llvm-project.git/libc/src/math, branch users/mingmingl-llvm/samplefdo-profile-format

[libc][math][c++23] Add sqrtbf16 math function (#156654)

2025-09-06T09:06:02+00:00

This PR adds sqrtbf16 higher math function for BFloat16 type along with
the tests.

---------

Signed-off-by: Krishna Pandey

[libc][math][c++23] Add fmodbf16 math function (#155575)

2025-09-03T16:50:07+00:00

This PR adds fmodbf16 basic math function for BFloat16 type along with
the tests.

---------

Signed-off-by: Krishna Pandey

[libc][math][c23] Implement C23 math function atanpif16 (#150400)

2025-09-02T19:50:50+00:00

This PR implements `atanpif16(x)` which computes
$\frac{\arctan(x)}{\pi}$ for half-precision floating-point numbers using
polynomial approximation with domain reduction.

## Mathematical Implementation

The implementation uses a 15th-degree Taylor polynomial expansion of
$\frac{\arctan(x)}{\pi}$ that's computed using
[`python-sympy`](https://www.sympy.org/en/index.html) and it's accurate
in $|x| \in [0, 0.5)$:

$$
g(x) = \frac{\arctan(x)}{\pi} \approx 
\begin{aligned}[t]
    & 0.318309886183791x \\
    & - 0.106103295394597x^3 \\
    & + 0.0636619772367581x^5 \\
    & - 0.0454728408833987x^7 \\
    & + 0.0353677651315323x^9 \\
    & - 0.0289372623803446x^{11} \\
    & + 0.0244853758602916x^{13} \\
    & - 0.0212206590789194x^{15} + O(x^{17})
\end{aligned}
$$


--- 

To ensure accuracy across all real inputs, the domain is divided into
three cases with appropriate transformations:

**Case 1: $|x| \leq 0.5$**  
Direct polynomial evaluation: 

$$\text{atanpi}(x) = \text{sign}(x) \cdot g(|x|)$$

**Case 2: $0.5 < |x| \leq 1$**  
Double-angle reduction using:

$$\arctan(x) = 2\arctan\left(\frac{x}{1 + \sqrt{1 + x^2}}\right)$$

$$\text{atanpi}(x) = \text{sign}(x) \cdot 2g\left(\frac{|x|}{1 + \sqrt{1
+ x^2}}\right)$$

**Case 3: $|x| > 1$**  
Reciprocal transformation using 

$$\arctan(x) = \frac{\pi}{2} - \arctan\left(\frac{1}{x}\right) \
\text{for} \ x \gt 0$$

$$\text{atanpi}(x) = \text{sign}(x) \cdot \left(\frac{1}{2} -
g\left(\frac{1}{|x|}\right)\right)$$


Closes #132212

[libc][math][c++23] Add scalb{,l}nbf16 math functions (#155569)

2025-08-28T14:59:03+00:00

This PR adds the following basic math functions for BFloat16 type along
with the tests:
- scalbnbf16
- scalblnbf16

---------

Signed-off-by: Krishna Pandey

[libc][math][c++23] Add totalorder{,mag}bf16 math functions (#155568)

2025-08-28T14:07:07+00:00

This PR adds the following basic math functions for BFloat16 type along
with the tests:
- totalorderbf16
- totalordermagbf16

---------

Signed-off-by: Krishna Pandey

[libc][math][c++23] Add {canonicalize,iscanonical,fdim,copysign,issignaling}bf16 math functions (#155567)

2025-08-27T17:01:17+00:00

This PR adds the following basic math functions for BFloat16 type along
with the tests:
- canonicalizebf16
- iscanonicalbf16
- fdimbf16
- copysignbf16
- issignalingbf16

---------

Signed-off-by: Krishna Pandey

[libc][math][c++23] Add {modf,remainder,remquo}bf16 math functions (#154652)

2025-08-27T06:12:09+00:00

This PR adds the following basic math functions for BFloat16 type along
with the tests:
- modfbf16
- remainderbf16
- remquobf16

---------

Signed-off-by: Krishna Pandey 
Co-authored-by: OverMighty

[libc][math][c++23] Add {frexp,ilogb,ldexp,llogb,logb}bf16 math functions (#154427)

2025-08-25T06:16:25+00:00

This PR adds the following basic math functions for BFloat16 type along
with the tests:
- frexpbf16
- ilobbf16
- ldexpbf16
- llogbbf16
- logbbf16

---------

Signed-off-by: Krishna Pandey

[libc][math] Do not use float16 basic operations in hypotf16. (#155177)

2025-08-25T00:01:00+00:00

Compiler runtimes for float16 basic operations might not be correctly
rounded for all rounding modes.

[libc][math][c++23] Add {nearbyint,rint,lrint,llrint,lround,llround}bf16 math functions (#153882)

2025-08-24T15:21:02+00:00

This PR adds the following basic math functions for BFloat16 type along
with the tests:
- nearbyintbf16
- rintbf16
- lrintbf16
- llrintbf16
- lroundbf16
- llroundbf16

---------

Signed-off-by: Krishna Pandey