Visit only statements in always live locals

[tmiasko]

No functional changes intended.

[rust-highfive]

r? @matthewjasper

(rust-highfive has picked a reviewer for you, use r? to override)

[tmiasko]

@bors rollup=never

[tmiasko]

@bors try @rust-timer queue

[rust-timer]

Awaiting bors try build completion.

[bors]

⌛ Trying commit 5686593 with merge 8baba0d590839b6c9f17ffc12b15300e5f1ef43a...

[bors]

☀️ Try build successful - checks-actions
Build commit: 8baba0d590839b6c9f17ffc12b15300e5f1ef43a (8baba0d590839b6c9f17ffc12b15300e5f1ef43a)

[rust-timer]

Queued 8baba0d590839b6c9f17ffc12b15300e5f1ef43a with parent 613ef74, future comparison URL.

@rustbot label: +S-waiting-on-perf

[rust-timer]

Finished benchmarking try commit (8baba0d590839b6c9f17ffc12b15300e5f1ef43a): comparison url.

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. Please note that if the perf results are neutral, you should likely undo the rollup=never given below by specifying rollup- to bors.

Importantly, though, if the results of this run are non-neutral do not roll this PR up -- it will mask other regressions or improvements in the roll up.

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

[matthewjasper]

@bors r+

[bors]

📌 Commit 5686593 has been approved by matthewjasper

[bors]

⌛ Testing commit 5686593 with merge 099f27b...

[bors]

☀️ Test successful - checks-actions
Approved by: matthewjasper
Pushing 099f27b to master...

[tmiasko]

This PR was benchmarked once before merging and then again after, the results are quite telling in terms of measurement errors associated with some benchmarks (especially that PR itself is rather insignificant):

• https://perf.rust-lang.org/compare.html?start=613ef740f3f37702728c6324f948d0abd1e9c82b&end=8baba0d590839b6c9f17ffc12b15300e5f1ef43a
• https://perf.rust-lang.org/compare.html?start=c6bc46227ab57a844fc7a9ed3a6c9efb35c725a9&end=099f27b6cbf10e31d41e4a2e6cb55ec39dbb0527&stat=instructions:u

It was my general impression that keccak instruction count results shifts between two relatively stable positions, in otherwise unpredictable manner. FYI @rust-lang/wg-compiler-performance