try: remove define_opaque from unix split_paths

[CAD97]

#158936 had a regression in the serde benchmark doc profile. The only hypothesis I have as to how is that potentially the usage of #[define_opaque] is slow. This manually defines the #[define_opaque] in std::sys::paths::unix to test that hypothesis.

r? @JonathanBrouwer

[rustbot]

JonathanBrouwer is currently at their maximum review capacity.
They may take a while to respond.

[JonathanBrouwer]

@bors try @rust-timer queue

[rust-log-analyzer]

The job test-pr-check-2 failed! Check out the build log: (web) (plain enhanced) (plain)

Click to see the possible cause of the failure (guessed by this bot)

[RUSTC-TIMING] object test:false 8.903
error: unneeded `return` statement
  --> library/std/src/sys/paths/unix.rs:83:13
   |
83 |             return Some(Path::new(OsStr::from_bytes(split.0)));
   |             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   |
   = help: for further information visit https://rust-lang.github.io/rust-clippy/main/index.html#needless_return
   = note: requested on the command line with `-D clippy::needless-return`
help: remove `return`
   |
83 -             return Some(Path::new(OsStr::from_bytes(split.0)));
83 +             Some(Path::new(OsStr::from_bytes(split.0)))
   |

error: unneeded `return` statement
  --> library/std/src/sys/paths/unix.rs:86:13
   |

[CAD97]

To fix the clippy::needless-return error:

Suggested change

	return Some(Path::new(OsStr::from_bytes(split.0)));
	} else {
	self.unparsed = None;
	return Some(Path::new(OsStr::from_bytes(unparsed)));
	Some(Path::new(OsStr::from_bytes(split.0)));
	} else {
	self.unparsed = None;
	Some(Path::new(OsStr::from_bytes(unparsed)));

Not committing that as I don't know whether the try build cares

View changes since the review

[rust-bors]

☀️ Try build successful (CI)
Build commit: b82ccf1 (b82ccf19f74e8f5e8773a1a9e84c7c2a90ab7474)
Base parent: 21b707e (21b707e3f97e0b522ebd2f277a862339625ad83f)

[rust-timer]

Finished benchmarking commit (b82ccf1): comparison URL.

Overall result: ❌✅ regressions and improvements - no action needed

Benchmarking means the PR may be perf-sensitive. Consider adding rollup=never if this change is not fit for rolling up.

@rustbot label: -S-waiting-on-perf -perf-regression

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.2%	[0.2%, 0.2%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.3%	[-0.4%, -0.3%]	2
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary 3.0%, secondary 0.5%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	3.0%	[3.0%, 3.0%]	1
Regressions ❌ (secondary)	3.8%	[3.8%, 3.8%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.9%	[-2.9%, -2.9%]	1
All ❌✅ (primary)	3.0%	[3.0%, 3.0%]	1

Cycles

Results (primary -3.2%, secondary -5.0%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	26.6%	[26.6%, 26.6%]	1
Improvements ✅ (primary)	-3.2%	[-3.2%, -3.2%]	1
Improvements ✅ (secondary)	-20.7%	[-21.4%, -20.0%]	2
All ❌✅ (primary)	-3.2%	[-3.2%, -3.2%]	1

Binary size

Results (primary -0.1%, secondary -0.1%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.1%	[-0.2%, -0.0%]	88
Improvements ✅ (secondary)	-0.1%	[-0.2%, -0.0%]	57
All ❌✅ (primary)	-0.1%	[-0.2%, -0.0%]	88

Bootstrap: 487.961s -> 489.948s (0.41%)
Artifact size: 406.51 MiB -> 406.41 MiB (-0.02%)

[CAD97]

neutral result on different benchmarks