generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Finish mmx2, 3dnow optimiz. 15to16 should be tested. Better fix of can't compile

Browse Source 
Originally committed as revision 2539 to svn://svn.mplayerhq.hu/mplayer/trunk/postproc
This commit is contained in:
Nick Kurshev 2001-10-29 18:28:06 +00:00
parent e697a1413b
commit a3aece931b
2 changed files with 110 additions and 92 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

89
postproc/rgb2rgb.c
View File

@ -1,23 +1,32 @@
/*
*
* rgb2rgb.c, Software RGB to RGB convertor
* Written by Nick Kurshev.
*/
#include <inttypes.h> #include <inttypes.h>
#include "../config.h" #include "../config.h"
#include "rgb2rgb.h" #include "rgb2rgb.h"
#ifdef HAVE_MMX
#include "mmx.h"
#endif
#include "../mmx_defs.h" #include "../mmx_defs.h"
#ifdef HAVE_MMX
static const uint64_t mask32 __attribute__((aligned(8))) = 0x00FFFFFF00FFFFFFULL;
static const uint64_t mask24l __attribute__((aligned(8))) = 0x0000000000FFFFFFULL;
static const uint64_t mask24h __attribute__((aligned(8))) = 0x0000FFFFFF000000ULL;
static const uint64_t mask15b __attribute__((aligned(8))) = 0x001F001F001F001FULL; /* 00000000 00011111 xxB */
static const uint64_t mask15rg __attribute__((aligned(8))) = 0x7FE07FE07FE07FE0ULL; /* 01111111 11100000 RGx */
#endif
void rgb24to32(uint8_t *src,uint8_t *dst,uint32_t src_size) void rgb24to32(uint8_t *src,uint8_t *dst,uint32_t src_size)
{ {
uint8_t *dest = dst; uint8_t *dest = dst;
uint8_t *s = src; uint8_t *s = src;
uint8_t *end; uint8_t *end;
#ifdef HAVE_MMX #ifdef HAVE_MMX
const uint64_t mask32 = 0x00FFFFFF00FFFFFFULL;
uint8_t *mm_end; uint8_t *mm_end;
#endif #endif
end = s + src_size; end = s + src_size;
#ifdef HAVE_MMX #ifdef HAVE_MMX
__asm __volatile(PREFETCH" %0\n\t"::"m"(*s):"memory"); __asm __volatile(PREFETCH" %0"::"m"(*s):"memory");
mm_end = (uint8_t*)((((unsigned long)end)/(MMREG_SIZE*2))*(MMREG_SIZE*2)); mm_end = (uint8_t*)((((unsigned long)end)/(MMREG_SIZE*2))*(MMREG_SIZE*2));
__asm __volatile("movq %0, %%mm7"::"m"(mask32):"memory"); __asm __volatile("movq %0, %%mm7"::"m"(mask32):"memory");
if(mm_end == end) mm_end -= MMREG_SIZE*2; if(mm_end == end) mm_end -= MMREG_SIZE*2;
@ -59,17 +68,15 @@ void rgb32to24(uint8_t *src,uint8_t *dst,uint32_t src_size)
uint8_t *s = src; uint8_t *s = src;
uint8_t *end; uint8_t *end;
#ifdef HAVE_MMX #ifdef HAVE_MMX
const uint64_t mask24l = 0x0000000000FFFFFFULL;
const uint64_t mask24h = 0x0000FFFFFF000000ULL;
uint8_t *mm_end; uint8_t *mm_end;
#endif #endif
end = s + src_size; end = s + src_size;
#ifdef HAVE_MMX #ifdef HAVE_MMX
__asm __volatile(PREFETCH" %0\n\t"::"m"(*s):"memory"); __asm __volatile(PREFETCH" %0"::"m"(*s):"memory");
mm_end = (uint8_t*)((((unsigned long)end)/(MMREG_SIZE*2))*(MMREG_SIZE*2)); mm_end = (uint8_t*)((((unsigned long)end)/(MMREG_SIZE*2))*(MMREG_SIZE*2));
__asm __volatile( __asm __volatile(
"movq %0, %%mm7\n\t" "movq %0, %%mm7\n\t"
"movq %1, %%mm6\n\t" "movq %1, %%mm6"
::"m"(mask24l),"m"(mask24h):"memory"); ::"m"(mask24l),"m"(mask24h):"memory");
if(mm_end == end) mm_end -= MMREG_SIZE*2; if(mm_end == end) mm_end -= MMREG_SIZE*2;
while(s < mm_end) while(s < mm_end)
@ -108,45 +115,47 @@ void rgb32to24(uint8_t *src,uint8_t *dst,uint32_t src_size)
} }
} }
/* TODO: 3DNOW, MMX2 optimization */ /*
Original by Strepto/Astral
/* Original by Strepto/Astral ported to gcc & bugfixed : A'rpi
ported to gcc & bugfixed : A'rpi */ MMX, 3DNOW optimization by Nick Kurshev
*/
void rgb15to16(uint8_t *src,uint8_t *dst,uint32_t src_size) void rgb15to16(uint8_t *src,uint8_t *dst,uint32_t src_size)
{ {
#ifdef HAVE_MMX #ifdef HAVE_MMX
static uint64_t mask_b = 0x001F001F001F001FLL; // 00000000 00011111 xxB
static uint64_t mask_rg = 0x7FE07FE07FE07FE0LL; // 01111111 11100000 RGx
register char* s=src+src_size; register char* s=src+src_size;
register char* d=dst+src_size; register char* d=dst+src_size;
register int offs=-src_size; register int offs=-src_size;
movq_m2r (mask_b, mm4); __asm __volatile(PREFETCH" %0"::"m"(*(s+offs)):"memory");
movq_m2r (mask_rg, mm5); __asm __volatile(
while(offs<0){ "movq %0, %%mm4\n\t"
movq_m2r (*(s+offs), mm0); "movq %1, %%mm5"
movq_r2r (mm0, mm1); ::"m"(mask15b), "m"(mask15rg):"memory");
while(offs<0)
movq_m2r (*(s+8+offs), mm2); {
movq_r2r (mm2, mm3); __asm __volatile(
PREFETCH" 32%1\n\t"
pand_r2r (mm4, mm0); "movq %1, %%mm0\n\t"
pand_r2r (mm5, mm1); "movq 8%1, %%mm2\n\t"
"movq %%mm0, %%mm1\n\t"
psllq_i2r(1,mm1); "movq %%mm2, %%mm3\n\t"
pand_r2r (mm4, mm2); "pand %%mm4, %%mm0\n\t"
"pand %%mm5, %%mm1\n\t"
pand_r2r (mm5, mm3); "pand %%mm4, %%mm2\n\t"
por_r2r (mm1, mm0); "pand %%mm5, %%mm3\n\t"
"psllq $1, %%mm1\n\t"
psllq_i2r(1,mm3); "psllq $1, %%mm3\n\t"
movq_r2m (mm0,*(d+offs)); "por %%mm1, %%mm0\n\t"
"por %%mm3, %%mm2\n\t"
por_r2r (mm3,mm2); MOVNTQ" %%mm0, %0\n\t"
movq_r2m (mm2,*(d+8+offs)); MOVNTQ" %%mm2, 8%0"
:"=m"(*(d+offs))
:"m"(*(s+offs))
:"memory");
offs+=16; offs+=16;
} }
emms(); __asm __volatile(SFENCE:::"memory");
__asm __volatile(EMMS:::"memory");
#else #else
uint16_t *s1=( uint16_t * )src; uint16_t *s1=( uint16_t * )src;
uint16_t *d1=( uint16_t * )dst; uint16_t *d1=( uint16_t * )dst;

89
postproc/rgb2rgb_template.c
View File

@ -1,23 +1,32 @@
/*
*
* rgb2rgb.c, Software RGB to RGB convertor
* Written by Nick Kurshev.
*/
#include <inttypes.h> #include <inttypes.h>
#include "../config.h" #include "../config.h"
#include "rgb2rgb.h" #include "rgb2rgb.h"
#ifdef HAVE_MMX
#include "mmx.h"
#endif
#include "../mmx_defs.h" #include "../mmx_defs.h"
#ifdef HAVE_MMX
static const uint64_t mask32 __attribute__((aligned(8))) = 0x00FFFFFF00FFFFFFULL;
static const uint64_t mask24l __attribute__((aligned(8))) = 0x0000000000FFFFFFULL;
static const uint64_t mask24h __attribute__((aligned(8))) = 0x0000FFFFFF000000ULL;
static const uint64_t mask15b __attribute__((aligned(8))) = 0x001F001F001F001FULL; /* 00000000 00011111 xxB */
static const uint64_t mask15rg __attribute__((aligned(8))) = 0x7FE07FE07FE07FE0ULL; /* 01111111 11100000 RGx */
#endif
void rgb24to32(uint8_t *src,uint8_t *dst,uint32_t src_size) void rgb24to32(uint8_t *src,uint8_t *dst,uint32_t src_size)
{ {
uint8_t *dest = dst; uint8_t *dest = dst;
uint8_t *s = src; uint8_t *s = src;
uint8_t *end; uint8_t *end;
#ifdef HAVE_MMX #ifdef HAVE_MMX
const uint64_t mask32 = 0x00FFFFFF00FFFFFFULL;
uint8_t *mm_end; uint8_t *mm_end;
#endif #endif
end = s + src_size; end = s + src_size;
#ifdef HAVE_MMX #ifdef HAVE_MMX
__asm __volatile(PREFETCH" %0\n\t"::"m"(*s):"memory"); __asm __volatile(PREFETCH" %0"::"m"(*s):"memory");
mm_end = (uint8_t*)((((unsigned long)end)/(MMREG_SIZE*2))*(MMREG_SIZE*2)); mm_end = (uint8_t*)((((unsigned long)end)/(MMREG_SIZE*2))*(MMREG_SIZE*2));
__asm __volatile("movq %0, %%mm7"::"m"(mask32):"memory"); __asm __volatile("movq %0, %%mm7"::"m"(mask32):"memory");
if(mm_end == end) mm_end -= MMREG_SIZE*2; if(mm_end == end) mm_end -= MMREG_SIZE*2;
@ -59,17 +68,15 @@ void rgb32to24(uint8_t *src,uint8_t *dst,uint32_t src_size)
uint8_t *s = src; uint8_t *s = src;
uint8_t *end; uint8_t *end;
#ifdef HAVE_MMX #ifdef HAVE_MMX
const uint64_t mask24l = 0x0000000000FFFFFFULL;
const uint64_t mask24h = 0x0000FFFFFF000000ULL;
uint8_t *mm_end; uint8_t *mm_end;
#endif #endif
end = s + src_size; end = s + src_size;
#ifdef HAVE_MMX #ifdef HAVE_MMX
__asm __volatile(PREFETCH" %0\n\t"::"m"(*s):"memory"); __asm __volatile(PREFETCH" %0"::"m"(*s):"memory");
mm_end = (uint8_t*)((((unsigned long)end)/(MMREG_SIZE*2))*(MMREG_SIZE*2)); mm_end = (uint8_t*)((((unsigned long)end)/(MMREG_SIZE*2))*(MMREG_SIZE*2));
__asm __volatile( __asm __volatile(
"movq %0, %%mm7\n\t" "movq %0, %%mm7\n\t"
"movq %1, %%mm6\n\t" "movq %1, %%mm6"
::"m"(mask24l),"m"(mask24h):"memory"); ::"m"(mask24l),"m"(mask24h):"memory");
if(mm_end == end) mm_end -= MMREG_SIZE*2; if(mm_end == end) mm_end -= MMREG_SIZE*2;
while(s < mm_end) while(s < mm_end)
@ -108,45 +115,47 @@ void rgb32to24(uint8_t *src,uint8_t *dst,uint32_t src_size)
} }
} }
/* TODO: 3DNOW, MMX2 optimization */ /*
Original by Strepto/Astral
/* Original by Strepto/Astral ported to gcc & bugfixed : A'rpi
ported to gcc & bugfixed : A'rpi */ MMX, 3DNOW optimization by Nick Kurshev
*/
void rgb15to16(uint8_t *src,uint8_t *dst,uint32_t src_size) void rgb15to16(uint8_t *src,uint8_t *dst,uint32_t src_size)
{ {
#ifdef HAVE_MMX #ifdef HAVE_MMX
static uint64_t mask_b = 0x001F001F001F001FLL; // 00000000 00011111 xxB
static uint64_t mask_rg = 0x7FE07FE07FE07FE0LL; // 01111111 11100000 RGx
register char* s=src+src_size; register char* s=src+src_size;
register char* d=dst+src_size; register char* d=dst+src_size;
register int offs=-src_size; register int offs=-src_size;
movq_m2r (mask_b, mm4); __asm __volatile(PREFETCH" %0"::"m"(*(s+offs)):"memory");
movq_m2r (mask_rg, mm5); __asm __volatile(
while(offs<0){ "movq %0, %%mm4\n\t"
movq_m2r (*(s+offs), mm0); "movq %1, %%mm5"
movq_r2r (mm0, mm1); ::"m"(mask15b), "m"(mask15rg):"memory");
while(offs<0)
movq_m2r (*(s+8+offs), mm2); {
movq_r2r (mm2, mm3); __asm __volatile(
PREFETCH" 32%1\n\t"
pand_r2r (mm4, mm0); "movq %1, %%mm0\n\t"
pand_r2r (mm5, mm1); "movq 8%1, %%mm2\n\t"
"movq %%mm0, %%mm1\n\t"
psllq_i2r(1,mm1); "movq %%mm2, %%mm3\n\t"
pand_r2r (mm4, mm2); "pand %%mm4, %%mm0\n\t"
"pand %%mm5, %%mm1\n\t"
pand_r2r (mm5, mm3); "pand %%mm4, %%mm2\n\t"
por_r2r (mm1, mm0); "pand %%mm5, %%mm3\n\t"
"psllq $1, %%mm1\n\t"
psllq_i2r(1,mm3); "psllq $1, %%mm3\n\t"
movq_r2m (mm0,*(d+offs)); "por %%mm1, %%mm0\n\t"
"por %%mm3, %%mm2\n\t"
por_r2r (mm3,mm2); MOVNTQ" %%mm0, %0\n\t"
movq_r2m (mm2,*(d+8+offs)); MOVNTQ" %%mm2, 8%0"
:"=m"(*(d+offs))
:"m"(*(s+offs))
:"memory");
offs+=16; offs+=16;
} }
emms(); __asm __volatile(SFENCE:::"memory");
__asm __volatile(EMMS:::"memory");
#else #else
uint16_t *s1=( uint16_t * )src; uint16_t *s1=( uint16_t * )src;
uint16_t *d1=( uint16_t * )dst; uint16_t *d1=( uint16_t * )dst;