libavcodec/arm/mathops.h

   1 /*
   2  * simple math operations
   3  * Copyright (c) 2006 Michael Niedermayer <michaelni@gmx.at> et al
   4  *
   5  * This file is part of FFmpeg.
   6  *
   7  * FFmpeg is free software; you can redistribute it and/or
   8  * modify it under the terms of the GNU Lesser General Public
   9  * License as published by the Free Software Foundation; either
  10  * version 2.1 of the License, or (at your option) any later version.
  11  *
  12  * FFmpeg is distributed in the hope that it will be useful,
  13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  15  * Lesser General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU Lesser General Public
  18  * License along with FFmpeg; if not, write to the Free Software
  19  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  20  */
  21
  22 #ifndef AVCODEC_ARM_MATHOPS_H
  23 #define AVCODEC_ARM_MATHOPS_H
  24
  25 #include <stdint.h>
  26 #include "libavutil/common.h"
  27
  28 #if HAVE_INLINE_ASM
  29
  30 #   define MULL MULL
  31 static inline av_const int MULL(int a, int b, unsigned shift)
  32 {
  33     int lo, hi;
  34     __asm__("smull %0, %1, %2, %3     \n\t"
  35             "mov   %0, %0,     lsr %4 \n\t"
  36             "add   %1, %0, %1, lsl %5 \n\t"
  37             : "=&r"(lo), "=&r"(hi)
  38             : "r"(b), "r"(a), "ir"(shift), "ir"(32-shift));
  39     return hi;
  40 }
  41
  42 #define MULH MULH
  43 #if HAVE_ARMV6
  44 static inline av_const int MULH(int a, int b)
  45 {
  46     int r;
  47     __asm__ ("smmul %0, %1, %2" : "=r"(r) : "r"(a), "r"(b));
  48     return r;
  49 }
  50 #else
  51 static inline av_const int MULH(int a, int b)
  52 {
  53     int lo, hi;
  54     __asm__ ("smull %0, %1, %2, %3" : "=&r"(lo), "=&r"(hi) : "r"(b), "r"(a));
  55     return hi;
  56 }
  57 #endif
  58
  59 static inline av_const int64_t MUL64(int a, int b)
  60 {
  61     union { uint64_t x; unsigned hl[2]; } x;
  62     __asm__ ("smull %0, %1, %2, %3"
  63              : "=r"(x.hl[0]), "=r"(x.hl[1]) : "r"(a), "r"(b));
  64     return x.x;
  65 }
  66 #define MUL64 MUL64
  67
  68 static inline av_const int64_t MAC64(int64_t d, int a, int b)
  69 {
  70     union { uint64_t x; unsigned hl[2]; } x = { d };
  71     __asm__ ("smlal %0, %1, %2, %3"
  72              : "+r"(x.hl[0]), "+r"(x.hl[1]) : "r"(a), "r"(b));
  73     return x.x;
  74 }
  75 #define MAC64(d, a, b) ((d) = MAC64(d, a, b))
  76 #define MLS64(d, a, b) MAC64(d, -(a), b)
  77
  78 #if HAVE_ARMV5TE
  79
  80 /* signed 16x16 -> 32 multiply add accumulate */
  81 #   define MAC16(rt, ra, rb)                                            \
  82     __asm__ ("smlabb %0, %1, %2, %0" : "+r"(rt) : "r"(ra), "r"(rb));
  83
  84 /* signed 16x16 -> 32 multiply */
  85 #   define MUL16 MUL16
  86 static inline av_const int MUL16(int ra, int rb)
  87 {
  88     int rt;
  89     __asm__ ("smulbb %0, %1, %2" : "=r"(rt) : "r"(ra), "r"(rb));
  90     return rt;
  91 }
  92
  93 #endif
  94
  95 #define mid_pred mid_pred
  96 static inline av_const int mid_pred(int a, int b, int c)
  97 {
  98     int m;
  99     __asm__ volatile (
 100         "mov   %0, %2  \n\t"
 101         "cmp   %1, %2  \n\t"
 102         "movgt %0, %1  \n\t"
 103         "movgt %1, %2  \n\t"
 104         "cmp   %1, %3  \n\t"
 105         "movle %1, %3  \n\t"
 106         "cmp   %0, %1  \n\t"
 107         "movgt %0, %1  \n\t"
 108         : "=&r"(m), "+r"(a)
 109         : "r"(b), "r"(c));
 110     return m;
 111 }
 112
 113 #endif /* HAVE_INLINE_ASM */
 114
 115 #endif /* AVCODEC_ARM_MATHOPS_H */