Move avutil tables only used in libavcodec to libavcodec.

10 files changed, 127 insertions, 11 deletions

diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index 6aa13aff4f..5dbd60e067 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Makefile
@@ -22,6 +22,7 @@ OBJS = allcodecs.o                                                      \
        fmtconvert.o                                                     \
        imgconvert.o                                                     \
        jrevdct.o                                                        \
+       mathtables.o                                                     \
        options.o                                                        \
        parser.o                                                         \
        raw.o                                                            \
@@ -655,12 +656,6 @@ OBJS-$(CONFIG_TEXT2MOVSUB_BSF)            += movsub_bsf.o
 OBJS-$(HAVE_PTHREADS)                  += pthread.o
 OBJS-$(HAVE_W32THREADS)                += pthread.o
 
-# inverse.o contains the ff_inverse table definition, which is used by
-# the FASTDIV macro (from libavutil); since referencing the external
-# table has a negative effect on performance, copy it in libavcodec as
-# well.
-OBJS-$(!CONFIG_SMALL)                  += inverse.o
-
 SKIPHEADERS                            += %_tablegen.h                  \
                                           %_tables.h                    \
                                           aac_tablegen_decl.h           \
diff --git a/libavcodec/arm/mathops.h b/libavcodec/arm/mathops.h
index 3803fcde8c..17a687da6a 100644
--- a/libavcodec/arm/mathops.h
+++ b/libavcodec/arm/mathops.h
@@ -36,6 +36,30 @@ static inline av_const int MULH(int a, int b)
     __asm__ ("smmul %0, %1, %2" : "=r"(r) : "r"(a), "r"(b));
     return r;
 }
+
+#define FASTDIV FASTDIV
+static av_always_inline av_const int FASTDIV(int a, int b)
+{
+    int r;
+    __asm__ ("cmp     %2, #2               \n\t"
+             "ldr     %0, [%3, %2, lsl #2] \n\t"
+             "ite     le                   \n\t"
+             "lsrle   %0, %1, #1           \n\t"
+             "smmulgt %0, %0, %1           \n\t"
+             : "=&r"(r) : "r"(a), "r"(b), "r"(ff_inverse) : "cc");
+    return r;
+}
+
+#else /* HAVE_ARMV6 */
+
+#define FASTDIV FASTDIV
+static av_always_inline av_const int FASTDIV(int a, int b)
+{
+    int r, t;
+    __asm__ ("umull %1, %0, %2, %3"
+             : "=&r"(r), "=&r"(t) : "r"(a), "r"(ff_inverse[b]));
+    return r;
+}
 #endif
 
 #define MLS64(d, a, b) MAC64(d, -(a), b)
diff --git a/libavcodec/inverse.c b/libavcodec/inverse.c
deleted file mode 100644
index 04681d256b..0000000000
--- a/libavcodec/inverse.c
+++ /dev/null
@@ -1 +0,0 @@
-#include "libavutil/inverse.c"
diff --git a/libavcodec/mathops.h b/libavcodec/mathops.h
index ab545ef504..218c459acc 100644
--- a/libavcodec/mathops.h
+++ b/libavcodec/mathops.h
@@ -22,9 +22,14 @@
 #ifndef AVCODEC_MATHOPS_H
 #define AVCODEC_MATHOPS_H
 
+#include <stdint.h>
+
 #include "libavutil/common.h"
 #include "config.h"
 
+extern const uint32_t ff_inverse[257];
+extern const uint8_t ff_sqrt_tab[256];
+
 #if   ARCH_ARM
 #   include "arm/mathops.h"
 #elif ARCH_AVR32
@@ -185,4 +190,28 @@ if ((y) < (x)) {\
 #   define PACK_2S16(a,b)    PACK_2U16((a)&0xffff, (b)&0xffff)
 #endif
 
+#ifndef FASTDIV
+#   define FASTDIV(a,b) ((uint32_t)((((uint64_t)a) * ff_inverse[b]) >> 32))
+#endif /* FASTDIV */
+
+static inline av_const unsigned int ff_sqrt(unsigned int a)
+{
+    unsigned int b;
+
+    if (a < 255) return (ff_sqrt_tab[a + 1] - 1) >> 4;
+    else if (a < (1 << 12)) b = ff_sqrt_tab[a >> 4] >> 2;
+#if !CONFIG_SMALL
+    else if (a < (1 << 14)) b = ff_sqrt_tab[a >> 6] >> 1;
+    else if (a < (1 << 16)) b = ff_sqrt_tab[a >> 8]   ;
+#endif
+    else {
+        int s = av_log2_16bit(a >> 16) >> 1;
+        unsigned int c = a >> (s + 2);
+        b = ff_sqrt_tab[c >> (s + 8)];
+        b = FASTDIV(c,b) + (b << s);
+    }
+
+    return b - (a < b * b);
+}
+
 #endif /* AVCODEC_MATHOPS_H */
diff --git a/libavcodec/mathtables.c b/libavcodec/mathtables.c
new file mode 100644
index 0000000000..45d06ebcdb
--- /dev/null
+++ b/libavcodec/mathtables.c
@@ -0,0 +1,68 @@
+/*
+ * This file is part of Libav.
+ *
+ * Libav 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,
+ * 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
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include <stdint.h>
+
+/* a*inverse[b]>>32 == a/b for all 0<=a<=16909558 && 2<=b<=256
+ * for a>16909558, is an overestimate by less than 1 part in 1<<24 */
+const uint32_t ff_inverse[257]={
+         0, 4294967295U,2147483648U,1431655766, 1073741824,  858993460,  715827883,  613566757,
+ 536870912,  477218589,  429496730,  390451573,  357913942,  330382100,  306783379,  286331154,
+ 268435456,  252645136,  238609295,  226050911,  214748365,  204522253,  195225787,  186737709,
+ 178956971,  171798692,  165191050,  159072863,  153391690,  148102321,  143165577,  138547333,
+ 134217728,  130150525,  126322568,  122713352,  119304648,  116080198,  113025456,  110127367,
+ 107374183,  104755300,  102261127,   99882961,   97612894,   95443718,   93368855,   91382283,
+  89478486,   87652394,   85899346,   84215046,   82595525,   81037119,   79536432,   78090315,
+  76695845,   75350304,   74051161,   72796056,   71582789,   70409300,   69273667,   68174085,
+  67108864,   66076420,   65075263,   64103990,   63161284,   62245903,   61356676,   60492498,
+  59652324,   58835169,   58040099,   57266231,   56512728,   55778797,   55063684,   54366675,
+  53687092,   53024288,   52377650,   51746594,   51130564,   50529028,   49941481,   49367441,
+  48806447,   48258060,   47721859,   47197443,   46684428,   46182445,   45691142,   45210183,
+  44739243,   44278014,   43826197,   43383509,   42949673,   42524429,   42107523,   41698712,
+  41297763,   40904451,   40518560,   40139882,   39768216,   39403370,   39045158,   38693400,
+  38347923,   38008561,   37675152,   37347542,   37025581,   36709123,   36398028,   36092163,
+  35791395,   35495598,   35204650,   34918434,   34636834,   34359739,   34087043,   33818641,
+  33554432,   33294321,   33038210,   32786010,   32537632,   32292988,   32051995,   31814573,
+  31580642,   31350127,   31122952,   30899046,   30678338,   30460761,   30246249,   30034737,
+  29826162,   29620465,   29417585,   29217465,   29020050,   28825284,   28633116,   28443493,
+  28256364,   28071682,   27889399,   27709467,   27531842,   27356480,   27183338,   27012373,
+  26843546,   26676816,   26512144,   26349493,   26188825,   26030105,   25873297,   25718368,
+  25565282,   25414008,   25264514,   25116768,   24970741,   24826401,   24683721,   24542671,
+  24403224,   24265352,   24129030,   23994231,   23860930,   23729102,   23598722,   23469767,
+  23342214,   23216040,   23091223,   22967740,   22845571,   22724695,   22605092,   22486740,
+  22369622,   22253717,   22139007,   22025474,   21913099,   21801865,   21691755,   21582751,
+  21474837,   21367997,   21262215,   21157475,   21053762,   20951060,   20849356,   20748635,
+  20648882,   20550083,   20452226,   20355296,   20259280,   20164166,   20069941,   19976593,
+  19884108,   19792477,   19701685,   19611723,   19522579,   19434242,   19346700,   19259944,
+  19173962,   19088744,   19004281,   18920561,   18837576,   18755316,   18673771,   18592933,
+  18512791,   18433337,   18354562,   18276457,   18199014,   18122225,   18046082,   17970575,
+  17895698,   17821442,   17747799,   17674763,   17602325,   17530479,   17459217,   17388532,
+  17318417,   17248865,   17179870,   17111424,   17043522,   16976156,   16909321,   16843010,
+  16777216
+};
+
+const uint8_t ff_sqrt_tab[256]={
+  0, 16, 23, 28, 32, 36, 40, 43, 46, 48, 51, 54, 56, 58, 60, 62, 64, 66, 68, 70, 72, 74, 76, 77, 79, 80, 82, 84, 85, 87, 88, 90,
+ 91, 92, 94, 95, 96, 98, 99,100,102,103,104,105,107,108,109,110,111,112,114,115,116,117,118,119,120,121,122,123,124,125,126,127,
+128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,144,145,146,147,148,149,150,151,151,152,153,154,155,156,156,
+157,158,159,160,160,161,162,163,164,164,165,166,167,168,168,169,170,171,171,172,173,174,174,175,176,176,177,178,179,179,180,181,
+182,182,183,184,184,185,186,186,187,188,188,189,190,190,191,192,192,193,194,194,195,196,196,197,198,198,199,200,200,201,202,202,
+203,204,204,205,205,206,207,207,208,208,209,210,210,211,212,212,213,213,214,215,215,216,216,217,218,218,219,219,220,220,221,222,
+222,223,223,224,224,225,226,226,227,227,228,228,229,230,230,231,231,232,232,233,233,234,235,235,236,236,237,237,238,238,239,239,
+240,240,241,242,242,243,243,244,244,245,245,246,246,247,247,248,248,249,249,250,250,251,251,252,252,253,253,254,254,255,255,255
+};
diff --git a/libavcodec/motion_est.c b/libavcodec/motion_est.c
index 932406d35b..3577ada50c 100644
--- a/libavcodec/motion_est.c
+++ b/libavcodec/motion_est.c
@@ -30,7 +30,7 @@
 #include <stdlib.h>
 #include <stdio.h>
 #include <limits.h>
-#include "libavutil/intmath.h"
+
 #include "avcodec.h"
 #include "dsputil.h"
 #include "mathops.h"
diff --git a/libavcodec/mpegvideo.c b/libavcodec/mpegvideo.c
index f739c2dc5a..571d9af489 100644
--- a/libavcodec/mpegvideo.c
+++ b/libavcodec/mpegvideo.c
@@ -27,11 +27,11 @@
  * The simplest mpeg encoder (well, it was the simplest!).
  */
 
-#include "libavutil/intmath.h"
 #include "libavutil/imgutils.h"
 #include "avcodec.h"
 #include "dsputil.h"
 #include "internal.h"
+#include "mathops.h"
 #include "mpegvideo.h"
 #include "mjpegenc.h"
 #include "msmpeg4.h"
diff --git a/libavcodec/mpegvideo_enc.c b/libavcodec/mpegvideo_enc.c
index 59476a023f..d2719fe8ee 100644
--- a/libavcodec/mpegvideo_enc.c
+++ b/libavcodec/mpegvideo_enc.c
@@ -34,6 +34,7 @@
 #include "dsputil.h"
 #include "mpegvideo.h"
 #include "h263.h"
+#include "mathops.h"
 #include "mjpegenc.h"
 #include "msmpeg4.h"
 #include "faandct.h"
diff --git a/libavcodec/ra144.c b/libavcodec/ra144.c
index 3c76a0ab7c..3705610fbb 100644
--- a/libavcodec/ra144.c
+++ b/libavcodec/ra144.c
@@ -22,8 +22,8 @@
 #include <stdint.h>
 #include "avcodec.h"
 #include "celp_filters.h"
+#include "mathops.h"
 #include "ra144.h"
-#include "libavutil/common.h"
 
 const int16_t ff_gain_val_tab[256][3] = {
     { 541, 956,  768}, { 877, 581,  568}, { 675,1574,  635}, {1248,1464,  668},
diff --git a/libavcodec/roqaudioenc.c b/libavcodec/roqaudioenc.c
index 8ee6002373..3cc9931a33 100644
--- a/libavcodec/roqaudioenc.c
+++ b/libavcodec/roqaudioenc.c
@@ -21,10 +21,10 @@
  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  */
 
-#include "libavutil/intmath.h"
 #include "avcodec.h"
 #include "bytestream.h"
 #include "internal.h"
+#include "mathops.h"
 
 #define ROQ_FRAME_SIZE           735
 #define ROQ_HEADER_SIZE   8