1 /*****************************************************************************
3 *****************************************************************************
4 * Copyright (C) 2004-2016 x264 project
6 * Authors: Loren Merritt <lorenm@u.washington.edu>
8 * This program is free software; you can redistribute it and/or modify
9 * it under the terms of the GNU General Public License as published by
10 * the Free Software Foundation; either version 2 of the License, or
11 * (at your option) any later version.
13 * This program is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 * GNU General Public License for more details.
18 * You should have received a copy of the GNU General Public License
19 * along with this program; if not, write to the Free Software
20 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02111, USA.
22 * This program is also available under a commercial proprietary license.
23 * For more information, contact us at licensing@x264.com.
24 *****************************************************************************/
29 uint32_t x264_cpu_detect( void );
30 int x264_cpu_num_processors( void );
31 void x264_cpu_emms( void );
32 void x264_cpu_sfence( void );
34 /* There is no way to forbid the compiler from using float instructions
35 * before the emms so miscompilation could theoretically occur in the
36 * unlikely event that the compiler reorders emms and float instructions. */
37 #if HAVE_X86_INLINE_ASM
38 /* Clobbering memory makes the compiler less likely to reorder code. */
39 #define x264_emms() asm volatile( "emms":::"memory","st","st(1)","st(2)", \
40 "st(3)","st(4)","st(5)","st(6)","st(7)" )
42 #define x264_emms() x264_cpu_emms()
47 #define x264_sfence x264_cpu_sfence
50 * gcc can't give variables any greater alignment than the stack frame has.
51 * We need 32 byte alignment for AVX2, so here we make sure that the stack is
52 * aligned to 32 bytes.
53 * gcc 4.2 introduced __attribute__((force_align_arg_pointer)) to fix this
54 * problem, but I don't want to require such a new version.
55 * aligning to 32 bytes only works if the compiler supports keeping that
56 * alignment between functions (osdep.h handles manual alignment of arrays
59 #if (ARCH_X86 || STACK_ALIGNMENT > 16) && HAVE_MMX
60 intptr_t x264_stack_align( void (*func)(), ... );
61 #define x264_stack_align(func,...) x264_stack_align((void (*)())func, __VA_ARGS__)
63 #define x264_stack_align(func,...) func(__VA_ARGS__)
71 extern const x264_cpu_name_t x264_cpu_names[];