Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Document which methods on f64 are precise #118217

Merged
merged 1 commit into from
Feb 29, 2024
Merged

Document which methods on f64 are precise #118217

merged 1 commit into from
Feb 29, 2024

Conversation

tbu-
Copy link
Contributor

@tbu- tbu- commented Nov 23, 2023

No description provided.

@rustbot
Copy link
Collaborator

rustbot commented Nov 23, 2023

r? @m-ou-se

(rustbot has picked a reviewer for you, use r? to override)

@rustbot rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-libs Relevant to the library team, which will review and decide on the PR/issue. labels Nov 23, 2023
@tbu-
Copy link
Contributor Author

tbu- commented Nov 23, 2023

It has surprised me in the past that floating-point calculations executed on macOS can be different than the ones on Linux or Windows. This is why I wanted to clearly document which method's exact results can be relied upon and which cannot.

If this is desired, I'll do the same for f32.

@m-ou-se
Copy link
Member

m-ou-se commented Nov 28, 2023

@workingjubilee This seems like something you might have an opinion on?

@m-ou-se m-ou-se assigned Amanieu and unassigned m-ou-se Feb 14, 2024
Amanieu
Amanieu reviewed Feb 29, 2024
View reviewed changes
Copy link
Member

@Amanieu Amanieu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't think "precise" is the correct word to use here because it has a very specific meaning in floating-point arithmetic: it means that the float-encoded result is exactly identical to the mathematical result, which may sometimes not be true due to limited float precision.

A better word might be "consistent" to indicate that the result will be consistent across platforms.

library/std/src/f64.rs Outdated Show resolved Hide resolved
library/std/src/f64.rs Outdated Show resolved Hide resolved
library/std/src/f64.rs Outdated Show resolved Hide resolved
library/std/src/f64.rs Outdated Show resolved Hide resolved
@Amanieu Amanieu added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Feb 29, 2024
@tbu-
Copy link
Contributor Author

tbu- commented Feb 29, 2024

I don't think "precise" is the correct word to use here because it has a very specific meaning in floating-point arithmetic: it means that the float-encoded result is exactly identical to the mathematical result, which may sometimes not be true due to limited float precision.

The word precise is only used on the methods that produce results identical to the mathematical result: floor, ceil, round, round_ties_even, trunc, fract and abs.

@tbu-
Copy link
Contributor Author

tbu- commented Feb 29, 2024

@rustbot ready

@rustbot rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Feb 29, 2024
@Amanieu
Copy link
Member

Amanieu commented Feb 29, 2024

@bors r+ rollup

@bors
Copy link
Contributor

bors commented Feb 29, 2024

📌 Commit b5307f5 has been approved by Amanieu

It is now in the queue for this repository.

@bors bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Feb 29, 2024
tbu- added a commit to tbu-/rust that referenced this pull request Feb 29, 2024
@tbu-
Document which methods on f32 are precise
7400f22 
Same as rust-lang#118217 but for `f32`.
@tbu- tbu- mentioned this pull request Feb 29, 2024
Merged
@GuillaumeGomez GuillaumeGomez mentioned this pull request Feb 29, 2024
Merged
GuillaumeGomez added a commit to GuillaumeGomez/rust that referenced this pull request Feb 29, 2024
@GuillaumeGomez
Rollup merge of rust-lang#121793 - tbu-:pr_floating_point_32, r=Amanieu
bc23b84 
Document which methods on `f32` are precise

Same as rust-lang#118217 but for `f32`.
bors added a commit to rust-lang-ci/rust that referenced this pull request Feb 29, 2024
@bors
Auto merge of rust-lang#121800 - GuillaumeGomez:rollup-wob2qcz, r=Gui…
384d26f 
…llaumeGomez

Rollup of 10 pull requests

Successful merges:

 - rust-lang#118217 (Document which methods on `f64` are precise)
 - rust-lang#119748 (Increase visibility of `join_path` and `split_paths`)
 - rust-lang#121412 (platform docs: clarify hexagon-unknown-none-elf example, add hexagon-unknown-linux-musl)
 - rust-lang#121654 (Fix `async Fn` confirmation for `FnDef`/`FnPtr`/`Closure` types)
 - rust-lang#121700 (CFI: Don't compress user-defined builtin types)
 - rust-lang#121765 (add platform-specific function to get the error number for HermitOS)
 - rust-lang#121781 (bootstrap/format: send larger batches to rustfmt)
 - rust-lang#121788 (bootstrap: fix clap deprecated warnings)
 - rust-lang#121792 (Improve renaming suggestion when item starts with underscore)
 - rust-lang#121793 (Document which methods on `f32` are precise)

r? `@ghost`
`@rustbot` modify labels: rollup
rust-timer added a commit to rust-lang-ci/rust that referenced this pull request Feb 29, 2024
@rust-timer
Unrolled build for rust-lang#121793
7d4b4c3 
Rollup merge of rust-lang#121793 - tbu-:pr_floating_point_32, r=Amanieu

Document which methods on `f32` are precise

Same as rust-lang#118217 but for `f32`.
@bors bors merged commit 0c2cb39 into rust-lang:master Feb 29, 2024
11 checks passed
@rustbot rustbot added this to the 1.78.0 milestone Feb 29, 2024
rust-timer added a commit to rust-lang-ci/rust that referenced this pull request Feb 29, 2024
@rust-timer
Unrolled build for rust-lang#118217
bac88b2 
Rollup merge of rust-lang#118217 - tbu-:pr_floating_point, r=Amanieu

Document which methods on `f64` are precise
This was referenced May 2, 2024
Merged
Open
Merged
bors added a commit to rust-lang/miri that referenced this pull request May 2, 2024
@bors
Auto merge of #3535 - RalfJung:host-float, r=RalfJung
fe1424a 
update host-float comments

Turns out most of these do not have guaranteed precision anyway so it's fine to use host floats (see rust-lang/rust#121793 and rust-lang/rust#118217). The exception are sqrt and mul_add, tracked at #3534 and #2995.
fmease added a commit to fmease/rust that referenced this pull request May 3, 2024
@fmease
Rollup merge of rust-lang#124609 - RalfJung:float-precision, r=cuviper
9f44f55 
variable-precision float operations can differ depending on optimization levels

Follow-up to rust-lang#121793 and rust-lang#118217 that accounts for optimizations changing the precision of these functions.

Fixes rust-lang#109118
Fixes rust-lang#71355
fmease added a commit to fmease/rust that referenced this pull request May 3, 2024
@fmease
Rollup merge of rust-lang#124609 - RalfJung:float-precision, r=cuviper
ab8fda5 
variable-precision float operations can differ depending on optimization levels

Follow-up to rust-lang#121793 and rust-lang#118217 that accounts for optimizations changing the precision of these functions.

Fixes rust-lang#109118
Fixes rust-lang#71355
fmease added a commit to fmease/rust that referenced this pull request May 3, 2024
@fmease
Rollup merge of rust-lang#124609 - RalfJung:float-precision, r=cuviper
52055c6 
variable-precision float operations can differ depending on optimization levels

Follow-up to rust-lang#121793 and rust-lang#118217 that accounts for optimizations changing the precision of these functions.

Fixes rust-lang#109118
Fixes rust-lang#71355
matthiaskrgr added a commit to matthiaskrgr/rust that referenced this pull request May 3, 2024
@matthiaskrgr
Rollup merge of rust-lang#124609 - RalfJung:float-precision, r=cuviper
c412751 
variable-precision float operations can differ depending on optimization levels

Follow-up to rust-lang#121793 and rust-lang#118217 that accounts for optimizations changing the precision of these functions.

Fixes rust-lang#109118
Fixes rust-lang#71355
rust-timer added a commit to rust-lang-ci/rust that referenced this pull request May 3, 2024
@rust-timer
Unrolled build for rust-lang#124609
17a3434 
Rollup merge of rust-lang#124609 - RalfJung:float-precision, r=cuviper

variable-precision float operations can differ depending on optimization levels

Follow-up to rust-lang#121793 and rust-lang#118217 that accounts for optimizations changing the precision of these functions.

Fixes rust-lang#109118
Fixes rust-lang#71355
RalfJung pushed a commit to RalfJung/rust that referenced this pull request May 4, 2024
@bors
Auto merge of rust-lang#3535 - RalfJung:host-float, r=RalfJung
f6c0090 
update host-float comments

Turns out most of these do not have guaranteed precision anyway so it's fine to use host floats (see rust-lang#121793 and rust-lang#118217). The exception are sqrt and mul_add, tracked at rust-lang/miri#3534 and rust-lang/miri#2995.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@Amanieu Amanieu Amanieu left review comments

Assignees

@Amanieu Amanieu

Labels
S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. T-libs Relevant to the library team, which will review and decide on the PR/issue.
Projects
None yet
Milestone
1.78.0
Development

Successfully merging this pull request may close these issues.
5 participants
@tbu- @rustbot @m-ou-se @Amanieu @bors