Add unicode fast path to is_printable

[Noratrieb]

Before, it would enter the full expensive check even for normal ascii characters. Now, it skips the check for the ascii characters in 32..127. This range was checked manually from the current behavior.

I ran the tracing test suite in miri, and it was really slow. I looked at a profile, and miri spent most of the time in core::char::methods::escape_debug_ext, where half of that was dominated by core::unicode::printable::is_printable. So I optimized it here.

The tracing profile:

[rust-highfive]

Hey! It looks like you've submitted a new PR for the library teams!

If this PR contains changes to any rust-lang/rust public library APIs then please comment with r? rust-lang/libs-api @rustbot label +T-libs-api -T-libs to request review from a libs-api team reviewer. If you're unsure where your change falls no worries, just leave it as is and the reviewer will take a look and make a decision to forward on if necessary.

Examples of T-libs-api changes:

• Stabilizing library features
• Introducing insta-stable changes such as new implementations of existing stable traits on existing stable types
• Introducing new or changing existing unstable library APIs (excluding permanently unstable features / features without a tracking issue)
• Changing public documentation in ways that create new stability guarantees
• Changing observable runtime behavior of library APIs

[rust-highfive]

r? @joshtriplett

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

[Noratrieb]

I've added a new benchmark and ran it and an existing one locally, the results are promising. (write_str_macro_debug writes a single unicode character, and write_str_macro_debug_ascii writes several ascii characters to a String.)

It appears to be slightly slower for non-ascii, and a lot faster for ascii.

-- unicode-is-printable-fastpath run 1
test fmt::write_str_macro_debug                                    ... bench:     472,941 ns/iter (+/- 10,715)
test fmt::write_str_macro_debug_ascii                              ... bench:     213,816 ns/iter (+/- 17,440)

-- master run 1
test fmt::write_str_macro_debug                                    ... bench:     470,989 ns/iter (+/- 30,310)
test fmt::write_str_macro_debug_ascii                              ... bench:     352,146 ns/iter (+/- 18,440)

-- unicode-is-printable-fastpath run 2
test fmt::write_str_macro_debug                                    ... bench:     473,699 ns/iter (+/- 20,190)
test fmt::write_str_macro_debug_ascii                              ... bench:     213,746 ns/iter (+/- 11,796)

-- master run 2
test fmt::write_str_macro_debug                                    ... bench:     470,464 ns/iter (+/- 16,985)
test fmt::write_str_macro_debug_ascii                              ... bench:     348,037 ns/iter (+/- 14,689)

[joshtriplett]

Since the conditional is needed anyway, would it be worth returning false in the < 32 case, to include that in the fast path?

[Noratrieb]

Huh, I changed this locally but forgot to commit it.

[joshtriplett]

@bors r+ rollup=never

[bors]

📌 Commit 3358a41 has been approved by joshtriplett

[joshtriplett]

@bors rollup=never

[bors]

⌛ Testing commit 3358a41 with merge dcbd5f5...

[bors]

☀️ Test successful - checks-actions
Approved by: joshtriplett
Pushing dcbd5f5 to master...

[rust-timer]

Finished benchmarking commit (dcbd5f5): comparison url.

Instruction count

• Primary benchmarks: no relevant changes found
• Secondary benchmarks: 🎉 relevant improvements found

	mean1	max	count2
Regressions 😿 (primary)	N/A	N/A	0
Regressions 😿 (secondary)	N/A	N/A	0
Improvements 🎉 (primary)	N/A	N/A	0
Improvements 🎉 (secondary)	-0.1%	-0.1%	2
All 😿🎉 (primary)	N/A	N/A	0

Max RSS (memory usage)

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

Cycles

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

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

@rustbot label: -perf-regression

Footnotes

1. the arithmetic mean of the percent change ↩

2. number of relevant changes ↩