1 ;*****************************************************************************
3 ;*****************************************************************************
4 ;* Copyright (C) 2008-2010 x264 project
6 ;* Authors: Loren Merritt <lorenm@u.washington.edu>
7 ;* Holger Lubitz <holger@lubitz.org>
9 ;* This file is part of FFmpeg.
11 ;* FFmpeg is free software; you can redistribute it and/or
12 ;* modify it under the terms of the GNU Lesser General Public
13 ;* License as published by the Free Software Foundation; either
14 ;* version 2.1 of the License, or (at your option) any later version.
16 ;* FFmpeg is distributed in the hope that it will be useful,
17 ;* but WITHOUT ANY WARRANTY; without even the implied warranty of
18 ;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19 ;* Lesser General Public License for more details.
21 ;* You should have received a copy of the GNU Lesser General Public
22 ;* License along with FFmpeg; if not, write to the Free Software
23 ;* 51, Inc., Foundation Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
24 ;******************************************************************************
40 %macro TRANSPOSE4x4W 5
41 SBUTTERFLY wd, %1, %2, %5
42 SBUTTERFLY wd, %3, %4, %5
43 SBUTTERFLY dq, %1, %3, %5
44 SBUTTERFLY dq, %2, %4, %5
48 %macro TRANSPOSE2x4x4W 5
49 SBUTTERFLY wd, %1, %2, %5
50 SBUTTERFLY wd, %3, %4, %5
51 SBUTTERFLY dq, %1, %3, %5
52 SBUTTERFLY dq, %2, %4, %5
53 SBUTTERFLY qdq, %1, %2, %5
54 SBUTTERFLY qdq, %3, %4, %5
57 %macro TRANSPOSE4x4D 5
58 SBUTTERFLY dq, %1, %2, %5
59 SBUTTERFLY dq, %3, %4, %5
60 SBUTTERFLY qdq, %1, %3, %5
61 SBUTTERFLY qdq, %2, %4, %5
65 %macro TRANSPOSE8x8W 9-11
67 SBUTTERFLY wd, %1, %2, %9
68 SBUTTERFLY wd, %3, %4, %9
69 SBUTTERFLY wd, %5, %6, %9
70 SBUTTERFLY wd, %7, %8, %9
71 SBUTTERFLY dq, %1, %3, %9
72 SBUTTERFLY dq, %2, %4, %9
73 SBUTTERFLY dq, %5, %7, %9
74 SBUTTERFLY dq, %6, %8, %9
75 SBUTTERFLY qdq, %1, %5, %9
76 SBUTTERFLY qdq, %2, %6, %9
77 SBUTTERFLY qdq, %3, %7, %9
78 SBUTTERFLY qdq, %4, %8, %9
82 ; in: m0..m7, unless %11 in which case m6 is in %9
83 ; out: m0..m7, unless %11 in which case m4 is in %10
84 ; spills into %9 and %10
88 SBUTTERFLY wd, %1, %2, %7
91 SBUTTERFLY wd, %3, %4, %2
92 SBUTTERFLY wd, %5, %6, %2
93 SBUTTERFLY wd, %7, %8, %2
94 SBUTTERFLY dq, %1, %3, %2
97 SBUTTERFLY dq, %2, %4, %3
98 SBUTTERFLY dq, %5, %7, %3
99 SBUTTERFLY dq, %6, %8, %3
100 SBUTTERFLY qdq, %1, %5, %3
101 SBUTTERFLY qdq, %2, %6, %3
104 SBUTTERFLY qdq, %3, %7, %2
105 SBUTTERFLY qdq, %4, %8, %2
114 %macro ABS1_MMX 2 ; a, tmp
120 %macro ABS2_MMX 4 ; a, b, tmp0, tmp1
167 %define ABS1 ABS1_MMX
168 %define ABS2 ABS2_MMX
169 %define ABSB ABSB_MMX
170 %define ABSB2 ABSB2_MMX
173 movd %1, [%2-3] ;to avoid crossing a cacheline
183 %macro SPLATB_SSSE3 3
202 %macro PALIGNR_SSSE3 4
206 %macro DEINTB 5 ; mask, reg1, mask, reg2, optional src to fill masks from
214 pand m%1, m%2 ; dst .. y6 .. y4
215 pand m%3, m%4 ; src .. y6 .. y4
216 psrlw m%2, 8 ; dst .. y7 .. y5
217 psrlw m%4, 8 ; src .. y7 .. y5
232 %macro SUMSUB_BADC 4-5
265 psraw %2, 1 ; %2: %2>>1
266 psraw %1, 1 ; %1: %1>>1
267 paddw %2, %4 ; %2: %2>>1+%1
268 psubw %1, %3 ; %1: %1>>1-%2
273 SUMSUB_BADC m%4, m%1, m%3, m%2; m%5
274 SUMSUB_BA m%3, m%4, m%5
275 SUMSUB2_AB m%1, m%2, m%5
276 SWAP %1, %3, %4, %5, %2
278 SUMSUB_BADC m%4, m%1, m%3, m%2
281 SUMSUB2_AB m%1, [%5], m%2
288 SUMSUBD2_AB m%2, m%4, m%6, m%5
289 ; %2: %2>>1-%4 %4: %2+%4>>1
290 SUMSUB_BA m%3, m%1, m%6
291 ; %3: %1+%3 %1: %1-%3
292 SUMSUB_BADC m%4, m%3, m%2, m%1, m%6
293 ; %4: %1+%3 + (%2+%4>>1)
294 ; %3: %1+%3 - (%2+%4>>1)
295 ; %2: %1-%3 + (%2>>1-%4)
296 ; %1: %1-%3 - (%2>>1-%4)
298 SUMSUBD2_AB m%2, m%4, [%5], [%5+16]
300 SUMSUB_BADC m%4, m%3, m%2, m%1
303 ; %1: %1+%3 + (%2+%4>>1) row0
304 ; %2: %1-%3 + (%2>>1-%4) row1
305 ; %3: %1-%3 - (%2>>1-%4) row2
306 ; %4: %1+%3 - (%2+%4>>1) row3
331 movhps [%5+%6+32], m%1
332 movhps [%5+%6+40], m%2
333 movhps [%5+%6+48], m%3
334 movhps [%5+%6+56], m%4
337 %macro LOAD_DIFF_8x4P 7-10 r0,r2,0 ; 4x dest, 2x temp, 2x pointer, increment?
338 LOAD_DIFF m%1, m%5, m%7, [%8], [%9]
339 LOAD_DIFF m%2, m%6, m%7, [%8+r1], [%9+r3]
340 LOAD_DIFF m%3, m%5, m%7, [%8+2*r1], [%9+2*r3]
341 LOAD_DIFF m%4, m%6, m%7, [%8+r4], [%9+r5]