[Experiment] Eliminate possible Vec::push branches

[c410-f3r]

Related to #105156. Requesting a perf run.

pub fn push(v: &mut Vec<u8>) {
    let _ = v.reserve(4);
    v.push(1);
    v.push(2);
    v.push(3);
    v.push(4);
}

AFAICT, the codegen backend should infer the infallibility of these pushs but unfortunately with LLVM 18 we still have unnecessary reserve_for_push branches.

For the sake of curiosity, assert_unchecked was included in push to see any potential impact of such change. Take a look at the generated assembly at https://godbolt.org/z/b5jjPhsf8.

AFAICT (again), the assumption of more available capacity for each push is not valid for all situations.

[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

[Noratrieb]

fixed it for you, it should work now
@bors try @rust-timer queue

[bors]

⌛ Trying commit e64d4ba with merge ba16851...

[bors]

☀️ Try build successful - checks-actions
Build commit: ba16851 (ba168511861d72f06ae72268ed19ac3d62f126b2)

[rust-timer]

Finished benchmarking commit (ba16851): comparison URL.

Overall result: ❌✅ regressions and improvements - ACTION NEEDED

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please fix the regressions and do another perf run. If the next run shows neutral or positive results, the label will be automatically removed.

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

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.7%	[0.4%, 1.3%]	13
Regressions ❌ (secondary)	0.3%	[0.2%, 0.4%]	3
Improvements ✅ (primary)	-0.3%	[-0.4%, -0.2%]	13
Improvements ✅ (secondary)	-1.0%	[-1.9%, -0.3%]	17
All ❌✅ (primary)	0.2%	[-0.4%, 1.3%]	26

Max RSS (memory usage)

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	4.7%	[0.1%, 12.7%]	6
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-3.7%	[-7.7%, -0.7%]	3
Improvements ✅ (secondary)	-2.5%	[-2.5%, -2.5%]	1
All ❌✅ (primary)	1.9%	[-7.7%, 12.7%]	9

Cycles

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	1.0%	[0.8%, 1.2%]	2
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.0%	[0.8%, 1.2%]	2

Binary size

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.3%	[0.0%, 2.2%]	58
Regressions ❌ (secondary)	0.7%	[0.2%, 2.5%]	5
Improvements ✅ (primary)	-0.1%	[-0.2%, -0.0%]	5
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.3%	[-0.2%, 2.2%]	63

Bootstrap: 641.758s -> 640.665s (-0.17%)
Artifact size: 308.80 MiB -> 308.77 MiB (-0.01%)

[c410-f3r]

Thanks @Nilstrieb

[cynecx]

@c410-f3r

Just as a note: LLVM's constraint elimination can work better when this is written as:

// rustc emits: `add nuw`, so the pass can assume that the instruction does not overflow
self.len = unsafe { self.len.unchecked_add(1) };

This reduces some calls to reserve_for_push but does not eliminate all of them. I haven't really had the time to dig in further to find out...