yading@10: ;******************************************************************************
yading@10: ;* MMX optimized deinterlacing functions
yading@10: ;* Copyright (c) 2010 Vitor Sessak
yading@10: ;* Copyright (c) 2002 Michael Niedermayer
yading@10: ;*
yading@10: ;* This file is part of FFmpeg.
yading@10: ;*
yading@10: ;* FFmpeg is free software; you can redistribute it and/or
yading@10: ;* modify it under the terms of the GNU Lesser General Public
yading@10: ;* License as published by the Free Software Foundation; either
yading@10: ;* version 2.1 of the License, or (at your option) any later version.
yading@10: ;*
yading@10: ;* FFmpeg is distributed in the hope that it will be useful,
yading@10: ;* but WITHOUT ANY WARRANTY; without even the implied warranty of
yading@10: ;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
yading@10: ;* Lesser General Public License for more details.
yading@10: ;*
yading@10: ;* You should have received a copy of the GNU Lesser General Public
yading@10: ;* License along with FFmpeg; if not, write to the Free Software
yading@10: ;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
yading@10: ;******************************************************************************
yading@10: 
yading@10: %include "libavutil/x86/x86util.asm"
yading@10: 
yading@10: SECTION_RODATA
yading@10: 
yading@10: cextern pw_4
yading@10: 
yading@10: SECTION .text
yading@10: 
yading@10: %macro DEINTERLACE 1
yading@10: %ifidn %1, inplace
yading@10: ;void ff_deinterlace_line_inplace_mmx(const uint8_t *lum_m4, const uint8_t *lum_m3, const uint8_t *lum_m2, const uint8_t *lum_m1, const uint8_t *lum,  int size)
yading@10: cglobal deinterlace_line_inplace_mmx, 6,6,7,      lum_m4, lum_m3, lum_m2, lum_m1, lum, size
yading@10: %else
yading@10: ;void ff_deinterlace_line_mmx(uint8_t *dst, const uint8_t *lum_m4, const uint8_t *lum_m3, const uint8_t *lum_m2, const uint8_t *lum_m1, const uint8_t *lum,  int size)
yading@10: cglobal deinterlace_line_mmx,         7,7,7, dst, lum_m4, lum_m3, lum_m2, lum_m1, lum, size
yading@10: %endif
yading@10:     pxor  mm7, mm7
yading@10:     movq  mm6, [pw_4]
yading@10: .nextrow:
yading@10:     movd  mm0, [lum_m4q]
yading@10:     movd  mm1, [lum_m3q]
yading@10:     movd  mm2, [lum_m2q]
yading@10: %ifidn %1, inplace
yading@10:     movd [lum_m4q], mm2
yading@10: %endif
yading@10:     movd  mm3, [lum_m1q]
yading@10:     movd  mm4, [lumq]
yading@10:     punpcklbw mm0, mm7
yading@10:     punpcklbw mm1, mm7
yading@10:     punpcklbw mm2, mm7
yading@10:     punpcklbw mm3, mm7
yading@10:     punpcklbw mm4, mm7
yading@10:     paddw     mm1, mm3
yading@10:     psllw     mm2, 1
yading@10:     paddw     mm0, mm4
yading@10:     psllw     mm1, 2
yading@10:     paddw     mm2, mm6
yading@10:     paddw     mm1, mm2
yading@10:     psubusw   mm1, mm0
yading@10:     psrlw     mm1, 3
yading@10:     packuswb  mm1, mm7
yading@10: %ifidn %1, inplace
yading@10:     movd [lum_m2q], mm1
yading@10: %else
yading@10:     movd   [dstq], mm1
yading@10:     add       dstq, 4
yading@10: %endif
yading@10:     add    lum_m4q, 4
yading@10:     add    lum_m3q, 4
yading@10:     add    lum_m2q, 4
yading@10:     add    lum_m1q, 4
yading@10:     add       lumq, 4
yading@10:     sub      sized, 4
yading@10:     jg .nextrow
yading@10:     REP_RET
yading@10: %endmacro
yading@10: 
yading@10: DEINTERLACE ""
yading@10: 
yading@10: DEINTERLACE inplace