| 1 |
/* |
| 2 |
* // Copyright (c) Radzivon Bartoshyk 9/2025. All rights reserved. |
| 3 |
* // |
| 4 |
* // Redistribution and use in source and binary forms, with or without modification, |
| 5 |
* // are permitted provided that the following conditions are met: |
| 6 |
* // |
| 7 |
* // 1. Redistributions of source code must retain the above copyright notice, this |
| 8 |
* // list of conditions and the following disclaimer. |
| 9 |
* // |
| 10 |
* // 2. Redistributions in binary form must reproduce the above copyright notice, |
| 11 |
* // this list of conditions and the following disclaimer in the documentation |
| 12 |
* // and/or other materials provided with the distribution. |
| 13 |
* // |
| 14 |
* // 3. Neither the name of the copyright holder nor the names of its |
| 15 |
* // contributors may be used to endorse or promote products derived from |
| 16 |
* // this software without specific prior written permission. |
| 17 |
* // |
| 18 |
* // THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" |
| 19 |
* // AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
| 20 |
* // IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE |
| 21 |
* // DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE |
| 22 |
* // FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL |
| 23 |
* // DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR |
| 24 |
* // SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER |
| 25 |
* // CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, |
| 26 |
* // OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE |
| 27 |
* // OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
| 28 |
*/ |
| 29 |
use crate::polyeval::f_polyeval6; |
| 30 |
|
| 31 |
/// Computes sqrt(1+x) - 1 |
| 32 |
pub fn f_sqrt1pm1f(x: f32) -> f32 { |
| 33 |
// filter out exceptional cases |
| 34 |
let ix = x.to_bits(); |
| 35 |
let ux = ix.wrapping_shl(1); |
| 36 |
if ux >= 0xffu32 << 24 || ux <= 0x68000000u32 { |
| 37 |
// |x| <= f32::EPSILON, |x| == inf, x == NaN |
| 38 |
if ux == 0 { |
| 39 |
// |x| == 0 |
| 40 |
return 0.; |
| 41 |
} |
| 42 |
if ux.wrapping_shl(8) == 0 { |
| 43 |
// |x| == Inf |
| 44 |
return if x.is_sign_negative() { |
| 45 |
f32::NAN |
| 46 |
} else { |
| 47 |
f32::INFINITY |
| 48 |
}; |
| 49 |
} |
| 50 |
|
| 51 |
if ux <= 0x68000000u32 { |
| 52 |
// |x| <= f32::EPSILON |
| 53 |
// Sqrt(1+x) - 1 ~ x/2-x^2/8 + O(x^3) |
| 54 |
#[cfg(any( |
| 55 |
all( |
| 56 |
any(target_arch = "x86", target_arch = "x86_64"), |
| 57 |
target_feature = "fma" |
| 58 |
), |
| 59 |
target_arch = "aarch64" |
| 60 |
))] |
| 61 |
{ |
| 62 |
use crate::common::f_fmlaf; |
| 63 |
return f_fmlaf(-0.25 * x, 0.25 * x, x * 0.5); |
| 64 |
} |
| 65 |
#[cfg(not(any( |
| 66 |
all( |
| 67 |
any(target_arch = "x86", target_arch = "x86_64"), |
| 68 |
target_feature = "fma" |
| 69 |
), |
| 70 |
target_arch = "aarch64" |
| 71 |
)))] |
| 72 |
{ |
| 73 |
use crate::common::f_fmla; |
| 74 |
let dx = x as f64; |
| 75 |
return f_fmla(-0.25 * dx, 0.25 * dx, dx * 0.5) as f32; |
| 76 |
} |
| 77 |
} |
| 78 |
|
| 79 |
return f32::NAN; // x == NaN |
| 80 |
} |
| 81 |
if (ix >> 31) != 0 && ux >= 0x7f00_0000 { |
| 82 |
// x < 0 and x >= 1 |
| 83 |
if ux == 0x7f00_0000u32 { |
| 84 |
// x == -1 |
| 85 |
return -1.; |
| 86 |
} |
| 87 |
return f32::NAN; |
| 88 |
} |
| 89 |
|
| 90 |
if ux <= 0x78eb_851eu32 { |
| 91 |
// |x| <= 0.015 |
| 92 |
|
| 93 |
// Polynomial generated by Sollya: |
| 94 |
// d = [-0.015, 0.015]; |
| 95 |
// sqrt1pm1 = sqrt(x + 1) - 1; |
| 96 |
// Q = fpminimax(sqrt1pm1, [|1,2,3,4,5,6|], [|D...|], d); |
| 97 |
const C: [u64; 6] = [ |
| 98 |
0x3fe0000000000034, |
| 99 |
0xbfc00000000002ee, |
| 100 |
0x3faffffffc0d541c, |
| 101 |
0xbfa3fffffa546ce5, |
| 102 |
0x3f9c016d2be05c25, |
| 103 |
0xbf95016d1ae9e058, |
| 104 |
]; |
| 105 |
let dx = x as f64; |
| 106 |
let p = f_polyeval6( |
| 107 |
dx, |
| 108 |
f64::from_bits(C[0]), |
| 109 |
f64::from_bits(C[1]), |
| 110 |
f64::from_bits(C[2]), |
| 111 |
f64::from_bits(C[3]), |
| 112 |
f64::from_bits(C[4]), |
| 113 |
f64::from_bits(C[5]), |
| 114 |
) * dx; |
| 115 |
return p as f32; |
| 116 |
} |
| 117 |
|
| 118 |
let dx = x as f64; |
| 119 |
((dx + 1.).sqrt() - 1.) as f32 |
| 120 |
} |
| 121 |
|
| 122 |
#[cfg(test)] |
| 123 |
mod tests { |
| 124 |
use super::*; |
| 125 |
|
| 126 |
#[test] |
| 127 |
fn test_sqrt1pm1f() { |
| 128 |
assert_eq!(f_sqrt1pm1f(-0.9), -0.6837722); |
| 129 |
assert_eq!(f_sqrt1pm1f(24.), 4.); |
| 130 |
assert_eq!(f_sqrt1pm1f(-0.75), -0.5); |
| 131 |
assert_eq!(f_sqrt1pm1f(8.), 2.); |
| 132 |
assert_eq!(f_sqrt1pm1f(9.), 2.1622777); |
| 133 |
assert_eq!(f_sqrt1pm1f(12.31), 2.6482873); |
| 134 |
assert_eq!(f_sqrt1pm1f(0.005231), 0.0026120886); |
| 135 |
assert_eq!(f_sqrt1pm1f(0.00000000005231), 2.6155e-11); |
| 136 |
assert_eq!(f_sqrt1pm1f(-0.00000000005231), -2.6155e-11); |
| 137 |
assert_eq!(f_sqrt1pm1f(0.), 0.); |
| 138 |
assert_eq!(f_sqrt1pm1f(-0.), 0.); |
| 139 |
assert_eq!(f_sqrt1pm1f(f32::INFINITY), f32::INFINITY); |
| 140 |
assert!(f_sqrt1pm1f(f32::NEG_INFINITY).is_nan()); |
| 141 |
assert!(f_sqrt1pm1f(f32::NAN).is_nan()); |
| 142 |
assert!(f_sqrt1pm1f(-1.0001).is_nan()); |
| 143 |
} |
| 144 |
} |