Mirrors/bevy
2
0
Fork 0
mirror of https://github.com/bevyengine/bevy synced 2025-01-09 11:48:56 +00:00
Code Issues Projects Releases Packages Wiki Activity
bevy/crates/bevy_math/src/ops.rs

472 lines
13 KiB
Rust
Raw Normal View History

Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
//! This mod re-exports the correct versions of floating-point operations with
//! unspecified precision in the standard library depending on whether the `libm`
//! crate feature is enabled.
//!
//! All the functions here are named according to their versions in the standard
//! library.
#![allow(dead_code)]
#![allow(clippy::disallowed_methods)]
// Note: There are some Rust methods with unspecified precision without a `libm`
// equivalent:
// - `f32::powi` (integer powers)
// - `f32::log` (logarithm with specified base)
// - `f32::abs_sub` (actually unsure if `libm` has this, but don't use it regardless)
//
// Additionally, the following nightly API functions are not presently integrated
// into this, but they would be candidates once standardized:
// - `f32::gamma`
// - `f32::ln_gamma`
#[cfg(not(feature = "libm"))]
mod std_ops {
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Raises a number to a floating point power.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn powf(x: f32, y: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::powf(x, y)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns `e^(self)`, (the exponential function).
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn exp(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::exp(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns `2^(self)`.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn exp2(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::exp2(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns the natural logarithm of the number.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn ln(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::ln(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns the base 2 logarithm of the number.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn log2(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::log2(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns the base 10 logarithm of the number.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn log10(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::log10(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns the cube root of a number.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn cbrt(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::cbrt(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Compute the distance between the origin and a point `(x, y)` on the Euclidean plane.
/// Equivalently, compute the length of the hypotenuse of a right-angle triangle with other sides having length `x.abs()` and `y.abs()`.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn hypot(x: f32, y: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::hypot(x, y)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the sine of a number (in radians).
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn sin(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::sin(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the cosine of a number (in radians).
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn cos(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::cos(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the tangent of a number (in radians).
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn tan(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::tan(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the arcsine of a number. Return value is in radians in
/// the range [-pi/2, pi/2] or NaN if the number is outside the range
/// [-1, 1].
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn asin(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::asin(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the arccosine of a number. Return value is in radians in
/// the range [0, pi] or NaN if the number is outside the range
/// [-1, 1].
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn acos(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::acos(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the arctangent of a number. Return value is in radians in the
/// range [-pi/2, pi/2];
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn atan(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::atan(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the four quadrant arctangent of `self` (`y`) and `other` (`x`) in radians.
///
/// * `x = 0`, `y = 0`: `0`
/// * `x >= 0`: `arctan(y/x)` -> `[-pi/2, pi/2]`
/// * `y >= 0`: `arctan(y/x) + pi` -> `(pi/2, pi]`
/// * `y < 0`: `arctan(y/x) - pi` -> `(-pi, -pi/2)`
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn atan2(x: f32, y: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::atan2(x, y)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Simultaneously computes the sine and cosine of the number, `x`. Returns
/// `(sin(x), cos(x))`.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn sin_cos(x: f32) -> (f32, f32) {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::sin_cos(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns `e^(self) - 1` in a way that is accurate even if the
/// number is close to zero.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn exp_m1(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::exp_m1(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns `ln(1+n)` (natural logarithm) more accurately than if
/// the operations were performed separately.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn ln_1p(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::ln_1p(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Hyperbolic sine function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn sinh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::sinh(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Hyperbolic cosine function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn cosh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::cosh(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Hyperbolic tangent function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn tanh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::tanh(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Inverse hyperbolic sine function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn asinh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::asinh(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Inverse hyperbolic cosine function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn acosh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::acosh(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Inverse hyperbolic tangent function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn atanh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
f32::atanh(x)
}
}
#[cfg(feature = "libm")]
mod libm_ops {
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Raises a number to a floating point power.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn powf(x: f32, y: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::powf(x, y)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns `e^(self)`, (the exponential function).
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn exp(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::expf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns `2^(self)`.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn exp2(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::exp2f(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns the natural logarithm of the number.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn ln(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
// This isn't documented in `libm` but this is actually the base e logarithm.
libm::logf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns the base 2 logarithm of the number.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn log2(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::log2f(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns the base 10 logarithm of the number.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn log10(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::log10f(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns the cube root of a number.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn cbrt(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::cbrtf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Compute the distance between the origin and a point `(x, y)` on the Euclidean plane.
Simpler lint fixes: makes `ci lints` work but disables a lint for now (#15376) Takes the first two commits from #15375 and adds suggestions from this comment: https://github.com/bevyengine/bevy/pull/15375#issuecomment-2366968300 See #15375 for more reasoning/motivation. ## Rebasing (rerunning) ```rust git switch simpler-lint-fixes git reset --hard main cargo fmt --all -- --unstable-features --config normalize_comments=true,imports_granularity=Crate cargo fmt --all git add --update git commit --message "rustfmt" cargo clippy --workspace --all-targets --all-features --fix cargo fmt --all -- --unstable-features --config normalize_comments=true,imports_granularity=Crate cargo fmt --all git add --update git commit --message "clippy" git cherry-pick e6c0b94f6795222310fb812fa5c4512661fc7887 ```
2024-09-24 11:42:59 +00:00
///
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Equivalently, compute the length of the hypotenuse of a right-angle triangle with other sides having length `x.abs()` and `y.abs()`.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn hypot(x: f32, y: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::hypotf(x, y)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the sine of a number (in radians).
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn sin(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::sinf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the cosine of a number (in radians).
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn cos(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::cosf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the tangent of a number (in radians).
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn tan(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::tanf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the arcsine of a number. Return value is in radians in
/// the range [-pi/2, pi/2] or NaN if the number is outside the range
/// [-1, 1].
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn asin(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::asinf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the arccosine of a number. Return value is in radians in
/// Hyperbolic tangent function.
///
/// Precision is specified when the `libm` feature is enabled.
/// the range [0, pi] or NaN if the number is outside the range
/// [-1, 1].
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn acos(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::acosf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the arctangent of a number. Return value is in radians in the
/// range [-pi/2, pi/2];
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn atan(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::atanf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Computes the four quadrant arctangent of `self` (`y`) and `other` (`x`) in radians.
///
/// * `x = 0`, `y = 0`: `0`
/// * `x >= 0`: `arctan(y/x)` -> `[-pi/2, pi/2]`
/// * `y >= 0`: `arctan(y/x) + pi` -> `(pi/2, pi]`
/// * `y < 0`: `arctan(y/x) - pi` -> `(-pi, -pi/2)`
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn atan2(x: f32, y: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::atan2f(x, y)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Simultaneously computes the sine and cosine of the number, `x`. Returns
/// `(sin(x), cos(x))`.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn sin_cos(x: f32) -> (f32, f32) {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::sincosf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns `e^(self) - 1` in a way that is accurate even if the
/// number is close to zero.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn exp_m1(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::expm1f(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Returns `ln(1+n)` (natural logarithm) more accurately than if
/// the operations were performed separately.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn ln_1p(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::log1pf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Hyperbolic sine function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn sinh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::sinhf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Hyperbolic cosine function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn cosh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::coshf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Hyperbolic tangent function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn tanh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::tanhf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Inverse hyperbolic sine function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn asinh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::asinhf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Inverse hyperbolic cosine function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn acosh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::acoshf(x)
}
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Inverse hyperbolic tangent function.
///
/// Precision is specified when the `libm` feature is enabled.
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[inline(always)]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub fn atanh(x: f32) -> f32 {
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
libm::atanhf(x)
}
}
#[cfg(feature = "libm")]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub use libm_ops::*;
Fix floating point math (#15239) # Objective - Fixes #15236 ## Solution - Use bevy_math::ops instead of std floating point operations. ## Testing - Did you test these changes? If so, how? Unit tests and `cargo run -p ci -- test` - How can other people (reviewers) test your changes? Is there anything specific they need to know? Execute `cargo run -p ci -- test` on Windows. - If relevant, what platforms did you test these changes on, and are there any important ones you can't test? Windows ## Migration Guide - Not a breaking change - Projects should use bevy math where applicable --------- Co-authored-by: Alice Cecile <alice.i.cecile@gmail.com> Co-authored-by: IQuick 143 <IQuick143cz@gmail.com> Co-authored-by: Joona Aalto <jondolf.dev@gmail.com>
2024-09-16 23:28:12 +00:00
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
#[cfg(not(feature = "libm"))]
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub use std_ops::*;
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
/// This extension trait covers shortfall in determinacy from the lack of a `libm` counterpart
/// to `f32::powi`. Use this for the common small exponents.
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
pub trait FloatPow {
/// Squares the f32
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
fn squared(self) -> Self;
Expose bevy math ops (#14863) # Objective - Fixes #14796 ## Solution - Copy docs for wrapper methods, make sure they are consistent with the original docs except for the section on precision.
2024-08-22 17:07:00 +00:00
/// Cubes the f32
Make bevy_math's `libm` feature use `libm` for all `f32`methods with unspecified precision (#14693) # Objective Closes #14474 Previously, the `libm` feature of bevy_math would just pass the same feature flag down to glam. However, bevy_math itself had many uses of floating-point arithmetic with unspecified precision. For example, `f32::sin_cos` and `f32::powi` have unspecified precision, which means that the exact details of their output are not guaranteed to be stable across different systems and/or versions of Rust. This means that users of bevy_math could observe slightly different behavior on different systems if these methods were used. The goal of this PR is to make it so that the `libm` feature flag actually guarantees some degree of determinacy within bevy_math itself by switching to the libm versions of these functions when the `libm` feature is enabled. ## Solution bevy_math now has an internal module `bevy_math::ops`, which re-exports either the standard versions of the operations or the libm versions depending on whether the `libm` feature is enabled. For example, `ops::sin` compiles to `f32::sin` without the `libm` feature and to `libm::sinf` with it. This approach has a small shortfall, which is that `f32::powi` (integer powers of floating point numbers) does not have an equivalent in `libm`. On the other hand, this method is only used for squaring and cubing numbers in bevy_math. Accordingly, this deficit is covered by the introduction of a trait `ops::FloatPow`: ```rust pub(crate) trait FloatPow { fn squared(self) -> Self; fn cubed(self) -> Self; } ``` Next, each current usage of the unspecified-precision methods has been replaced by its equivalent in `ops`, so that when `libm` is enabled, the libm version is used instead. The exception, of course, is that `.powi(2)`/`.powi(3)` have been replaced with `.squared()`/`.cubed()`. Finally, the usage of the plain `f32` methods with unspecified precision is now linted out of bevy_math (and hence disallowed in CI). For example, using `f32::sin` within bevy_math produces a warning that tells the user to use the `ops::sin` version instead. ## Testing Ran existing tests. It would be nice to check some benchmarks on NURBS things once #14677 merges. I'm happy to wait until then if the rest of this PR is fine. --- ## Discussion In the future, it might make sense to actually expose `bevy_math::ops` as public if any downstream Bevy crates want to provide similar determinacy guarantees. For now, it's all just `pub(crate)`. This PR also only covers `f32`. If we find ourselves using `f64` internally in parts of bevy_math for better robustness, we could extend the module and lints to cover the `f64` versions easily enough. I don't know how feasible it is, but it would also be nice if we could standardize the bevy_math tests with the `libm` feature in CI, since their success is currently platform-dependent (e.g. 8 of them fail on my machine when run locally). --------- Co-authored-by: IQuick 143 <IQuick143cz@gmail.com>
2024-08-12 16:13:36 +00:00
fn cubed(self) -> Self;
}
impl FloatPow for f32 {
#[inline]
fn squared(self) -> Self {
self * self
}
#[inline]
fn cubed(self) -> Self {
self * self * self
}
}
Reference in a new issue Copy permalink