Avoid getting dep_dep_node unnecessarily.

[nnethercote]

r? @ghost

[nnethercote]

@bors try @rust-timer queue

[hkBst]

Change seems obviously good, since it avoids a function call on both Some branches of the match, which only need the result for debugging information.

I have no idea of the potential performance impact of this, but I'm curious to see the results of the perf run.

View changes since this review

[rust-bors]

☀️ Try build successful (CI)
Build commit: 7b6fbc1 (7b6fbc1b2e10c887bd6f992a6f0525a844b1ac01, parent: 3b8665c5ab3aeced9b01672404c3764583e722ca)

[rust-timer]

Finished benchmarking commit (7b6fbc1): comparison URL.

Overall result: ✅ improvements - 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)	-	-	0
Regressions ❌ (secondary)	0.2%	[0.2%, 0.2%]	2
Improvements ✅ (primary)	-1.4%	[-3.5%, -0.2%]	110
Improvements ✅ (secondary)	-1.1%	[-2.6%, -0.2%]	57
All ❌✅ (primary)	-1.4%	[-3.5%, -0.2%]	110

Max RSS (memory usage)

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

Cycles

Results (primary -2.8%, secondary -2.7%)

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)	-2.8%	[-3.6%, -2.2%]	17
Improvements ✅ (secondary)	-2.7%	[-3.4%, -2.0%]	2
All ❌✅ (primary)	-2.8%	[-3.6%, -2.2%]	17

Binary size

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

Bootstrap: 472.69s -> 471.95s (-0.16%)
Artifact size: 387.73 MiB -> 387.81 MiB (0.02%)

[nnethercote]

The icount results were almost twice as good as what I saw locally. And there are even some green cycle count results, which is nice, and uncommon for a micro-optimization like this.

There are a few double-digit wall-time regressions, but only among secondary benchmarks, and only among the really short-running ones, which appear to be prone to large swings with some frequency; the change is so simple that I think they can be ignored.

r? @saethlin