Add [f32]::sort_floats and [f64]::sort_floats

[joshtriplett]

It's inconvenient to sort a slice or Vec of floats, compared to sorting integers. To simplify numeric code, add a convenience method to [f32] and [f64] to sort them using sort_unstable_by with total_cmp.

[rust-highfive]

r? @kennytm

(rust-highfive has picked a reviewer for you, use r? to override)

[joshtriplett]

Discussed on Zulip at https://rust-lang.zulipchat.com/#narrow/stream/219381-t-libs/topic/float.20total_cmp .

[cuviper]

They'll need distinct new lang items, just like slice_u8 for its ascii methods.

[joshtriplett]

@cuviper Thanks, done.

[joshtriplett]

Marked as cfg(not(bootstrap)) since the bootstrap compiler doesn't have these lang items.

[joshtriplett]

...works better if I actually sort the sorted array in the right order...

[joshtriplett]

This looks like the mir-opt segfault hitting CI for various PRs right now (#93447); waiting for that to get sorted out before retrying.

[joshtriplett]

I've rebased this to eliminate the merge conflicts. Thanks to #94963 this got a lot simpler.

[joshtriplett]

I continue to believe that this doesn't hide NaN behavior any more than total_cmp itself does.

However, if people feel that a name like sort_floats_total would address this concern more explicitly, I can rename the method. Would that address the objections, or are there other concerns beyond that?

[pnkfelix]

r? rust-lang/compiler

[apiraino]

@rustbot r? rust/compiler

[petrochenkov]

Not compiler, @rustbot r? rust/libs-api

[petrochenkov]

@rustbot r? rust-lang/libs-api

[petrochenkov]

@rustbot r? rust-lang/library

[petrochenkov]

r? @rust-lang/libs-api

[petrochenkov]

(╯°□°)╯︵ ┻━┻
r? @Amanieu

[Amanieu]

I find these questionably useful, but I guess we can defer this discussion until stabilization.

@bors r+

[bors]

📌 Commit bded8fc has been approved by Amanieu

It is now in the queue for this repository.

[bors]

⌛ Testing commit bded8fc with merge ed793d8...

[bors]

☀️ Test successful - checks-actions
Approved by: Amanieu
Pushing ed793d8 to master...

[rust-timer]

Finished benchmarking commit (ed793d8): comparison url.

Instruction count

This benchmark run did not return any relevant results for this metric.

Max RSS (memory usage)

Results

• Primary benchmarks: no relevant changes found
• Secondary benchmarks: mixed results

	mean1	max	count2
Regressions 😿 (primary)	N/A	N/A	0
Regressions 😿 (secondary)	2.3%	2.3%	1
Improvements 🎉 (primary)	N/A	N/A	0
Improvements 🎉 (secondary)	-3.0%	-3.0%	1
All 😿🎉 (primary)	N/A	N/A	0

Cycles

Results

• Primary benchmarks: 😿 relevant regression found
• Secondary benchmarks: no relevant changes found

	mean1	max	count2
Regressions 😿 (primary)	3.4%	3.4%	1
Regressions 😿 (secondary)	N/A	N/A	0
Improvements 🎉 (primary)	N/A	N/A	0
Improvements 🎉 (secondary)	N/A	N/A	0
All 😿🎉 (primary)	3.4%	3.4%	1

If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf.

@rustbot label: -perf-regression

Footnotes

1. the arithmetic mean of the percent change ↩ ↩²

2. number of relevant changes ↩ ↩²