[perf] Try to skip some early lints with --cap-lints

[Kobzol]

Opening for perf.

r? @ghost

[Kobzol]

@bors2 try @rust-timer queue

[rust-bors]

⌛ Trying commit 6bc7c78 with merge aaf546c…

To cancel the try build, run the command @bors2 try cancel.

[Kobzol]

@bors2 try

[rust-bors]

⌛ Trying commit 951e697 with merge 702ebc3…

(The previously running try build was automatically cancelled.)

To cancel the try build, run the command @bors2 try cancel.

[rust-bors]

☀️ Try build successful (CI)
Build commit: 702ebc3 (702ebc3c678c9326b6cb9ac784ddfc1c715293b9, parent: 42245d34d22ade32b3f276dcf74deb826841594c)

[rust-timer]

Finished benchmarking commit (702ebc3): 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 (primary 1.4%, secondary 1.8%)

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

	mean	range	count
Regressions ❌ (primary)	1.4%	[1.0%, 1.9%]	2
Regressions ❌ (secondary)	1.8%	[1.5%, 2.3%]	11
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.4%	[1.0%, 1.9%]	2

Cycles

Results (secondary 6.8%)

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)	6.8%	[6.8%, 6.8%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Binary size

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

Bootstrap: 688.878s -> 688.462s (-0.06%)
Artifact size: 371.92 MiB -> 372.06 MiB (0.04%)

[Kobzol]

Hmm, I didn't expect any perf. wins on rustbot, because it doesn't measure dependencies, but I don't see any wins even locally :/

[Kobzol]

@bors2 try

[rust-bors]

⌛ Trying commit 8b9d949 with merge 95fcbf6…

To cancel the try build, run the command @bors2 try cancel.

[rust-bors]

☀️ Try build successful (CI)
Build commit: 95fcbf6 (95fcbf6a25feab8b420a47931c73fa42f1f47b43, parent: 99b18d6c5062449db8e7ccded4cb69b555a239c3)

[Kobzol]

No, still nothing. I wonder if these 4 excluded lints are the most expensive ones 🤔

[lqd]

My recollection from a couple years ago when I looked at this is that linting dependencies in our benchmarks were <100ms or so. Do you see something different now?

[Kobzol]

I benchmarked this a few months ago, late lints were super fast, but the early ones seemed to be a bit more expensive. I don't remember the details though, I'll need to re-check. I also benchmarked it locally, not with PGO, which is always a bit misleading.

[Kobzol]

Hmm, it seems like the early lints are "quick" now. Maybe I benchmarked it wrongly before.