util: Optimize _mesa_roundeven with SSE 4.1.
[mesa.git] / src / util / rounding.h
1 /*
2 * Copyright © 2015 Intel Corporation
3 *
4 * Permission is hereby granted, free of charge, to any person obtaining a
5 * copy of this software and associated documentation files (the "Software"),
6 * to deal in the Software without restriction, including without limitation
7 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
8 * and/or sell copies of the Software, and to permit persons to whom the
9 * Software is furnished to do so, subject to the following conditions:
10 *
11 * The above copyright notice and this permission notice (including the next
12 * paragraph) shall be included in all copies or substantial portions of the
13 * Software.
14 *
15 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
18 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
20 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
21 * IN THE SOFTWARE.
22 */
23
24 #include <math.h>
25
26 #ifdef __SSE4_1__
27 #include <smmintrin.h>
28 #endif
29
30 /* The C standard library has functions round()/rint()/nearbyint() that round
31 * their arguments according to the rounding mode set in the floating-point
32 * control register. While there are trunc()/ceil()/floor() functions that do
33 * a specific operation without modifying the rounding mode, there is no
34 * roundeven() in any version of C.
35 *
36 * Technical Specification 18661 (ISO/IEC TS 18661-1:2014) adds roundeven(),
37 * but it's unfortunately not implemented by glibc.
38 *
39 * This implementation differs in that it does not raise the inexact exception.
40 *
41 * We use rint() to implement these functions, with the assumption that the
42 * floating-point rounding mode has not been changed from the default Round
43 * to Nearest.
44 */
45
46 /**
47 * \brief Rounds \c x to the nearest integer, with ties to the even integer.
48 */
49 static inline float
50 _mesa_roundevenf(float x)
51 {
52 #ifdef __SSE4_1__
53 float ret;
54 __m128 m = _mm_load_ss(&x);
55 m = _mm_round_ss(m, m, _MM_FROUND_CUR_DIRECTION | _MM_FROUND_NO_EXC);
56 _mm_store_ss(&ret, m);
57 return ret;
58 #else
59 return rintf(x);
60 #endif
61 }
62
63 /**
64 * \brief Rounds \c x to the nearest integer, with ties to the even integer.
65 */
66 static inline double
67 _mesa_roundeven(double x)
68 {
69 #ifdef __SSE4_1__
70 double ret;
71 __m128d m = _mm_load_sd(&x);
72 m = _mm_round_sd(m, m, _MM_FROUND_CUR_DIRECTION | _MM_FROUND_NO_EXC);
73 _mm_store_sd(&ret, m);
74 return ret;
75 #else
76 return rint(x);
77 #endif
78 }