Ensure that ./x.py doc --stage 0 src/libstd works via CI

[ecstatic-morse]

This was split off from #71645, which recommends that users first try building libstd docs with the bootstrap rustdoc. This should work in most cases, but will fail if we start using a very recent rustdoc feature outside a #[cfg(not(bootstrap))].

It would be very nice to guarantee that ./x.py doc --stage 0 src/libstd works, since it allows documentation changes to be rendered locally without needing to build the compiler. However, it may put too big a burden on rustdoc developers who presumably want to dogfood new features.

[rust-highfive]

r? @pietroalbini

(rust_highfive has picked a reviewer for you, use r? to override)

[ecstatic-morse]

cc @rust-lang/rustdoc

[ecstatic-morse]

r? @GuillaumeGomez

[Mark-Simulacrum]

To be clear from a "implementation" perspective this looks good to me, but I agree it would be good to get rustdoc sign off as the team that's closest to docs today... basically, I think this means "new shiny features" would be less easy to use in std's docs, unless they're e.g. top-level attributes that can easily be cfg'd away.

[Mark-Simulacrum]

I'm going to go ahead and approve this, but if we run into trouble due to it and there's no nice fixes I would be open to reverting.

@bors r+

[bors]

📌 Commit 1f7c896 has been approved by Mark-Simulacrum

[bors]

⌛ Testing commit 1f7c896 with merge d88c74853164295717ffef1f8dc80efc39a63788...

[bors]

💔 Test failed - checks-azure

[ecstatic-morse]

Seemingly unrelated failure upon invoking the linker on MacOS: nul byte found in provided data. That's...weird. Doesn't seem like your typical spurious error to me. WDYT @Mark-Simulacrum?

https://dev.azure.com/rust-lang/rust/_build/results?buildId=28262&view=logs&j=1c64c3ff-4c37-56fe-a133-b407354bbfbf&t=07a412a6-d5a9-54c7-0ba5-234866c24f9d&l=29278

[Mark-Simulacrum]

Perhaps a race condition? Doesn't seem like it should/can be related to this PR though so let's @bors try

macOS linkers in general are pretty questionable so I wouldn't put it past this being spurious as well

[bors]

⌛ Trying commit 1f7c896 with merge 5d3f35f32f98377da91122864f55ad3a2607d09f...

[bors]

☀️ Try build successful - checks-azure
Build commit: 5d3f35f32f98377da91122864f55ad3a2607d09f (5d3f35f32f98377da91122864f55ad3a2607d09f)

[Mark-Simulacrum]

What I meant was @bors retry

[bors]

⌛ Trying commit 1f7c896 with merge 95d1050f4d7ae3b8fca4736c31b7afb598ca3032...

[Mark-Simulacrum]

@bors retry- r+

[ecstatic-morse]

@bors try-

[ecstatic-morse]

For some reason, the "try" build is still pending in the queue but is no longer marked as a "try" build after I did try-. I don't think retry- is a command.
@bors r-

[Mark-Simulacrum]

@bors r- retry

[Mark-Simulacrum]

@bors r+

Okay, I think this should do the trick...

[bors]

📌 Commit 1f7c896 has been approved by Mark-Simulacrum

[RalfJung]

Will this help ensure that ./x.py test src/libstd --stage 0 --doc works? This failed until the last bootstrap bump (#70770).

[Mark-Simulacrum]

Hm, yes, it should help I think -- may not be a 100% fix for that command though as internally the plumbing of arguments is a little different between doc and test IIRC

[bors]

⌛ Testing commit 1f7c896 with merge 4802f09...

[bors]

☀️ Test successful - checks-actions, checks-azure
Approved by: Mark-Simulacrum
Pushing 4802f09 to master...