virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-23 12:43:46 +02:00
Code Issues Releases Activity

replacing MULH by asm for x86

Browse Source 
about 30% faster imdct36()

Originally committed as revision 6048 to svn://svn.ffmpeg.org/ffmpeg/trunk
This commit is contained in:
Michael Niedermayer 2006-08-22 11:51:09 +00:00
parent 5ab1972b53
commit 47a0cd7408
1 changed files with 15 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
libavcodec/mpegaudiodec.c
View File

@ -43,18 +43,26 @@
#define FRAC_ONE (1 << FRAC_BITS)
#define MULL(a,b) (((int64_t)(a) * (int64_t)(b)) >> FRAC_BITS)
#define MUL64(a,b) ((int64_t)(a) * (int64_t)(b))
#ifdef ARCH_X86
# define MULL(a,b) (((int64_t)(a) * (int64_t)(b)) >> FRAC_BITS)
# define MUL64(ra, rb) \
({ int64_t rt; asm ("imull %2\n\t" : "=A"(rt) : "a" (ra), "g" (rb)); rt; })
# define MULH(ra, rb) \
({ int rt, dummy; asm ("imull %3\n\t" : "=d"(rt), "=a"(dummy): "a" (ra), "rm" (rb)); rt; })
#else
# define MULL(a,b) (((int64_t)(a) * (int64_t)(b)) >> FRAC_BITS)
# define MUL64(a,b) ((int64_t)(a) * (int64_t)(b))
//#define MULH(a,b) (((int64_t)(a) * (int64_t)(b))>>32) //gcc 3.4 creates an incredibly bloated mess out of this
static always_inline int MULH(int a, int b){
return ((int64_t)(a) * (int64_t)(b))>>32;
}
#endif
#define FIX(a) ((int)((a) * FRAC_ONE))
/* WARNING: only correct for posititive numbers */
#define FIXR(a) ((int)((a) * FRAC_ONE + 0.5))
#define FRAC_RND(a) (((a) + (FRAC_ONE/2)) >> FRAC_BITS)
#define FIXHR(a) ((int)((a) * (1LL<<32) + 0.5))
//#define MULH(a,b) (((int64_t)(a) * (int64_t)(b))>>32) //gcc 3.4 creates an incredibly bloated mess out of this
static always_inline int MULH(int a, int b){
return ((int64_t)(a) * (int64_t)(b))>>32;
}
/****************/
@ -779,13 +787,7 @@ static inline int round_sample(int64_t *sum)
return sum1;
}
# ifdef ARCH_X86
/* ask gcc devels why this is 3 times faster then the generic code below */
# define MULS(ra, rb) \
({ int64_t rt; asm ("imull %2\n\t" : "=A"(rt) : "a" (ra), "g" (rb)); rt; })
# else
# define MULS(ra, rb) MUL64(ra, rb)
# endif
# define MULS(ra, rb) MUL64(ra, rb)
#endif
#define SUM8(sum, op, w, p) \