GVN: Elide more intermediate transmutes

[scottmcm]

We already skipped intermediate steps like u32 or i32 that support any (initialized) value.

This extends that to also allow skipping intermediate steps whose values are a superset of either the source or destination type. Most importantly, that means that usize → NonZeroUsize → ptr::Alignment and ptr::Alignment → NonZeroUsize → usize can skip the middle because NonZeroUsize is a superset of Alignment.

Then Alignment::as_usize is updated to take advantage of that and let us remove some more locals in a few places.

r? cjgillot

[rustbot]

Some changes occurred to MIR optimizations

cc @rust-lang/wg-mir-opt

[scottmcm]

I'm not expecting much, but just in case:
@bors try @rust-timer queue

[rust-bors]

☀️ Try build successful (CI)
Build commit: cd2b201 (cd2b201cee18477fa23de21b4a5bbee37815f3dd, parent: 75963ce795666bc1f961e5d60060809809f6bc68)

[rust-timer]

Finished benchmarking commit (cd2b201): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

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.

Next Steps: If you can justify the regressions found in this try perf run, please do so in sufficient writing along with @rustbot label: +perf-regression-triaged. If not, please fix the regressions and do another perf run. If its results are neutral or positive, the label will be automatically removed.

@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)	0.3%	[0.3%, 0.3%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.5%	[-0.8%, -0.2%]	2
Improvements ✅ (secondary)	-0.1%	[-0.1%, -0.1%]	1
All ❌✅ (primary)	-0.2%	[-0.8%, 0.3%]	3

Max RSS (memory usage)

Results (primary -4.0%)

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)	-	-	0
Improvements ✅ (primary)	-4.0%	[-5.7%, -2.2%]	3
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-4.0%	[-5.7%, -2.2%]	3

Cycles

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

Binary size

Results (primary -0.0%, secondary -0.0%)

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

	mean	range	count
Regressions ❌ (primary)	0.1%	[0.0%, 0.5%]	11
Regressions ❌ (secondary)	0.0%	[0.0%, 0.0%]	1
Improvements ✅ (primary)	-0.1%	[-1.1%, -0.0%]	13
Improvements ✅ (secondary)	-0.0%	[-0.1%, -0.0%]	6
All ❌✅ (primary)	-0.0%	[-1.1%, 0.5%]	24

Bootstrap: 471.746s -> 470.956s (-0.17%)
Artifact size: 383.60 MiB -> 383.53 MiB (-0.02%)

[cjgillot]

Can this change be done in its own commit? This would make easier to understand what changes from the mir opt diff.

[scottmcm]

Absolutely. I split it to three:

1. Just the compiler change to the mir-opt pass
2. The change to Alignment::as_usize on its own
3. The field change in Alignment

That way all the library changes are separated.

[cjgillot]

Do you mind commenting the logic? This looks good to me, but it took me a few frowns to understand, for instance why ranges are compared the way you wrote it

[rustbot]

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.