Optimize performance by inline in macro hygiene system

[xizheyin]

I inline some small method in rustc_span/src/hygiene.rs and so on.

[compiler-errors]

@bors2 try @rust-timer queue

[rust-bors]

⌛ Trying commit e233072 with merge 70a86f7…

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

[rust-bors]

☀️ Try build successful (CI)
Build commit: 70a86f7 (70a86f741181f85bf71587a6464a6364479c2806, parent: efd420c770bb179537c01063e98cb6990c439654)

[rust-timer]

Finished benchmarking commit (70a86f7): 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
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.1%	[-0.1%, -0.1%]	5
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary 5.1%, secondary 0.2%)

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

	mean	range	count
Regressions ❌ (primary)	5.1%	[5.1%, 5.1%]	1
Regressions ❌ (secondary)	2.6%	[2.6%, 2.6%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.3%	[-2.3%, -2.3%]	1
All ❌✅ (primary)	5.1%	[5.1%, 5.1%]	1

Cycles

Results (secondary 1.1%)

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)	2.7%	[2.2%, 3.3%]	2
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.2%	[-2.2%, -2.2%]	1
All ❌✅ (primary)	-	-	0

Binary size

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

Bootstrap: 468.582s -> 468.539s (-0.01%)
Artifact size: 374.67 MiB -> 374.67 MiB (0.00%)

[petrochenkov]

Did you check that the #[inline] annotations actually have effect?
I.e. without them the functions are not inlined on some hot path, and with them they are inlined.
Most of the marked functions are crate-private so LLVM already can inline them if it thinks it makes sense.

[xizheyin]

I tested the execution efficiency of the first commit locally, and it does show a slight improvement in execution time, but I'm not sure if all the functions marked by #[inline] are actually inlined. do you have a good idea?

[xizheyin]

@rustbot ready
The code in the master branch now works just fine. I removed the second commit, which did not result in a performance improvement. It seems that the improvement may only come from #[inline]. Do we need to run the bot again to test performance?

[petrochenkov]

Since we are adding inline to small functions, it probably makes sense to add it to HygieneData::with and with_session_globals as well.
We can run benchmarks after that.

[petrochenkov]

The PR description also needs to be updated.
@rustbot author

[rustbot]

Reminder, once the PR becomes ready for a review, use @rustbot ready.

[xizheyin]

@rustbot ready

[petrochenkov]

@bors2 try @rust-timer queue

[rust-bors]

⌛ Trying commit 60d6980 with merge dc79eb3…

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

[rust-bors]

☀️ Try build successful (CI)
Build commit: dc79eb3 (dc79eb38141d9436db26729956e443911d304dad, parent: fc5af1813307d25a84d633f21e2e53c9376eb547)

[rust-timer]

Finished benchmarking commit (dc79eb3): 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
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.1%	[-0.1%, -0.1%]	4
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (secondary -1.0%)

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)	-	-	0
Improvements ✅ (secondary)	-1.0%	[-1.0%, -1.0%]	1
All ❌✅ (primary)	-	-	0

Cycles

Results (secondary -3.9%)

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)	-	-	0
Improvements ✅ (secondary)	-3.9%	[-4.6%, -3.2%]	2
All ❌✅ (primary)	-	-	0

Binary size

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

Bootstrap: 468.258s -> 469.253s (0.21%)
Artifact size: 374.68 MiB -> 374.65 MiB (-0.01%)

[xizheyin]

The results look good, all of them are improved.

[petrochenkov]

@bors r+ rollup=maybe

[bors]

📌 Commit 60d6980 has been approved by petrochenkov

It is now in the queue for this repository.