[rustdoc] Cleanup "highlight::end_expansion"

[yotamofek]

~Looks like a 5% improvement on the highlight benchmark.
Obviously, highlighting is only a small part of rustdoc's runtime, so improvement won't be as large on rustc-perf (if there's even an improvement), but holding fingers for a nice gain.

Perf seems neutral, but IMHO this is a nice small cleanup regardless.

r? @GuillaumeGomez (& perf run please!)

[yotamofek]

This basically just removes two String allocations, instead just using the final output string as the write buffer (whilst also reserving the actual amount of memory needed)

[GuillaumeGomez]

I'm curious to see the impact too.

@bors try @rust-timer queue

[rust-bors]

☀️ Try build successful (CI)
Build commit: ce6801f (ce6801fb79029c882e41451398a9638c867123b5, parent: d4ae855111df8c7ee255bea4c112e74b7d72cf45)

[rust-timer]

Finished benchmarking commit (ce6801f): comparison URL.

Overall result: ❌ regressions - no action needed

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

@bors rollup=never
@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)	1.5%	[1.5%, 1.5%]	1
Regressions ❌ (secondary)	1.4%	[1.4%, 1.4%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.5%	[1.5%, 1.5%]	1

Max RSS (memory usage)

Results (secondary -2.9%)

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
Improvements ✅ (secondary)	-2.9%	[-2.9%, -2.9%]	1
All ❌✅ (primary)	-	-	0

Cycles

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

Binary size

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

Bootstrap: 470.999s -> 472.425s (0.30%)
Artifact size: 387.75 MiB -> 387.76 MiB (0.00%)

[yotamofek]

Regression is on "opt" profile test, so most likely just noise. Otherwise perf is neutral :(
I still kinda like this change though. Maybe without the part for pre-calculating how long the string is gonna be, that's just complexity for (apparently) no gain

[GuillaumeGomez]

Go ahead then, we can run another perf check once you pushed the change.

[yotamofek]

Go ahead then, we can run another perf check once you pushed the change.

Done.
Ended up also making another small change where we use Cow::to_mut to turn the Cow::Borrowed into a Cow::Owned, only if it needs to be written to. Allows us to get rid of a special-casing early-return.

[GuillaumeGomez]

Round 2!

@bors try @rust-timer queue

[rust-bors]

☀️ Try build successful (CI)
Build commit: 67e3f35 (67e3f3599b7ce31555960f8039aab973321fba35, parent: 94ecb52bbeeccc990545d4ebc9e9c35dcaf285e3)

[rust-timer]

Finished benchmarking commit (67e3f35): comparison URL.

Overall result: no relevant changes - no action needed

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

@bors rollup=never
@rustbot label: -S-waiting-on-perf -perf-regression

Instruction count

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

Max RSS (memory usage)

Results (primary -0.3%, secondary 2.3%)

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

	mean	range	count
Regressions ❌ (primary)	0.7%	[0.7%, 0.7%]	1
Regressions ❌ (secondary)	2.3%	[2.3%, 2.3%]	1
Improvements ✅ (primary)	-1.3%	[-1.3%, -1.3%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-0.3%	[-1.3%, 0.7%]	2

Cycles

Results (primary -2.6%)

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)	-2.6%	[-2.6%, -2.6%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-2.6%	[-2.6%, -2.6%]	1

Binary size

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

Bootstrap: 470.352s -> 470.242s (-0.02%)
Artifact size: 387.80 MiB -> 387.80 MiB (0.00%)

[yotamofek]

@GuillaumeGomez perf is neutral, WDYT about merging this in? :)

[GuillaumeGomez]

Code is simpler so definitely!

@bors r+ rollup

[bors]

📌 Commit 78c2e58 has been approved by GuillaumeGomez

It is now in the queue for this repository.