixels16_8_c(dst+16, src[0]+16, stride, h);\
+}\
+void ff_ ## OPNAME ## _dirac_pixels8_l2_c(uint8_t *dst, const uint8_t *src[5], int stride, int h)\
+{\
+ OPNAME ## _pixels8_l2_8(dst, src[0], src[1], stride, stride, stride, h);\
+}\
+void ff_ ## OPNAME ## _dirac_pixels16_l2_c(uint8_t *dst, const uint8_t *src[5], int stride, int h)\
+{\
+ OPNAME ## _pixels16_l2_8(dst, src[0], src[1], stride, stride, stride, h);\
+}\
+void ff_ ## OPNAME ## _dirac_pixels32_l2_c(uint8_t *dst, const uint8_t *src[5], int stride, int h)\
+{\
+ OPNAME ## _pixels16_l2_8(dst , src[0] , src[1] , stride, stride, stride, h);\
+ OPNAME ## _pixels16_l2_8(dst+16, src[0]+16, src[1]+16, stride, stride, stride, h);\
+}\
+void ff_ ## OPNAME ## _dirac_pixels8_l4_c(uint8_t *dst, const uint8_t *src[5], int stride, int h)\
+{\
+ OPNAME ## _pixels8_l4_8(dst, src[0], src[1], src[2], src[3], stride, stride, stride, stride, stride, h);\
+}\
+void ff_ ## OPNAME ## _dirac_pixels16_l4_c(uint8_t *dst, const uint8_t *src[5], int stride, int h)\
+{\
+ OPNAME ## _pixels16_l4_8(dst, src[0], src[1], src[2], src[3], stride, stride, stride, stride, stride, h);\
+}\
+void ff_ ## OPNAME ## _dirac_pixels32_l4_c(uint8_t *dst, const uint8_t *src[5], int stride, int h)\
+{\
+ OPNAME ## _pixels16_l4_8(dst , src[0] , src[1] , src[2] , src[3] , stride, stride, stride, stride, stride, h);\
+ OPNAME ## _pixels16_l4_8(dst+16, src[0]+16, src[1]+16, src[2]+16, src[3]+16, stride, stride, stride, stride, stride, h);\
+}
+DIRAC_MC(put)
+DIRAC_MC(avg)
+#endif
+
static void wmv2_mspel8_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int w){
uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
int i;
@@ -1810,7 +1918,7 @@ static void add_bytes_c(uint8_t *dst, uint8_t *src, int w){
dst[i+0] += src[i+0];
}
-static void diff_bytes_c(uint8_t *dst, uint8_t *src1, uint8_t *src2, int w){
+static void diff_bytes_c(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int w){
long i;
#if !HAVE_FAST_UNALIGNED
if((long)src2 & (sizeof(long)-1)){
@@ -2409,7 +2517,7 @@ static void vector_fmac_scalar_c(float *dst, const float *src, float mul,
dst[i] += src[i] * mul;
}
-static void butterflies_float_c(float *restrict v1, float *restrict v2,
+static void butterflies_float_c(float *av_restrict v1, float *av_restrict v2,
int len)
{
int i;
@@ -2632,6 +2740,37 @@ static void ff_jref_idct_add(uint8_t *dest, int line_size, DCTELEM *block)
ff_add_pixels_clamped_c(block, dest, line_size);
}
+static void ff_jref_idct4_put(uint8_t *dest, int line_size, DCTELEM *block)
+{
+ ff_j_rev_dct4 (block);
+ put_pixels_clamped4_c(block, dest, line_size);
+}
+static void ff_jref_idct4_add(uint8_t *dest, int line_size, DCTELEM *block)
+{
+ ff_j_rev_dct4 (block);
+ add_pixels_clamped4_c(block, dest, line_size);
+}
+
+static void ff_jref_idct2_put(uint8_t *dest, int line_size, DCTELEM *block)
+{
+ ff_j_rev_dct2 (block);
+ put_pixels_clamped2_c(block, dest, line_size);
+}
+static void ff_jref_idct2_add(uint8_t *dest, int line_size, DCTELEM *block)
+{
+ ff_j_rev_dct2 (block);
+ add_pixels_clamped2_c(block, dest, line_size);
+}
+
+static void ff_jref_idct1_put(uint8_t *dest, int line_size, DCTELEM *block)
+{
+ dest[0] = av_clip_uint8((block[0] + 4)>>3);
+}
+static void ff_jref_idct1_add(uint8_t *dest, int line_size, DCTELEM *block)
+{
+ dest[0] = av_clip_uint8(dest[0] + ((block[0] + 4)>>3));
+}
+
static void just_return(void *mem av_unused, int stride av_unused, int h av_unused) { return; }
/* init static data */
@@ -2663,7 +2802,7 @@ int ff_check_alignment(void){
"Compiler did not align stack variables. Libavcodec has been miscompiled\n"
"and may be very slow or crash. This is not a bug in libavcodec,\n"
"but in the compiler. You may try recompiling using gcc >= 4.2.\n"
- "Do not report crashes to Libav developers.\n");
+ "Do not report crashes to FFmpeg developers.\n");
#endif
did_fail=1;
}
@@ -2698,12 +2837,28 @@ av_cold void ff_dsputil_init(DSPContext* c, AVCodecContext *avctx)
}
#endif //CONFIG_ENCODERS
- if (avctx->bits_per_raw_sample == 10) {
- c->idct_put = ff_simple_idct_put_10;
- c->idct_add = ff_simple_idct_add_10;
- c->idct = ff_simple_idct_10;
- c->idct_permutation_type = FF_NO_IDCT_PERM;
- } else {
+ if(avctx->lowres==1){
+ c->idct_put= ff_jref_idct4_put;
+ c->idct_add= ff_jref_idct4_add;
+ c->idct = ff_j_rev_dct4;
+ c->idct_permutation_type= FF_NO_IDCT_PERM;
+ }else if(avctx->lowres==2){
+ c->idct_put= ff_jref_idct2_put;
+ c->idct_add= ff_jref_idct2_add;
+ c->idct = ff_j_rev_dct2;
+ c->idct_permutation_type= FF_NO_IDCT_PERM;
+ }else if(avctx->lowres==3){
+ c->idct_put= ff_jref_idct1_put;
+ c->idct_add= ff_jref_idct1_add;
+ c->idct = ff_j_rev_dct1;
+ c->idct_permutation_type= FF_NO_IDCT_PERM;
+ }else{
+ if (avctx->bits_per_raw_sample == 10) {
+ c->idct_put = ff_simple_idct_put_10;
+ c->idct_add = ff_simple_idct_add_10;
+ c->idct = ff_simple_idct_10;
+ c->idct_permutation_type = FF_NO_IDCT_PERM;
+ } else {
if(avctx->idct_algo==FF_IDCT_INT){
c->idct_put= ff_jref_idct_put;
c->idct_add= ff_jref_idct_add;
@@ -2734,6 +2889,7 @@ av_cold void ff_dsputil_init(DSPContext* c, AVCodecContext *avctx)
c->idct = ff_simple_idct_8;
c->idct_permutation_type= FF_NO_IDCT_PERM;
}
+ }
}
c->diff_pixels = diff_pixels_c;
@@ -3029,3 +3185,8 @@ av_cold void ff_dsputil_init(DSPContext* c, AVCodecContext *avctx)
ff_init_scantable_permutation(c->idct_permutation,
c->idct_permutation_type);
}
+
+av_cold void dsputil_init(DSPContext* c, AVCodecContext *avctx)
+{
+ ff_dsputil_init(c, avctx);
+}
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-10 00:33:20 +0000