1 | /* |
2 | * Copyright (c) 2017, Intel Corporation |
3 | * |
4 | * Redistribution and use in source and binary forms, with or without |
5 | * modification, are permitted provided that the following conditions are met: |
6 | * |
7 | * * Redistributions of source code must retain the above copyright notice, |
8 | * this list of conditions and the following disclaimer. |
9 | * * Redistributions in binary form must reproduce the above copyright |
10 | * notice, this list of conditions and the following disclaimer in the |
11 | * documentation and/or other materials provided with the distribution. |
12 | * * Neither the name of Intel Corporation nor the names of its contributors |
13 | * may be used to endorse or promote products derived from this software |
14 | * without specific prior written permission. |
15 | * |
16 | * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" |
17 | * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
18 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
19 | * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE |
20 | * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR |
21 | * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF |
22 | * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS |
23 | * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN |
24 | * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) |
25 | * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE |
26 | * POSSIBILITY OF SUCH DAMAGE. |
27 | */ |
28 | |
29 | #ifndef UTIL_MATH_H_ |
30 | #define UTIL_MATH_H_ |
31 | |
32 | #include "arch.h" |
33 | #include "intrinsics.h" |
34 | |
35 | #include <math.h> |
36 | |
37 | static really_inline |
38 | double our_pow(double x, double y) { |
39 | #if defined(HAVE_AVX) |
40 | /* |
41 | * Clear the upper half of AVX registers before calling into the math lib. |
42 | * On some versions of glibc this can save thousands of AVX-to-SSE |
43 | * transitions. |
44 | */ |
45 | _mm256_zeroupper(); |
46 | #endif |
47 | return pow(x, y); |
48 | } |
49 | |
50 | #endif // UTIL_MATH_H_ |
51 | |