alloc: specialize String::extend for slices of str

[lolbinarycat]

Here's a small optimization via specialization that can potentially eliminate a fair few reallocations when building strings using specific patterns, unsure if this justifies the use of unsafe, but I had the code implemented from #148604, so I thought it was worth submitting to see.

[rustbot]

r? @Mark-Simulacrum

rustbot has assigned @Mark-Simulacrum.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[Kobzol]

@bors try @rust-timer queue

[hanna-kruppe]

This needs to check for overflow, otherwise it's unsound to do unchecked ptr.adds and copy_nonoverlappings below.

[lolbinarycat]

You're right, I forgot about the case where multiple strs reuse the same memory by having their addresses overlap, and was operating under the assumption all arguments were disjoint, in which case overflow would be impossible.

What do you recommend doing in the overflow case? just panicing with a generic message? triggering the oom handler to preserve behavior (as that's what the generic implementation would do in the overflow case)?

[hanna-kruppe]

Maybe just saturate and let reserve panic from capacity overflow, just like the previous push_str calls would have? I don't know off-hand if there's any significant benefit to actual overflow check vs. saturation in this context.

[rust-bors]

☀️ Try build successful (CI)
Build commit: 281de2c (281de2c09f393bab648058208a34357b68b660d1, parent: 66428d92bec337ed4785d695d0127276a482278c)

[lolbinarycat]

Fixed CI failure and used saturating arithmetic

[rust-timer]

Finished benchmarking commit (281de2c): 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 (secondary -1.0%)

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)	1.7%	[0.9%, 2.5%]	2
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.7%	[-6.2%, -1.2%]	2
All ❌✅ (primary)	-	-	0

Cycles

Results (primary -6.4%, 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)	2.4%	[2.4%, 2.4%]	1
Improvements ✅ (primary)	-6.4%	[-6.4%, -6.4%]	1
Improvements ✅ (secondary)	-2.2%	[-2.2%, -2.2%]	1
All ❌✅ (primary)	-6.4%	[-6.4%, -6.4%]	1

Binary size

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

Bootstrap: 470.249s -> 470.338s (0.02%)
Artifact size: 386.85 MiB -> 386.82 MiB (-0.01%)