[workingjubilee]

This PR slides a bunch of crate versions forward until suddenly a bunch of deps fall out of the tree!
In doing so this mostly picks up a version bump in the redox_users crate which makes most of the features default to optional.

crossbeam-utils 0.7 => 0.8.3 (where applicable)
https://github.com/crossbeam-rs/crossbeam/blob/master/crossbeam-utils/CHANGELOG.md
directories 3.0.1 => 3.0.2
ignore 0.4.16 => 0.4.17
tempfile 3.0.5 => tempfile 3.2

Removes constant_time_eq from deps exceptions
Removes arrayref from deps exceptions
And also removes:

• blake2b_simd
• const_fn (the package, not the feature)
• constant_time_eq
• redox_users 0.3.4
• rust-argon2

[rust-highfive]

r? @Mark-Simulacrum

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

[Mark-Simulacrum]

Generally it's best to split these up in case there's any regressions, and provide links to bumped dependency changelogs if possible. But let's try a perf run to start:

@bors try @rust-timer queue

[rust-timer]

Awaiting bors try build completion.

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

[bors]

⌛ Trying commit d6c9dcc1a394990c78da209a76ee07cdd7d8df1d with merge 196e018e74cf08fbb02b8040a535e431fb9ff460...

[workingjubilee]

Alas, crossbeam has a changelog but none of the others do and instead I went crawling individual commits to find this out! :D
e.g. dirs-dev/dirs-sys-rs@e776c08

[workingjubilee]

Huh, that's weird, I thought I rebased that already. Sorry!

[workingjubilee]

My latest commit assumes the instructions given by the logbot are correct and therefore adds rand_core 0.5.1 to rustc-workspace-hack with the superset of requested features. Apologies if it is not, I have not encountered this flavor of error before and I am not entirely sure of how to proceed.

[JohnTitor]

@bors try @rust-timer queue

[rust-timer]

Awaiting bors try build completion.

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

[bors]

⌛ Trying commit 7d3186d with merge c05d0daf28bf4ee17b5aeeb79a80e57e74d37139...

[bors]

☀️ Try build successful - checks-actions
Build commit: c05d0daf28bf4ee17b5aeeb79a80e57e74d37139 (c05d0daf28bf4ee17b5aeeb79a80e57e74d37139)

[rust-timer]

Queued c05d0daf28bf4ee17b5aeeb79a80e57e74d37139 with parent bb491ed, future comparison URL.

[rust-timer]

Finished benchmarking try commit (c05d0daf28bf4ee17b5aeeb79a80e57e74d37139): 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

[workingjubilee]

Nothing dramatic (indeed, almost entirely <1.0% and only on the best case), but almost no red and a lot of green. 💚

[workingjubilee]

Generally it's best to split these up in case there's any regressions, and provide links to bumped dependency changelogs if possible.

I could probably have left the tempfile change out, it's true.

The rest kind of seemed like it "goes together" because of the motion towards crossbeam-utils 0.8.x (and the bugfix in crossbeam-utils 0.8.2), and I discovered the sudden "dep cascade" only when poking around after the tempfile tweak. But I don't really understand the resolver from the user point of view. ^^;

[Mark-Simulacrum]

@bors r+ rollup=iffy

[bors]

📌 Commit 7d3186d has been approved by Mark-Simulacrum

[bors]

⌛ Testing commit 7d3186d with merge 537544b...

[bors]

☀️ Test successful - checks-actions
Approved by: Mark-Simulacrum
Pushing 537544b to master...