1;******************************************************************************
2;* SIMD-optimized IDCT-related routines
3;* Copyright (c) 2008 Loren Merritt
4;* Copyright (c) 2003-2013 Michael Niedermayer
5;* Copyright (c) 2013 Daniel Kang
6;*
7;* This file is part of FFmpeg.
8;*
9;* FFmpeg is free software; you can redistribute it and/or
10;* modify it under the terms of the GNU Lesser General Public
11;* License as published by the Free Software Foundation; either
12;* version 2.1 of the License, or (at your option) any later version.
13;*
14;* FFmpeg is distributed in the hope that it will be useful,
15;* but WITHOUT ANY WARRANTY; without even the implied warranty of
16;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
17;* Lesser General Public License for more details.
18;*
19;* You should have received a copy of the GNU Lesser General Public
20;* License along with FFmpeg; if not, write to the Free Software
21;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
22;******************************************************************************
23
24%include "libavutil/x86/x86util.asm"
25
26SECTION_RODATA
27
28cextern pb_80
29
30SECTION .text
31
32;--------------------------------------------------------------------------
33;void ff_put_signed_pixels_clamped(const int16_t *block, uint8_t *pixels,
34;                                  ptrdiff_t line_size)
35;--------------------------------------------------------------------------
36
37%macro PUT_SIGNED_PIXELS_CLAMPED_HALF 1
38    mova     m1, [blockq+mmsize*0+%1]
39    mova     m2, [blockq+mmsize*2+%1]
40    packsswb m1, [blockq+mmsize*1+%1]
41    packsswb m2, [blockq+mmsize*3+%1]
42    paddb    m1, m0
43    paddb    m2, m0
44    movq     [pixelsq+lsizeq*0], m1
45    movhps   [pixelsq+lsizeq*1], m1
46    movq     [pixelsq+lsizeq*2], m2
47    movhps   [pixelsq+lsize3q ], m2
48%endmacro
49
50INIT_XMM sse2
51cglobal put_signed_pixels_clamped, 3, 4, 3, block, pixels, lsize, lsize3
52    mova     m0, [pb_80]
53    lea      lsize3q, [lsizeq*3]
54    PUT_SIGNED_PIXELS_CLAMPED_HALF 0
55    lea      pixelsq, [pixelsq+lsizeq*4]
56    PUT_SIGNED_PIXELS_CLAMPED_HALF 64
57    RET
58
59;--------------------------------------------------------------------------
60; void ff_put_pixels_clamped(const int16_t *block, uint8_t *pixels,
61;                            ptrdiff_t line_size);
62;--------------------------------------------------------------------------
63; %1 = block offset
64%macro PUT_PIXELS_CLAMPED_HALF 1
65    mova     m0, [blockq+mmsize*0+%1]
66    mova     m1, [blockq+mmsize*2+%1]
67    packuswb m0, [blockq+mmsize*1+%1]
68    packuswb m1, [blockq+mmsize*3+%1]
69    movq           [pixelsq], m0
70    movhps  [lsizeq+pixelsq], m0
71    movq  [2*lsizeq+pixelsq], m1
72    movhps [lsize3q+pixelsq], m1
73%endmacro
74
75INIT_XMM sse2
76cglobal put_pixels_clamped, 3, 4, 2, block, pixels, lsize, lsize3
77    lea lsize3q, [lsizeq*3]
78    PUT_PIXELS_CLAMPED_HALF 0
79    lea pixelsq, [pixelsq+lsizeq*4]
80    PUT_PIXELS_CLAMPED_HALF 64
81    RET
82
83;--------------------------------------------------------------------------
84; void ff_add_pixels_clamped(const int16_t *block, uint8_t *pixels,
85;                            ptrdiff_t line_size);
86;--------------------------------------------------------------------------
87; %1 = block offset
88%macro ADD_PIXELS_CLAMPED 1
89    mova       m0, [blockq+mmsize*0+%1]
90    mova       m1, [blockq+mmsize*1+%1]
91    movq       m2, [pixelsq]
92    movq       m3, [pixelsq+lsizeq]
93    punpcklbw  m2, m4
94    punpcklbw  m3, m4
95    paddsw     m0, m2
96    paddsw     m1, m3
97    packuswb   m0, m1
98    movq       [pixelsq], m0
99    movhps     [pixelsq+lsizeq], m0
100%endmacro
101
102INIT_XMM sse2
103cglobal add_pixels_clamped, 3, 3, 5, block, pixels, lsize
104    pxor       m4, m4
105    ADD_PIXELS_CLAMPED 0
106    lea        pixelsq, [pixelsq+lsizeq*2]
107    ADD_PIXELS_CLAMPED 32
108    lea        pixelsq, [pixelsq+lsizeq*2]
109    ADD_PIXELS_CLAMPED 64
110    lea        pixelsq, [pixelsq+lsizeq*2]
111    ADD_PIXELS_CLAMPED 96
112    RET
113