2002-05-29 20:16:22 +03:00
|
|
|
/*
|
|
|
|
* DSP utils mmx functions are compiled twice for rnd/no_rnd
|
2009-01-19 17:46:40 +02:00
|
|
|
* Copyright (c) 2000, 2001 Fabrice Bellard
|
2004-01-10 18:04:55 +02:00
|
|
|
* Copyright (c) 2003-2004 Michael Niedermayer <michaelni@gmx.at>
|
2002-05-29 20:16:22 +03:00
|
|
|
*
|
2007-07-05 13:37:29 +03:00
|
|
|
* MMX optimization by Nick Kurshev <nickols_k@mail.ru>
|
|
|
|
* mostly rewritten by Michael Niedermayer <michaelni@gmx.at>
|
|
|
|
* and improved by Zdenek Kabelac <kabi@users.sf.net>
|
|
|
|
*
|
2006-10-07 18:30:46 +03:00
|
|
|
* This file is part of FFmpeg.
|
|
|
|
*
|
|
|
|
* FFmpeg is free software; you can redistribute it and/or
|
2002-05-29 20:16:22 +03:00
|
|
|
* modify it under the terms of the GNU Lesser General Public
|
|
|
|
* License as published by the Free Software Foundation; either
|
2006-10-07 18:30:46 +03:00
|
|
|
* version 2.1 of the License, or (at your option) any later version.
|
2002-05-29 20:16:22 +03:00
|
|
|
*
|
2006-10-07 18:30:46 +03:00
|
|
|
* FFmpeg is distributed in the hope that it will be useful,
|
2002-05-29 20:16:22 +03:00
|
|
|
* 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
|
2006-10-07 18:30:46 +03:00
|
|
|
* License along with FFmpeg; if not, write to the Free Software
|
2006-01-13 00:43:26 +02:00
|
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
|
2002-05-29 20:16:22 +03:00
|
|
|
*/
|
|
|
|
|
2013-05-07 01:42:22 +03:00
|
|
|
#include <stddef.h>
|
|
|
|
#include <stdint.h>
|
|
|
|
|
2002-05-29 20:16:22 +03:00
|
|
|
// put_pixels
|
2013-05-07 01:42:22 +03:00
|
|
|
STATIC void DEF(put, pixels8_xy2)(uint8_t *block, const uint8_t *pixels,
|
|
|
|
ptrdiff_t line_size, int h)
|
2002-05-29 20:16:22 +03:00
|
|
|
{
|
2002-05-30 18:14:56 +03:00
|
|
|
MOVQ_ZERO(mm7);
|
|
|
|
SET_RND(mm6); // =2 for rnd and =1 for no_rnd version
|
2008-10-16 16:34:09 +03:00
|
|
|
__asm__ volatile(
|
2005-12-22 03:10:11 +02:00
|
|
|
"movq (%1), %%mm0 \n\t"
|
|
|
|
"movq 1(%1), %%mm4 \n\t"
|
|
|
|
"movq %%mm0, %%mm1 \n\t"
|
|
|
|
"movq %%mm4, %%mm5 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm0 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm4 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm1 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm5 \n\t"
|
|
|
|
"paddusw %%mm0, %%mm4 \n\t"
|
|
|
|
"paddusw %%mm1, %%mm5 \n\t"
|
|
|
|
"xor %%"REG_a", %%"REG_a" \n\t"
|
|
|
|
"add %3, %1 \n\t"
|
2011-01-16 00:09:35 +02:00
|
|
|
".p2align 3 \n\t"
|
2005-12-22 03:10:11 +02:00
|
|
|
"1: \n\t"
|
|
|
|
"movq (%1, %%"REG_a"), %%mm0 \n\t"
|
|
|
|
"movq 1(%1, %%"REG_a"), %%mm2 \n\t"
|
|
|
|
"movq %%mm0, %%mm1 \n\t"
|
|
|
|
"movq %%mm2, %%mm3 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm0 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm2 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm1 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm3 \n\t"
|
|
|
|
"paddusw %%mm2, %%mm0 \n\t"
|
|
|
|
"paddusw %%mm3, %%mm1 \n\t"
|
|
|
|
"paddusw %%mm6, %%mm4 \n\t"
|
|
|
|
"paddusw %%mm6, %%mm5 \n\t"
|
|
|
|
"paddusw %%mm0, %%mm4 \n\t"
|
|
|
|
"paddusw %%mm1, %%mm5 \n\t"
|
|
|
|
"psrlw $2, %%mm4 \n\t"
|
|
|
|
"psrlw $2, %%mm5 \n\t"
|
|
|
|
"packuswb %%mm5, %%mm4 \n\t"
|
|
|
|
"movq %%mm4, (%2, %%"REG_a") \n\t"
|
|
|
|
"add %3, %%"REG_a" \n\t"
|
2002-05-29 20:16:22 +03:00
|
|
|
|
2005-12-22 03:10:11 +02:00
|
|
|
"movq (%1, %%"REG_a"), %%mm2 \n\t" // 0 <-> 2 1 <-> 3
|
|
|
|
"movq 1(%1, %%"REG_a"), %%mm4 \n\t"
|
|
|
|
"movq %%mm2, %%mm3 \n\t"
|
|
|
|
"movq %%mm4, %%mm5 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm2 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm4 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm3 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm5 \n\t"
|
|
|
|
"paddusw %%mm2, %%mm4 \n\t"
|
|
|
|
"paddusw %%mm3, %%mm5 \n\t"
|
|
|
|
"paddusw %%mm6, %%mm0 \n\t"
|
|
|
|
"paddusw %%mm6, %%mm1 \n\t"
|
|
|
|
"paddusw %%mm4, %%mm0 \n\t"
|
|
|
|
"paddusw %%mm5, %%mm1 \n\t"
|
|
|
|
"psrlw $2, %%mm0 \n\t"
|
|
|
|
"psrlw $2, %%mm1 \n\t"
|
|
|
|
"packuswb %%mm1, %%mm0 \n\t"
|
|
|
|
"movq %%mm0, (%2, %%"REG_a") \n\t"
|
|
|
|
"add %3, %%"REG_a" \n\t"
|
2002-05-29 20:16:22 +03:00
|
|
|
|
2005-12-22 03:10:11 +02:00
|
|
|
"subl $2, %0 \n\t"
|
|
|
|
"jnz 1b \n\t"
|
|
|
|
:"+g"(h), "+S"(pixels)
|
2008-05-09 00:11:24 +03:00
|
|
|
:"D"(block), "r"((x86_reg)line_size)
|
2005-12-22 03:10:11 +02:00
|
|
|
:REG_a, "memory");
|
2002-05-29 20:16:22 +03:00
|
|
|
}
|
|
|
|
|
2013-04-20 00:14:44 +03:00
|
|
|
// avg_pixels
|
2002-06-03 20:31:46 +03:00
|
|
|
// this routine is 'slightly' suboptimal but mostly unused
|
2013-05-07 01:42:22 +03:00
|
|
|
STATIC void DEF(avg, pixels8_xy2)(uint8_t *block, const uint8_t *pixels,
|
|
|
|
ptrdiff_t line_size, int h)
|
2002-06-03 20:31:46 +03:00
|
|
|
{
|
|
|
|
MOVQ_ZERO(mm7);
|
|
|
|
SET_RND(mm6); // =2 for rnd and =1 for no_rnd version
|
2008-10-16 16:34:09 +03:00
|
|
|
__asm__ volatile(
|
2005-12-22 03:10:11 +02:00
|
|
|
"movq (%1), %%mm0 \n\t"
|
|
|
|
"movq 1(%1), %%mm4 \n\t"
|
|
|
|
"movq %%mm0, %%mm1 \n\t"
|
|
|
|
"movq %%mm4, %%mm5 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm0 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm4 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm1 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm5 \n\t"
|
|
|
|
"paddusw %%mm0, %%mm4 \n\t"
|
|
|
|
"paddusw %%mm1, %%mm5 \n\t"
|
|
|
|
"xor %%"REG_a", %%"REG_a" \n\t"
|
|
|
|
"add %3, %1 \n\t"
|
2011-01-16 00:09:35 +02:00
|
|
|
".p2align 3 \n\t"
|
2005-12-22 03:10:11 +02:00
|
|
|
"1: \n\t"
|
|
|
|
"movq (%1, %%"REG_a"), %%mm0 \n\t"
|
|
|
|
"movq 1(%1, %%"REG_a"), %%mm2 \n\t"
|
|
|
|
"movq %%mm0, %%mm1 \n\t"
|
|
|
|
"movq %%mm2, %%mm3 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm0 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm2 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm1 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm3 \n\t"
|
|
|
|
"paddusw %%mm2, %%mm0 \n\t"
|
|
|
|
"paddusw %%mm3, %%mm1 \n\t"
|
|
|
|
"paddusw %%mm6, %%mm4 \n\t"
|
|
|
|
"paddusw %%mm6, %%mm5 \n\t"
|
|
|
|
"paddusw %%mm0, %%mm4 \n\t"
|
|
|
|
"paddusw %%mm1, %%mm5 \n\t"
|
|
|
|
"psrlw $2, %%mm4 \n\t"
|
|
|
|
"psrlw $2, %%mm5 \n\t"
|
|
|
|
"movq (%2, %%"REG_a"), %%mm3 \n\t"
|
|
|
|
"packuswb %%mm5, %%mm4 \n\t"
|
|
|
|
"pcmpeqd %%mm2, %%mm2 \n\t"
|
|
|
|
"paddb %%mm2, %%mm2 \n\t"
|
2013-05-07 01:15:41 +03:00
|
|
|
PAVGB_MMX(%%mm3, %%mm4, %%mm5, %%mm2)
|
2005-12-22 03:10:11 +02:00
|
|
|
"movq %%mm5, (%2, %%"REG_a") \n\t"
|
|
|
|
"add %3, %%"REG_a" \n\t"
|
2002-06-03 20:31:46 +03:00
|
|
|
|
2005-12-22 03:10:11 +02:00
|
|
|
"movq (%1, %%"REG_a"), %%mm2 \n\t" // 0 <-> 2 1 <-> 3
|
|
|
|
"movq 1(%1, %%"REG_a"), %%mm4 \n\t"
|
|
|
|
"movq %%mm2, %%mm3 \n\t"
|
|
|
|
"movq %%mm4, %%mm5 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm2 \n\t"
|
|
|
|
"punpcklbw %%mm7, %%mm4 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm3 \n\t"
|
|
|
|
"punpckhbw %%mm7, %%mm5 \n\t"
|
|
|
|
"paddusw %%mm2, %%mm4 \n\t"
|
|
|
|
"paddusw %%mm3, %%mm5 \n\t"
|
|
|
|
"paddusw %%mm6, %%mm0 \n\t"
|
|
|
|
"paddusw %%mm6, %%mm1 \n\t"
|
|
|
|
"paddusw %%mm4, %%mm0 \n\t"
|
|
|
|
"paddusw %%mm5, %%mm1 \n\t"
|
|
|
|
"psrlw $2, %%mm0 \n\t"
|
|
|
|
"psrlw $2, %%mm1 \n\t"
|
|
|
|
"movq (%2, %%"REG_a"), %%mm3 \n\t"
|
|
|
|
"packuswb %%mm1, %%mm0 \n\t"
|
|
|
|
"pcmpeqd %%mm2, %%mm2 \n\t"
|
|
|
|
"paddb %%mm2, %%mm2 \n\t"
|
2013-05-07 01:15:41 +03:00
|
|
|
PAVGB_MMX(%%mm3, %%mm0, %%mm1, %%mm2)
|
2005-12-22 03:10:11 +02:00
|
|
|
"movq %%mm1, (%2, %%"REG_a") \n\t"
|
|
|
|
"add %3, %%"REG_a" \n\t"
|
2002-06-03 20:31:46 +03:00
|
|
|
|
2005-12-22 03:10:11 +02:00
|
|
|
"subl $2, %0 \n\t"
|
|
|
|
"jnz 1b \n\t"
|
|
|
|
:"+g"(h), "+S"(pixels)
|
2008-05-09 00:11:24 +03:00
|
|
|
:"D"(block), "r"((x86_reg)line_size)
|
2005-12-22 03:10:11 +02:00
|
|
|
:REG_a, "memory");
|
2002-06-03 20:31:46 +03:00
|
|
|
}
|