view liba52/resample_mmx.c @ 23572:a00685941686

demux_mkv very long seek fix The seek code searching for the closest position in the index used "int64_t min_diff=0xFFFFFFFL" as the initial "further from the goal than any real alternative" value. The unit is milliseconds so seeks more than about 75 hours past the end of the file would fail to recognize the last index position as the best match. This was triggered in practice by chapter seek code which apparently uses a seek of 1000000000 seconds forward to mean "seek to the end". The practical effect was that trying to seek to the next chapter in a file without chapters made MPlayer block until it finished reading the file from the current position to the end. Fixed by increasing the initial value from FFFFFFF to FFFFFFFFFFFFFFF.
author uau
date Wed, 20 Jun 2007 18:19:03 +0000
parents d6219ce521e9
children a180e69cc18c
line wrap: on
line source


// MMX optimizations from Michael Niedermayer (michaelni@gmx.at) (under GPL)

/* optimization TODO / NOTES 
    movntq is slightly faster (0.5% with the current test.c benchmark) 
	(but thats just test.c so that needs to be testd in reallity)
	and it would mean (C / MMX2 / MMX / 3DNOW) versions 
*/

#include "a52_internal.h"


static uint64_t attribute_used __attribute__((aligned(8))) magicF2W= 0x43c0000043c00000LL;
static uint64_t attribute_used __attribute__((aligned(8))) wm1010= 0xFFFF0000FFFF0000LL;
static uint64_t attribute_used __attribute__((aligned(8))) wm0101= 0x0000FFFF0000FFFFLL;
static uint64_t attribute_used __attribute__((aligned(8))) wm1100= 0xFFFFFFFF00000000LL;

static int a52_resample_MONO_to_5_MMX(float * _f, int16_t * s16){
    int32_t * f = (int32_t *) _f;
	asm volatile(
		"mov $-512, %%"REG_S"		\n\t"
		"movq "MANGLE(magicF2W)", %%mm7	\n\t"
		"movq "MANGLE(wm1100)", %%mm3	\n\t"
		"movq "MANGLE(wm0101)", %%mm4	\n\t"
		"movq "MANGLE(wm1010)", %%mm5	\n\t"
		"pxor %%mm6, %%mm6		\n\t"
		"1:				\n\t"
		"movq (%1, %%"REG_S", 2), %%mm0	\n\t"
		"movq 8(%1, %%"REG_S", 2), %%mm1\n\t"
		"lea (%%"REG_S", %%"REG_S", 4), %%"REG_D"\n\t"
		"psubd %%mm7, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"packssdw %%mm1, %%mm0		\n\t"
		"movq %%mm0, %%mm1		\n\t"
		"pand %%mm4, %%mm0		\n\t"
		"pand %%mm5, %%mm1		\n\t"
		"movq %%mm6, (%0, %%"REG_D")	\n\t" // 0 0 0 0
		"movd %%mm0, 8(%0, %%"REG_D")	\n\t" // A 0
		"pand %%mm3, %%mm0		\n\t"
		"movd %%mm6, 12(%0, %%"REG_D")	\n\t" // 0 0
		"movd %%mm1, 16(%0, %%"REG_D")	\n\t" // 0 B
		"pand %%mm3, %%mm1		\n\t"
		"movd %%mm6, 20(%0, %%"REG_D")	\n\t" // 0 0
		"movq %%mm0, 24(%0, %%"REG_D")	\n\t" // 0 0 C 0
		"movq %%mm1, 32(%0, %%"REG_D")	\n\t" // 0 0 0 B
		"add $8, %%"REG_S"		\n\t"
		" jnz 1b			\n\t"
		"emms				\n\t"
		:: "r" (s16+1280), "r" (f+256)
		:"%"REG_S, "%"REG_D, "memory"
	);
    return 5*256;
}

static int a52_resample_STEREO_to_2_MMX(float * _f, int16_t * s16){
    int32_t * f = (int32_t *) _f;
/* benchmark scores are 0.3% better with SSE but we would need to set bias=0 and premultiply it
#ifdef HAVE_SSE
	asm volatile(
		"mov $-1024, %%"REG_S"		\n\t"
		"1:				\n\t"
		"cvtps2pi (%1, %%"REG_S"), %%mm0\n\t"
		"cvtps2pi 1024(%1, %%"REG_S"), %%mm2\n\t"
		"movq %%mm0, %%mm1		\n\t"
		"punpcklwd %%mm2, %%mm0		\n\t"
		"punpckhwd %%mm2, %%mm1		\n\t"
		"movq %%mm0, (%0, %%"REG_S")	\n\t"
		"movq %%mm1, 8(%0, %%"REG_S")	\n\t"
		"add $16, %%"REG_S"		\n\t"
		" jnz 1b			\n\t"
		"emms				\n\t"
		:: "r" (s16+512), "r" (f+256)
		:"%"REG_S, "memory"
	);*/
	asm volatile(
		"mov $-1024, %%"REG_S"		\n\t"
		"movq "MANGLE(magicF2W)", %%mm7	\n\t"
		"1:				\n\t"
		"movq (%1, %%"REG_S"), %%mm0	\n\t"
		"movq 8(%1, %%"REG_S"), %%mm1	\n\t"
		"movq 1024(%1, %%"REG_S"), %%mm2\n\t"
		"movq 1032(%1, %%"REG_S"), %%mm3\n\t"
		"psubd %%mm7, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"psubd %%mm7, %%mm2		\n\t"
		"psubd %%mm7, %%mm3		\n\t"
		"packssdw %%mm1, %%mm0		\n\t"
		"packssdw %%mm3, %%mm2		\n\t"
		"movq %%mm0, %%mm1		\n\t"
		"punpcklwd %%mm2, %%mm0		\n\t"
		"punpckhwd %%mm2, %%mm1		\n\t"
		"movq %%mm0, (%0, %%"REG_S")	\n\t"
		"movq %%mm1, 8(%0, %%"REG_S")	\n\t"
		"add $16, %%"REG_S"		\n\t"
		" jnz 1b			\n\t"
		"emms				\n\t"
		:: "r" (s16+512), "r" (f+256)
		:"%"REG_S, "memory"
	);
    return 2*256;
}

static int a52_resample_3F_to_5_MMX(float * _f, int16_t * s16){
    int32_t * f = (int32_t *) _f;
	asm volatile(
		"mov $-1024, %%"REG_S"		\n\t"
		"movq "MANGLE(magicF2W)", %%mm7	\n\t"
		"pxor %%mm6, %%mm6		\n\t"
		"movq %%mm7, %%mm5		\n\t"
		"punpckldq %%mm6, %%mm5		\n\t"
		"1:				\n\t"
		"movd (%1, %%"REG_S"), %%mm0	\n\t"
		"punpckldq 2048(%1, %%"REG_S"), %%mm0\n\t"
		"movd 1024(%1, %%"REG_S"), %%mm1\n\t"
		"punpckldq 4(%1, %%"REG_S"), %%mm1\n\t"
		"movd 2052(%1, %%"REG_S"), %%mm2\n\t"
		"movq %%mm7, %%mm3		\n\t"
		"punpckldq 1028(%1, %%"REG_S"), %%mm3\n\t"
		"movd 8(%1, %%"REG_S"), %%mm4	\n\t"
		"punpckldq 2056(%1, %%"REG_S"), %%mm4\n\t"
		"lea (%%"REG_S", %%"REG_S", 4), %%"REG_D"\n\t"
		"sar $1, %%"REG_D"		\n\t"
		"psubd %%mm7, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"psubd %%mm5, %%mm2		\n\t"
		"psubd %%mm7, %%mm3		\n\t"
		"psubd %%mm7, %%mm4		\n\t"
		"packssdw %%mm6, %%mm0		\n\t"
		"packssdw %%mm2, %%mm1		\n\t"
		"packssdw %%mm4, %%mm3		\n\t"
		"movq %%mm0, (%0, %%"REG_D")	\n\t"
		"movq %%mm1, 8(%0, %%"REG_D")	\n\t"
		"movq %%mm3, 16(%0, %%"REG_D")	\n\t"
		"movd 1032(%1, %%"REG_S"), %%mm1\n\t"
		"punpckldq 12(%1, %%"REG_S"), %%mm1\n\t"
		"movd 2060(%1, %%"REG_S"), %%mm2\n\t"
		"movq %%mm7, %%mm3		\n\t"
		"punpckldq 1036(%1, %%"REG_S"), %%mm3\n\t"
		"pxor %%mm0, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"psubd %%mm5, %%mm2		\n\t"
		"psubd %%mm7, %%mm3		\n\t"
		"packssdw %%mm1, %%mm0		\n\t"
		"packssdw %%mm3, %%mm2		\n\t"
		"movq %%mm0, 24(%0, %%"REG_D")	\n\t"
		"movq %%mm2, 32(%0, %%"REG_D")	\n\t"
				
		"add $16, %%"REG_S"		\n\t"
		" jnz 1b			\n\t"
		"emms				\n\t"
		:: "r" (s16+1280), "r" (f+256)
		:"%"REG_S, "%"REG_D, "memory"
	);
    return 5*256;
}

static int a52_resample_2F_2R_to_4_MMX(float * _f, int16_t * s16){
    int32_t * f = (int32_t *) _f;
	asm volatile(
		"mov $-1024, %%"REG_S"		\n\t"
		"movq "MANGLE(magicF2W)", %%mm7	\n\t"
		"1:				\n\t"
		"movq (%1, %%"REG_S"), %%mm0	\n\t"
		"movq 8(%1, %%"REG_S"), %%mm1	\n\t"
		"movq 1024(%1, %%"REG_S"), %%mm2\n\t"
		"movq 1032(%1, %%"REG_S"), %%mm3\n\t"
		"psubd %%mm7, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"psubd %%mm7, %%mm2		\n\t"
		"psubd %%mm7, %%mm3		\n\t"
		"packssdw %%mm1, %%mm0		\n\t"
		"packssdw %%mm3, %%mm2		\n\t"
		"movq 2048(%1, %%"REG_S"), %%mm3\n\t"
		"movq 2056(%1, %%"REG_S"), %%mm4\n\t"
		"movq 3072(%1, %%"REG_S"), %%mm5\n\t"
		"movq 3080(%1, %%"REG_S"), %%mm6\n\t"
		"psubd %%mm7, %%mm3		\n\t"
		"psubd %%mm7, %%mm4		\n\t"
		"psubd %%mm7, %%mm5		\n\t"
		"psubd %%mm7, %%mm6		\n\t"
		"packssdw %%mm4, %%mm3		\n\t"
		"packssdw %%mm6, %%mm5		\n\t"
		"movq %%mm0, %%mm1		\n\t"
		"movq %%mm3, %%mm4		\n\t"
		"punpcklwd %%mm2, %%mm0		\n\t"
		"punpckhwd %%mm2, %%mm1		\n\t"
		"punpcklwd %%mm5, %%mm3		\n\t"
		"punpckhwd %%mm5, %%mm4		\n\t"
		"movq %%mm0, %%mm2		\n\t"
		"movq %%mm1, %%mm5		\n\t"
		"punpckldq %%mm3, %%mm0		\n\t"
		"punpckhdq %%mm3, %%mm2		\n\t"
		"punpckldq %%mm4, %%mm1		\n\t"
		"punpckhdq %%mm4, %%mm5		\n\t"
		"movq %%mm0, (%0, %%"REG_S",2)	\n\t"
		"movq %%mm2, 8(%0, %%"REG_S",2)	\n\t"
		"movq %%mm1, 16(%0, %%"REG_S",2)\n\t"
		"movq %%mm5, 24(%0, %%"REG_S",2)\n\t"
		"add $16, %%"REG_S"		\n\t"
		" jnz 1b			\n\t"
		"emms				\n\t"
		:: "r" (s16+1024), "r" (f+256)
		:"%"REG_S, "memory"
	);
    return 4*256;
}

static int a52_resample_3F_2R_to_5_MMX(float * _f, int16_t * s16){
    int32_t * f = (int32_t *) _f;
	asm volatile(
		"mov $-1024, %%"REG_S"		\n\t"
		"movq "MANGLE(magicF2W)", %%mm7	\n\t"
		"1:				\n\t"
		"movd (%1, %%"REG_S"), %%mm0	\n\t"
		"punpckldq 2048(%1, %%"REG_S"), %%mm0\n\t"
		"movd 3072(%1, %%"REG_S"), %%mm1\n\t"
		"punpckldq 4096(%1, %%"REG_S"), %%mm1\n\t"
		"movd 1024(%1, %%"REG_S"), %%mm2\n\t"
		"punpckldq 4(%1, %%"REG_S"), %%mm2\n\t"
		"movd 2052(%1, %%"REG_S"), %%mm3\n\t"
		"punpckldq 3076(%1, %%"REG_S"), %%mm3\n\t"
		"movd 4100(%1, %%"REG_S"), %%mm4\n\t"
		"punpckldq 1028(%1, %%"REG_S"), %%mm4\n\t"
		"movd 8(%1, %%"REG_S"), %%mm5	\n\t"
		"punpckldq 2056(%1, %%"REG_S"), %%mm5\n\t"
		"lea (%%"REG_S", %%"REG_S", 4), %%"REG_D"\n\t"
		"sar $1, %%"REG_D"		\n\t"
		"psubd %%mm7, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"psubd %%mm7, %%mm2		\n\t"
		"psubd %%mm7, %%mm3		\n\t"
		"psubd %%mm7, %%mm4		\n\t"
		"psubd %%mm7, %%mm5		\n\t"
		"packssdw %%mm1, %%mm0		\n\t"
		"packssdw %%mm3, %%mm2		\n\t"
		"packssdw %%mm5, %%mm4		\n\t"
		"movq %%mm0, (%0, %%"REG_D")	\n\t"
		"movq %%mm2, 8(%0, %%"REG_D")	\n\t"
		"movq %%mm4, 16(%0, %%"REG_D")	\n\t"
		
		"movd 3080(%1, %%"REG_S"), %%mm0\n\t"
		"punpckldq 4104(%1, %%"REG_S"), %%mm0\n\t"
		"movd 1032(%1, %%"REG_S"), %%mm1\n\t"
		"punpckldq 12(%1, %%"REG_S"), %%mm1\n\t"
		"movd 2060(%1, %%"REG_S"), %%mm2\n\t"
		"punpckldq 3084(%1, %%"REG_S"), %%mm2\n\t"
		"movd 4108(%1, %%"REG_S"), %%mm3\n\t"
		"punpckldq 1036(%1, %%"REG_S"), %%mm3\n\t"
		"psubd %%mm7, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"psubd %%mm7, %%mm2		\n\t"
		"psubd %%mm7, %%mm3		\n\t"
		"packssdw %%mm1, %%mm0		\n\t"
		"packssdw %%mm3, %%mm2		\n\t"
		"movq %%mm0, 24(%0, %%"REG_D")	\n\t"
		"movq %%mm2, 32(%0, %%"REG_D")	\n\t"
				
		"add $16, %%"REG_S"		\n\t"
		" jnz 1b			\n\t"
		"emms				\n\t"
		:: "r" (s16+1280), "r" (f+256)
		:"%"REG_S, "%"REG_D, "memory"
	);
    return 5*256;
}

static int a52_resample_MONO_LFE_to_6_MMX(float * _f, int16_t * s16){
    int32_t * f = (int32_t *) _f;
	asm volatile(
		"mov $-1024, %%"REG_S"		\n\t"
		"movq "MANGLE(magicF2W)", %%mm7	\n\t"
		"pxor %%mm6, %%mm6		\n\t"
		"1:				\n\t"
		"movq 1024(%1, %%"REG_S"), %%mm0\n\t"
		"movq 1032(%1, %%"REG_S"), %%mm1\n\t"
		"movq (%1, %%"REG_S"), %%mm2	\n\t"
		"movq 8(%1, %%"REG_S"), %%mm3	\n\t"
		"psubd %%mm7, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"psubd %%mm7, %%mm2		\n\t"
		"psubd %%mm7, %%mm3		\n\t"
		"packssdw %%mm1, %%mm0		\n\t"
		"packssdw %%mm3, %%mm2		\n\t"
		"movq %%mm0, %%mm1		\n\t"
		"punpcklwd %%mm2, %%mm0		\n\t"
		"punpckhwd %%mm2, %%mm1		\n\t"
		"lea (%%"REG_S", %%"REG_S", 2), %%"REG_D"\n\t"
		"movq %%mm6, (%0, %%"REG_D")	\n\t"
		"movd %%mm0, 8(%0, %%"REG_D")	\n\t"
		"punpckhdq %%mm0, %%mm0		\n\t"
		"movq %%mm6, 12(%0, %%"REG_D")	\n\t"
		"movd %%mm0, 20(%0, %%"REG_D")	\n\t"
		"movq %%mm6, 24(%0, %%"REG_D")	\n\t"
		"movd %%mm1, 32(%0, %%"REG_D")	\n\t"
		"punpckhdq %%mm1, %%mm1		\n\t"
		"movq %%mm6, 36(%0, %%"REG_D")	\n\t"
		"movd %%mm1, 44(%0, %%"REG_D")	\n\t"
		"add $16, %%"REG_S"		\n\t"
		" jnz 1b			\n\t"
		"emms				\n\t"
		:: "r" (s16+1536), "r" (f+256)
		:"%"REG_S, "%"REG_D, "memory"
	);
    return 6*256;
}

static int a52_resample_STEREO_LFE_to_6_MMX(float * _f, int16_t * s16){
    int32_t * f = (int32_t *) _f;
	asm volatile(
		"mov $-1024, %%"REG_S"		\n\t"
		"movq "MANGLE(magicF2W)", %%mm7	\n\t"
		"pxor %%mm6, %%mm6		\n\t"
		"1:				\n\t"
		"movq 1024(%1, %%"REG_S"), %%mm0\n\t"
		"movq 2048(%1, %%"REG_S"), %%mm1\n\t"
		"movq (%1, %%"REG_S"), %%mm5	\n\t" 
		"psubd %%mm7, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"psubd %%mm7, %%mm5		\n\t"
		"lea (%%"REG_S", %%"REG_S", 2), %%"REG_D"\n\t"
		
		"pxor %%mm4, %%mm4		\n\t"
		"packssdw %%mm5, %%mm0		\n\t" // FfAa
		"packssdw %%mm4, %%mm1		\n\t" // 00Bb
		"punpckhwd %%mm0, %%mm4		\n\t" // F0f0
		"punpcklwd %%mm1, %%mm0		\n\t" // BAba
		"movq %%mm0, %%mm1		\n\t" // BAba
		"punpckldq %%mm4, %%mm3		\n\t" // f0XX
		"punpckldq %%mm6, %%mm0		\n\t" // 00ba
		"punpckhdq %%mm1, %%mm3		\n\t" // BAf0
		
		"movq %%mm0, (%0, %%"REG_D")	\n\t" // 00ba
		"punpckhdq %%mm4, %%mm0		\n\t" // F000
		"movq %%mm3, 8(%0, %%"REG_D")	\n\t" // BAf0
		"movq %%mm0, 16(%0, %%"REG_D")	\n\t" // F000
		"add $8, %%"REG_S"		\n\t"
		" jnz 1b			\n\t"
		"emms				\n\t"
		:: "r" (s16+1536), "r" (f+256)
		:"%"REG_S, "%"REG_D, "memory"
	);
    return 6*256;
}

static int a52_resample_3F_LFE_to_6_MMX(float * _f, int16_t * s16){
    int32_t * f = (int32_t *) _f;
	asm volatile(
		"mov $-1024, %%"REG_S"		\n\t"
		"movq "MANGLE(magicF2W)", %%mm7	\n\t"
		"pxor %%mm6, %%mm6		\n\t"
		"1:				\n\t"
		"movq 1024(%1, %%"REG_S"), %%mm0\n\t"
		"movq 3072(%1, %%"REG_S"), %%mm1\n\t"
		"movq 2048(%1, %%"REG_S"), %%mm4\n\t"
		"movq (%1, %%"REG_S"), %%mm5	\n\t" 
		"psubd %%mm7, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"psubd %%mm7, %%mm4		\n\t"
		"psubd %%mm7, %%mm5		\n\t"
		"lea (%%"REG_S", %%"REG_S", 2), %%"REG_D"\n\t"
		
		"packssdw %%mm4, %%mm0		\n\t" // EeAa
		"packssdw %%mm5, %%mm1		\n\t" // FfBb
		"movq %%mm0, %%mm2		\n\t" // EeAa
		"punpcklwd %%mm1, %%mm0		\n\t" // BAba
		"punpckhwd %%mm1, %%mm2		\n\t" // FEfe
		"movq %%mm0, %%mm1		\n\t" // BAba
		"punpckldq %%mm6, %%mm0		\n\t" // 00ba
		"punpckhdq %%mm1, %%mm1		\n\t" // BABA
		
		"movq %%mm0, (%0, %%"REG_D")	\n\t"
		"punpckhdq %%mm2, %%mm0		\n\t" // FE00
		"punpckldq %%mm1, %%mm2		\n\t" // BAfe
		"movq %%mm2, 8(%0, %%"REG_D")	\n\t"
		"movq %%mm0, 16(%0, %%"REG_D")	\n\t"
		"add $8, %%"REG_S"		\n\t"
		" jnz 1b			\n\t"
		"emms				\n\t"
		:: "r" (s16+1536), "r" (f+256)
		:"%"REG_S, "%"REG_D, "memory"
	);
    return 6*256;
}

static int a52_resample_2F_2R_LFE_to_6_MMX(float * _f, int16_t * s16){
    int32_t * f = (int32_t *) _f;
	asm volatile(
		"mov $-1024, %%"REG_S"		\n\t"
		"movq "MANGLE(magicF2W)", %%mm7	\n\t"
//		"pxor %%mm6, %%mm6		\n\t"
		"1:				\n\t"
		"movq 1024(%1, %%"REG_S"), %%mm0\n\t"
		"movq 2048(%1, %%"REG_S"), %%mm1\n\t"
		"movq 3072(%1, %%"REG_S"), %%mm2\n\t"
		"movq 4096(%1, %%"REG_S"), %%mm3\n\t"
		"movq (%1, %%"REG_S"), %%mm5	\n\t" 
		"psubd %%mm7, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"psubd %%mm7, %%mm2		\n\t"
		"psubd %%mm7, %%mm3		\n\t"
		"psubd %%mm7, %%mm5		\n\t"
		"lea (%%"REG_S", %%"REG_S", 2), %%"REG_D"\n\t"
		
		"packssdw %%mm2, %%mm0		\n\t" // CcAa
		"packssdw %%mm3, %%mm1		\n\t" // DdBb
		"packssdw %%mm5, %%mm5		\n\t" // FfFf
		"movq %%mm0, %%mm2		\n\t" // CcAa
		"punpcklwd %%mm1, %%mm0		\n\t" // BAba
		"punpckhwd %%mm1, %%mm2		\n\t" // DCdc
		"pxor %%mm4, %%mm4		\n\t" // 0000
		"punpcklwd %%mm5, %%mm4		\n\t" // F0f0
		"movq %%mm0, %%mm1		\n\t" // BAba
		"movq %%mm4, %%mm3		\n\t" // F0f0
		"punpckldq %%mm2, %%mm0		\n\t" // dcba
		"punpckhdq %%mm1, %%mm1		\n\t" // BABA
		"punpckldq %%mm1, %%mm4		\n\t" // BAf0
		"punpckhdq %%mm3, %%mm2		\n\t" // F0DC
		
		"movq %%mm0, (%0, %%"REG_D")	\n\t"
		"movq %%mm4, 8(%0, %%"REG_D")	\n\t"
		"movq %%mm2, 16(%0, %%"REG_D")	\n\t"
		"add $8, %%"REG_S"		\n\t"
		" jnz 1b			\n\t"
		"emms				\n\t"
		:: "r" (s16+1536), "r" (f+256)
		:"%"REG_S, "%"REG_D, "memory"
	);
    return 6*256;
}

static int a52_resample_3F_2R_LFE_to_6_MMX(float * _f, int16_t * s16){
    int32_t * f = (int32_t *) _f;
	asm volatile(
		"mov $-1024, %%"REG_S"		\n\t"
		"movq "MANGLE(magicF2W)", %%mm7	\n\t"
//		"pxor %%mm6, %%mm6		\n\t"
		"1:				\n\t"
		"movq 1024(%1, %%"REG_S"), %%mm0\n\t"
		"movq 3072(%1, %%"REG_S"), %%mm1\n\t"
		"movq 4096(%1, %%"REG_S"), %%mm2\n\t"
		"movq 5120(%1, %%"REG_S"), %%mm3\n\t"
		"movq 2048(%1, %%"REG_S"), %%mm4\n\t"
		"movq (%1, %%"REG_S"), %%mm5	\n\t" 
		"psubd %%mm7, %%mm0		\n\t"
		"psubd %%mm7, %%mm1		\n\t"
		"psubd %%mm7, %%mm2		\n\t"
		"psubd %%mm7, %%mm3		\n\t"
		"psubd %%mm7, %%mm4		\n\t"
		"psubd %%mm7, %%mm5		\n\t"
		"lea (%%"REG_S", %%"REG_S", 2), %%"REG_D"\n\t"
		
		"packssdw %%mm2, %%mm0		\n\t" // CcAa
		"packssdw %%mm3, %%mm1		\n\t" // DdBb
		"packssdw %%mm4, %%mm4		\n\t" // EeEe
		"packssdw %%mm5, %%mm5		\n\t" // FfFf
		"movq %%mm0, %%mm2		\n\t" // CcAa
		"punpcklwd %%mm1, %%mm0		\n\t" // BAba
		"punpckhwd %%mm1, %%mm2		\n\t" // DCdc
		"punpcklwd %%mm5, %%mm4		\n\t" // FEfe
		"movq %%mm0, %%mm1		\n\t" // BAba
		"movq %%mm4, %%mm3		\n\t" // FEfe
		"punpckldq %%mm2, %%mm0		\n\t" // dcba
		"punpckhdq %%mm1, %%mm1		\n\t" // BABA
		"punpckldq %%mm1, %%mm4		\n\t" // BAfe
		"punpckhdq %%mm3, %%mm2		\n\t" // FEDC
		
		"movq %%mm0, (%0, %%"REG_D")	\n\t"
		"movq %%mm4, 8(%0, %%"REG_D")	\n\t"
		"movq %%mm2, 16(%0, %%"REG_D")	\n\t"
		"add $8, %%"REG_S"		\n\t"
		" jnz 1b			\n\t"
		"emms				\n\t"
		:: "r" (s16+1536), "r" (f+256)
		:"%"REG_S, "%"REG_D, "memory"
	);
    return 6*256;
}


static void* a52_resample_MMX(int flags, int ch){
    switch (flags) {
    case A52_MONO:
	if(ch==5) return a52_resample_MONO_to_5_MMX;
	break;
    case A52_CHANNEL:
    case A52_STEREO:
    case A52_DOLBY:
	if(ch==2) return a52_resample_STEREO_to_2_MMX;
	break;
    case A52_3F:
	if(ch==5) return a52_resample_3F_to_5_MMX;
	break;
    case A52_2F2R:
	if(ch==4) return a52_resample_2F_2R_to_4_MMX;
	break;
    case A52_3F2R:
	if(ch==5) return a52_resample_3F_2R_to_5_MMX;
	break;
    case A52_MONO | A52_LFE:
	if(ch==6) return a52_resample_MONO_LFE_to_6_MMX;
	break;
    case A52_CHANNEL | A52_LFE:
    case A52_STEREO | A52_LFE:
    case A52_DOLBY | A52_LFE:
	if(ch==6) return a52_resample_STEREO_LFE_to_6_MMX;
	break;
    case A52_3F | A52_LFE:
	if(ch==6) return a52_resample_3F_LFE_to_6_MMX;
	break;
    case A52_2F2R | A52_LFE:
	if(ch==6) return a52_resample_2F_2R_LFE_to_6_MMX;
	break;
    case A52_3F2R | A52_LFE:
	if(ch==6) return a52_resample_3F_2R_LFE_to_6_MMX;
	break;
    }
    return NULL;
}