[PERF TEST] Make TokenKind Copy

[nnethercote]

By leaking nonterminals we can make them &'static instead of
ref-counted, which is enough for this test.

r? @ghost

[nnethercote]

@bors try @rust-timer queue

[rust-timer]

Awaiting bors try build completion.

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

[bors]

⌛ Trying commit f1db01ceb09c3a88caadc9664a6d012347780dc1 with merge 7698e668bd4f07761451bd67ab45301a7e6ac82b...

[bors]

☀️ Try build successful - checks-actions
Build commit: 7698e668bd4f07761451bd67ab45301a7e6ac82b (7698e668bd4f07761451bd67ab45301a7e6ac82b)

[rust-timer]

Queued 7698e668bd4f07761451bd67ab45301a7e6ac82b with parent 9714e13, future comparison URL.

[rust-timer]

Finished benchmarking commit (7698e668bd4f07761451bd67ab45301a7e6ac82b): comparison url.

Summary:

• Primary benchmarks: 🎉 relevant improvements found
• Secondary benchmarks: mixed results

	Regressions 😿 (primary)	Regressions 😿 (secondary)	Improvements 🎉 (primary)	Improvements 🎉 (secondary)	All 😿 🎉 (primary)
count1	2	12	144	95	146
mean2	0.6%	0.7%	-0.8%	-1.7%	-0.8%
max	0.8%	0.8%	-3.0%	-13.0%	-3.0%

If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf.

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

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

Footnotes

1. number of relevant changes ↩

2. the arithmetic mean of the percent change ↩

[nnethercote]

This was just an experiment to discover the potential speedups from making Token into a Copy type. Conclusion: decent improvements, but not amazing.

[nnethercote]

Note to self: a follow-up discussion was had here about better ways to implement this. It's difficult.

[nnethercote]

@bors try @rust-timer queue

[bors]

⌛ Trying commit d94a57e with merge 6deb69be8d85236a858b12f1813e580b8ad425b7...

[bors]

☀️ Try build successful - checks-actions
Build commit: 6deb69be8d85236a858b12f1813e580b8ad425b7 (6deb69be8d85236a858b12f1813e580b8ad425b7)

[rust-timer]

Finished benchmarking commit (6deb69be8d85236a858b12f1813e580b8ad425b7): comparison URL.

Overall result: ✅ improvements - no action needed

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

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

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.8%	[-2.4%, -0.3%]	133
Improvements ✅ (secondary)	-2.5%	[-12.8%, -0.2%]	56
All ❌✅ (primary)	-0.8%	[-2.4%, -0.3%]	133

Max RSS (memory usage)

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	2.4%	[0.9%, 5.1%]	70
Regressions ❌ (secondary)	19.2%	[1.8%, 33.3%]	20
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.9%	[-2.9%, -2.9%]	1
All ❌✅ (primary)	2.4%	[0.9%, 5.1%]	70

Cycles

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-2.1%	[-4.2%, -1.0%]	15
Improvements ✅ (secondary)	-4.6%	[-16.2%, -1.9%]	39
All ❌✅ (primary)	-2.1%	[-4.2%, -1.0%]	15

Binary size

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

Bootstrap: 654.49s -> 649.326s (-0.79%)

[nnethercote]

Perf results were similar to earlier. Still not a feasible path forward, though. Getting rid of TokenKind::Interpolated is the real hope.

[Dylan-DPC]

Closing this as it was a perf test