/*
* Floating point AAN DCT
- * Copyright (c) 2003 Michael Niedermayer <michaelni@gmx.at>
- *
- * This library 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.
+ * this implementation is based upon the IJG integer AAN DCT (see jfdctfst.c)
*
- * This library 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.
+ * Copyright (c) 2003 Michael Niedermayer <michaelni@gmx.at>
+ * Copyright (c) 2003 Roman Shaposhnik
*
- * You should have received a copy of the GNU Lesser General Public
- * License along with this library; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ * Permission to use, copy, modify, and/or distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
*
- * this implementation is based upon the IJG integer AAN DCT (see jfdctfst.c)
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
/**
- * @file faandct.c
- * @brief
+ * @file
+ * @brief
* Floating point AAN DCT
* @author Michael Niedermayer <michaelni@gmx.at>
*/
-#include <math.h>
-
#include "dsputil.h"
#include "faandct.h"
#define A5 0.38268343236508977170 // cos(pi*6/16)
#define A4 1.30656296487637652774 // cos(pi*2/16)sqrt(2)
-static FLOAT postscale[64]={
+static const FLOAT postscale[64]={
B0*B0, B0*B1, B0*B2, B0*B3, B0*B4, B0*B5, B0*B6, B0*B7,
B1*B0, B1*B1, B1*B2, B1*B3, B1*B4, B1*B5, B1*B6, B1*B7,
B2*B0, B2*B1, B2*B2, B2*B3, B2*B4, B2*B5, B2*B6, B2*B7,
B7*B0, B7*B1, B7*B2, B7*B3, B7*B4, B7*B5, B7*B6, B7*B7,
};
-void ff_faandct(DCTELEM * data)
+static av_always_inline void row_fdct(FLOAT temp[64], DCTELEM * data)
{
FLOAT tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7;
FLOAT tmp10, tmp11, tmp12, tmp13;
- FLOAT z1, z2, z3, z4, z5, z11, z13;
- FLOAT temp[64];
+ FLOAT z2, z4, z11, z13;
+ FLOAT av_unused z5;
int i;
- emms_c();
-
for (i=0; i<8*8; i+=8) {
tmp0= data[0 + i] + data[7 + i];
tmp7= data[0 + i] - data[7 + i];
tmp5= data[2 + i] - data[5 + i];
tmp3= data[3 + i] + data[4 + i];
tmp4= data[3 + i] - data[4 + i];
-
+
tmp10= tmp0 + tmp3;
tmp13= tmp0 - tmp3;
tmp11= tmp1 + tmp2;
tmp12= tmp1 - tmp2;
-
+
temp[0 + i]= tmp10 + tmp11;
temp[4 + i]= tmp10 - tmp11;
-
- z1= (tmp12 + tmp13)*A1;
- temp[2 + i]= tmp13 + z1;
- temp[6 + i]= tmp13 - z1;
-
- tmp10= tmp4 + tmp5;
- tmp11= tmp5 + tmp6;
- tmp12= tmp6 + tmp7;
-
- z5= (tmp10 - tmp12) * A5;
- z2= tmp10*A2 + z5;
- z4= tmp12*A4 + z5;
- z3= tmp11*A1;
-
- z11= tmp7 + z3;
- z13= tmp7 - z3;
+
+ tmp12 += tmp13;
+ tmp12 *= A1;
+ temp[2 + i]= tmp13 + tmp12;
+ temp[6 + i]= tmp13 - tmp12;
+
+ tmp4 += tmp5;
+ tmp5 += tmp6;
+ tmp6 += tmp7;
+
+#if 0
+ z5= (tmp4 - tmp6) * A5;
+ z2= tmp4*A2 + z5;
+ z4= tmp6*A4 + z5;
+#else
+ z2= tmp4*(A2+A5) - tmp6*A5;
+ z4= tmp6*(A4-A5) + tmp4*A5;
+#endif
+ tmp5*=A1;
+
+ z11= tmp7 + tmp5;
+ z13= tmp7 - tmp5;
temp[5 + i]= z13 + z2;
temp[3 + i]= z13 - z2;
temp[1 + i]= z11 + z4;
temp[7 + i]= z11 - z4;
}
+}
+
+void ff_faandct(DCTELEM * data)
+{
+ FLOAT tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7;
+ FLOAT tmp10, tmp11, tmp12, tmp13;
+ FLOAT z2, z4, z11, z13;
+ FLOAT av_unused z5;
+ FLOAT temp[64];
+ int i;
+
+ emms_c();
+
+ row_fdct(temp, data);
for (i=0; i<8; i++) {
tmp0= temp[8*0 + i] + temp[8*7 + i];
tmp5= temp[8*2 + i] - temp[8*5 + i];
tmp3= temp[8*3 + i] + temp[8*4 + i];
tmp4= temp[8*3 + i] - temp[8*4 + i];
-
+
tmp10= tmp0 + tmp3;
tmp13= tmp0 - tmp3;
tmp11= tmp1 + tmp2;
tmp12= tmp1 - tmp2;
-
- data[8*0 + i]= lrint(SCALE(8*0 + i) * (tmp10 + tmp11));
- data[8*4 + i]= lrint(SCALE(8*4 + i) * (tmp10 - tmp11));
-
- z1= (tmp12 + tmp13)* A1;
- data[8*2 + i]= lrint(SCALE(8*2 + i) * (tmp13 + z1));
- data[8*6 + i]= lrint(SCALE(8*6 + i) * (tmp13 - z1));
-
- tmp10= tmp4 + tmp5;
- tmp11= tmp5 + tmp6;
- tmp12= tmp6 + tmp7;
-
- z5= (tmp10 - tmp12) * A5;
- z2= tmp10*A2 + z5;
- z4= tmp12*A4 + z5;
- z3= tmp11*A1;
-
- z11= tmp7 + z3;
- z13= tmp7 - z3;
-
- data[8*5 + i]= lrint(SCALE(8*5 + i) * (z13 + z2));
- data[8*3 + i]= lrint(SCALE(8*3 + i) * (z13 - z2));
- data[8*1 + i]= lrint(SCALE(8*1 + i) * (z11 + z4));
- data[8*7 + i]= lrint(SCALE(8*7 + i) * (z11 - z4));
+
+ data[8*0 + i]= lrintf(SCALE(8*0 + i) * (tmp10 + tmp11));
+ data[8*4 + i]= lrintf(SCALE(8*4 + i) * (tmp10 - tmp11));
+
+ tmp12 += tmp13;
+ tmp12 *= A1;
+ data[8*2 + i]= lrintf(SCALE(8*2 + i) * (tmp13 + tmp12));
+ data[8*6 + i]= lrintf(SCALE(8*6 + i) * (tmp13 - tmp12));
+
+ tmp4 += tmp5;
+ tmp5 += tmp6;
+ tmp6 += tmp7;
+
+#if 0
+ z5= (tmp4 - tmp6) * A5;
+ z2= tmp4*A2 + z5;
+ z4= tmp6*A4 + z5;
+#else
+ z2= tmp4*(A2+A5) - tmp6*A5;
+ z4= tmp6*(A4-A5) + tmp4*A5;
+#endif
+ tmp5*=A1;
+
+ z11= tmp7 + tmp5;
+ z13= tmp7 - tmp5;
+
+ data[8*5 + i]= lrintf(SCALE(8*5 + i) * (z13 + z2));
+ data[8*3 + i]= lrintf(SCALE(8*3 + i) * (z13 - z2));
+ data[8*1 + i]= lrintf(SCALE(8*1 + i) * (z11 + z4));
+ data[8*7 + i]= lrintf(SCALE(8*7 + i) * (z11 - z4));
+ }
+}
+
+void ff_faandct248(DCTELEM * data)
+{
+ FLOAT tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7;
+ FLOAT tmp10, tmp11, tmp12, tmp13;
+ FLOAT temp[64];
+ int i;
+
+ emms_c();
+
+ row_fdct(temp, data);
+
+ for (i=0; i<8; i++) {
+ tmp0 = temp[8*0 + i] + temp[8*1 + i];
+ tmp1 = temp[8*2 + i] + temp[8*3 + i];
+ tmp2 = temp[8*4 + i] + temp[8*5 + i];
+ tmp3 = temp[8*6 + i] + temp[8*7 + i];
+ tmp4 = temp[8*0 + i] - temp[8*1 + i];
+ tmp5 = temp[8*2 + i] - temp[8*3 + i];
+ tmp6 = temp[8*4 + i] - temp[8*5 + i];
+ tmp7 = temp[8*6 + i] - temp[8*7 + i];
+
+ tmp10 = tmp0 + tmp3;
+ tmp11 = tmp1 + tmp2;
+ tmp12 = tmp1 - tmp2;
+ tmp13 = tmp0 - tmp3;
+
+ data[8*0 + i] = lrintf(SCALE(8*0 + i) * (tmp10 + tmp11));
+ data[8*4 + i] = lrintf(SCALE(8*4 + i) * (tmp10 - tmp11));
+
+ tmp12 += tmp13;
+ tmp12 *= A1;
+ data[8*2 + i] = lrintf(SCALE(8*2 + i) * (tmp13 + tmp12));
+ data[8*6 + i] = lrintf(SCALE(8*6 + i) * (tmp13 - tmp12));
+
+ tmp10 = tmp4 + tmp7;
+ tmp11 = tmp5 + tmp6;
+ tmp12 = tmp5 - tmp6;
+ tmp13 = tmp4 - tmp7;
+
+ data[8*1 + i] = lrintf(SCALE(8*0 + i) * (tmp10 + tmp11));
+ data[8*5 + i] = lrintf(SCALE(8*4 + i) * (tmp10 - tmp11));
+
+ tmp12 += tmp13;
+ tmp12 *= A1;
+ data[8*3 + i] = lrintf(SCALE(8*2 + i) * (tmp13 + tmp12));
+ data[8*7 + i] = lrintf(SCALE(8*6 + i) * (tmp13 - tmp12));
}
}
projects / ffmpeg / blobdiff