• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1# How to contribute
2
3- Pick your favorite math function from the [issue tracker].
4- Look for the C implementation of the function in the [MUSL source code][src].
5- Copy paste the C code into a Rust file in the `src/math` directory and adjust
6  `src/math/mod.rs` accordingly. Also, uncomment the corresponding trait method
7  in `src/lib.rs`.
8- Write some simple tests in your module (using `#[test]`)
9- Run `cargo test` to make sure it works
10- Run `cargo test --features musl-reference-tests` to compare your
11  implementation against musl's
12- Send us a pull request! Make sure to run `cargo fmt` on your code before
13  sending the PR. Also include "closes #42" in the PR description to close the
14  corresponding issue.
15- :tada:
16
17[issue tracker]: https://github.com/rust-lang/libm/issues
18[src]: https://git.musl-libc.org/cgit/musl/tree/src/math
19[`src/math/truncf.rs`]: https://github.com/rust-lang/libm/blob/master/src/math/truncf.rs
20
21Check [PR #65] for an example.
22
23[PR #65]: https://github.com/rust-lang/libm/pull/65
24
25## Tips and tricks
26
27- *IMPORTANT* The code in this crate will end up being used in the `core` crate so it can **not**
28  have any external dependencies (other than `core` itself).
29
30- Only use relative imports within the `math` directory / module, e.g. `use self::fabs::fabs` or
31`use super::k_cos`. Absolute imports from core are OK, e.g. `use core::u64`.
32
33- To reinterpret a float as an integer use the `to_bits` method. The MUSL code uses the
34  `GET_FLOAT_WORD` macro, or a union, to do this operation.
35
36- To reinterpret an integer as a float use the `f32::from_bits` constructor. The MUSL code uses the
37  `SET_FLOAT_WORD` macro, or a union, to do this operation.
38
39- You may use other methods from core like `f64::is_nan`, etc. as appropriate.
40
41- If you're implementing one of the private double-underscore functions, take a look at the
42  "source" name in the comment at the top for an idea for alternate naming. For example, `__sin`
43  was renamed to `k_sin` after the FreeBSD source code naming. Do `use` these private functions in
44  `mod.rs`.
45
46- You may encounter weird literals like `0x1p127f` in the MUSL code. These are hexadecimal floating
47  point literals. Rust (the language) doesn't support these kind of literals. The best way I have
48  found to deal with these literals is to turn them into their integer representation using the
49  [`hexf!`] macro and then turn them back into floats. See below:
50
51[`hexf!`]: https://crates.io/crates/hexf
52
53``` rust
54// Step 1: write a program to convert the float into its integer representation
55#[macro_use]
56extern crate hexf;
57
58fn main() {
59    println!("{:#x}", hexf32!("0x1.0p127").to_bits());
60}
61```
62
63``` console
64$ # Step 2: run the program
65$ cargo run
660x7f000000
67```
68
69``` rust
70// Step 3: copy paste the output into libm
71let x1p127 = f32::from_bits(0x7f000000); // 0x1p127f === 2 ^ 12
72```
73
74- Rust code panics on arithmetic overflows when not optimized. You may need to use the [`Wrapping`]
75  newtype to avoid this problem.
76
77[`Wrapping`]: https://doc.rust-lang.org/std/num/struct.Wrapping.html
78
79## Testing
80
81Normal tests can be executed with:
82
83```
84cargo test
85```
86
87If you'd like to run tests with randomized inputs that get compared against musl
88itself, you'll need to be on a Linux system and then you can execute:
89
90```
91cargo test --features musl-reference-tests
92```
93
94Note that you may need to pass `--release` to Cargo if there are errors related
95to integer overflow.
96