Stream: t-compiler

Topic: #54818 weekly meeting 2018-10-11


pnkfelix (Oct 11 2018 at 13:52, on Zulip):

a heads up to @T-compiler , we'll be starting our weekly meeting here in about 8 minutes

pnkfelix (Oct 11 2018 at 13:52, on Zulip):

(through the magic of zulip plugins, the topic now links to our standard agenda)

nagisa (Oct 11 2018 at 13:54, on Zulip):

/me did not forget to turn on zulip this time around

Wesley Wiser (Oct 11 2018 at 14:00, on Zulip):

Hey everyone :wave:

pnkfelix (Oct 11 2018 at 14:03, on Zulip):

Alright, sorry I'm tardy

pnkfelix (Oct 11 2018 at 14:03, on Zulip):

I was trying to do a prepass over the agenda in the hopes of making hte meeting itself go more smoothly

pnkfelix (Oct 11 2018 at 14:04, on Zulip):

In case people are not aware, we moved the agenda link here: https://github.com/rust-lang/rust/issues/54818 right after the meeting last week

nagisa (Oct 11 2018 at 14:04, on Zulip):

/me did the same on stuff assigned to them

pnkfelix (Oct 11 2018 at 14:04, on Zulip):

So lets start: P-high

pnkfelix (Oct 11 2018 at 14:04, on Zulip):

"std-using paths work just fine in 2018 edition #![no_std]" #53166

pnkfelix (Oct 11 2018 at 14:05, on Zulip):

@Vadim Petrochenkov says that this will be fixed once #54671 lands

pnkfelix (Oct 11 2018 at 14:05, on Zulip):

we'll get to the question of that being beta-nominated when we go through that list

pnkfelix (Oct 11 2018 at 14:05, on Zulip):

next: "Compiler panic using 'static_assertions'" #54327

pnkfelix (Oct 11 2018 at 14:05, on Zulip):

did @eddyb or @nikomatsakis get a chance to look at this over the last week?

nikomatsakis (Oct 11 2018 at 14:06, on Zulip):

no, but I'm sure @eddyb is correct

pnkfelix (Oct 11 2018 at 14:06, on Zulip):

so, what shall we do: downgrade to P-medium and tagged as S-blocked ?

nikomatsakis (Oct 11 2018 at 14:07, on Zulip):

I guess so

oli (Oct 11 2018 at 14:07, on Zulip):

downgrade and blocked, yes, there's like 3 related issues to this same problem

pnkfelix (Oct 11 2018 at 14:07, on Zulip):

okay, next: "trait permitting extra members" #54665

pnkfelix (Oct 11 2018 at 14:08, on Zulip):

I trust that @Vadim Petrochenkov will address this

pnkfelix (Oct 11 2018 at 14:08, on Zulip):

does it need to be on the RC2 milestone?

nikomatsakis (Oct 11 2018 at 14:08, on Zulip):

it feels like it should be

nikomatsakis (Oct 11 2018 at 14:08, on Zulip):

pretty serious fwd compat risk

pnkfelix (Oct 11 2018 at 14:08, on Zulip):

okay

nikomatsakis (Oct 11 2018 at 14:09, on Zulip):

I could possibly push on it a bit more, but I got to parts of the code I didn't understand that well

nikomatsakis (Oct 11 2018 at 14:09, on Zulip):

I suspect @eddyb or @Vadim Petrochenkov would be better choices

pnkfelix (Oct 11 2018 at 14:09, on Zulip):

right, @Vadim Petrochenkov is already assigned and did say they would take care of it

nikomatsakis (Oct 11 2018 at 14:09, on Zulip):

should we just re-assign?

nikomatsakis (Oct 11 2018 at 14:09, on Zulip):

oh, ok

nikomatsakis (Oct 11 2018 at 14:09, on Zulip):

great

pnkfelix (Oct 11 2018 at 14:09, on Zulip):

so I'll wait for them to actually say they cannot

nikomatsakis (Oct 11 2018 at 14:09, on Zulip):

I'll unassign myself

pnkfelix (Oct 11 2018 at 14:09, on Zulip):

maybe you want to unassign yourself in meantime

pnkfelix (Oct 11 2018 at 14:09, on Zulip):

okay

pnkfelix (Oct 11 2018 at 14:10, on Zulip):

next: "The #[panic_handler] attribute can be applied to non-functions" #54896

oli (Oct 11 2018 at 14:11, on Zulip):

feels noncritical, but would be good to get fixed

pnkfelix (Oct 11 2018 at 14:11, on Zulip):

an odd thing

pnkfelix (Oct 11 2018 at 14:12, on Zulip):

when I'm trying it out, i get: "error[E0152]: duplicate lang item found: panic_impl."

oli (Oct 11 2018 at 14:12, on Zulip):

yea you need no_std I think

pnkfelix (Oct 11 2018 at 14:12, on Zulip):

I'm probably not testing it out in exactly the same manner as the bug filer

nikomatsakis (Oct 11 2018 at 14:12, on Zulip):

the fact that you get a compilation error — however bad — is something of a relief

pnkfelix (Oct 11 2018 at 14:13, on Zulip):

the bug filer claims they do not

oli (Oct 11 2018 at 14:13, on Zulip):

I can take this one if we want it fixed quickly

oli (Oct 11 2018 at 14:13, on Zulip):

otherwise I'll write mentoring instructions

pnkfelix (Oct 11 2018 at 14:13, on Zulip):

Well

pnkfelix (Oct 11 2018 at 14:13, on Zulip):

I mostly want to figure out

pnkfelix (Oct 11 2018 at 14:13, on Zulip):

whether this needs to be on RC2 or another milestone

pnkfelix (Oct 11 2018 at 14:13, on Zulip):

and for that I'd really like to have my hands on a test case

pnkfelix (Oct 11 2018 at 14:14, on Zulip):

my attempts to use #![no_std] in the playground have not yielded any meanginful difference from what I reported earlir

simulacrum (Oct 11 2018 at 14:15, on Zulip):

I think we can not have this on RC2

simulacrum (Oct 11 2018 at 14:15, on Zulip):

i.e. if it's done by edition that's fine

nikomatsakis (Oct 11 2018 at 14:15, on Zulip):

"can not"?

simulacrum (Oct 11 2018 at 14:15, on Zulip):

I don't think it's super bad

nikomatsakis (Oct 11 2018 at 14:15, on Zulip):

oh, like, it would be ok to not have this get fixed

simulacrum (Oct 11 2018 at 14:15, on Zulip):

yes

nagisa (Oct 11 2018 at 14:15, on Zulip):

I feel like attribute in other positions should jsut be ignored

nagisa (Oct 11 2018 at 14:15, on Zulip):

not necessarily even warned against

pnkfelix (Oct 11 2018 at 14:15, on Zulip):

lets put it on Release milestone

simulacrum (Oct 11 2018 at 14:15, on Zulip):

Probably P-medium, too?

pnkfelix (Oct 11 2018 at 14:16, on Zulip):

I don't see the reason to try to stress about resolving this by Oct 25

nagisa (Oct 11 2018 at 14:16, on Zulip):

I reviewed the implementation originally and I have no idea why it even triggers outside of functions

nikomatsakis (Oct 11 2018 at 14:16, on Zulip):

I prefer errors for attributes in wrong places (but often we just lint)

nikomatsakis (Oct 11 2018 at 14:16, on Zulip):

mostly because sometimes we want to extend the use of the attribute in that new context

davidtwco (Oct 11 2018 at 14:16, on Zulip):

If there isn't a rush on fixing this then I'd be happy to take a go at it.

pnkfelix (Oct 11 2018 at 14:16, on Zulip):

do others agree with @simulacrum that this can be P-medium ?

pnkfelix (Oct 11 2018 at 14:17, on Zulip):

Centril made it P-high, but that might have just been a hack to force us all to look at it first. :)

oli (Oct 11 2018 at 14:17, on Zulip):

medium sounds good

nagisa (Oct 11 2018 at 14:17, on Zulip):

Sure. The primary user of this feature (embedded people) are well behaved anyway, and will definitely not bite us for having a bug in a feature when it is used the way it wasn’t intended to be :slight_smile:

nikomatsakis (Oct 11 2018 at 14:18, on Zulip):

seems good

nikomatsakis (Oct 11 2018 at 14:18, on Zulip):

also :+1: to @davidtwco tackling — do you need a few tips? (Maybe @Oli can provide)

pnkfelix (Oct 11 2018 at 14:19, on Zulip):

next: "Underscore lifetimes are incorrectly accepted as lifetime bounds in impl headers" #54902

davidtwco (Oct 11 2018 at 14:19, on Zulip):

I'll start a topic for it.

nagisa (Oct 11 2018 at 14:19, on Zulip):

@davidtwco feel free to r? me for the fix.

nikomatsakis (Oct 11 2018 at 14:19, on Zulip):

Argh, we should fix this. I can mentor it.

nikomatsakis (Oct 11 2018 at 14:19, on Zulip):

and/or fix

pnkfelix (Oct 11 2018 at 14:19, on Zulip):

okay

pnkfelix (Oct 11 2018 at 14:20, on Zulip):

@nikomatsakis you think trying to mentor is okay if we want to hit RC2 ?

nikomatsakis (Oct 11 2018 at 14:20, on Zulip):

Depends on who gets mentored. I'll probably just open a fix.

nikomatsakis (Oct 11 2018 at 14:20, on Zulip):

I'm looking briefly at the code, I suspect it's just tweaking a line or two

pnkfelix (Oct 11 2018 at 14:21, on Zulip):

(i'm removing nominated tag)

pnkfelix (Oct 11 2018 at 14:21, on Zulip):

okay that's all the P-high stuff

pnkfelix (Oct 11 2018 at 14:21, on Zulip):

next, beta-nominations

pnkfelix (Oct 11 2018 at 14:21, on Zulip):

none of them are tagged T-compiler

pnkfelix (Oct 11 2018 at 14:22, on Zulip):

I do want to at least point at "resolve: Scale back hard-coded extern prelude additions on 2015 edition" #54671

pnkfelix (Oct 11 2018 at 14:22, on Zulip):

I am not certain whether the lang team will be the ones to approve a backport to beta there, or if they will delegate back to us

pnkfelix (Oct 11 2018 at 14:23, on Zulip):

skimming over the PR, it seems safe for backport to me

nikomatsakis (Oct 11 2018 at 14:23, on Zulip):

yeah

nikomatsakis (Oct 11 2018 at 14:23, on Zulip):

presuming it looks the same after @Vadim Petrochenkov rewrites it

pnkfelix (Oct 11 2018 at 14:23, on Zulip):

anyway, if any of you want to raise an objection about it, throw a comment on the PR

pnkfelix (Oct 11 2018 at 14:24, on Zulip):

no stable nominations

pnkfelix (Oct 11 2018 at 14:24, on Zulip):

next stable-to-beta regressions

pnkfelix (Oct 11 2018 at 14:24, on Zulip):

first: "ICE (stable segfault) in LLVM with creative FFI types" #54410

pnkfelix (Oct 11 2018 at 14:25, on Zulip):

I don't know if I'd call going from a segfault to an ICE exactly a regression ...

pnkfelix (Oct 11 2018 at 14:25, on Zulip):

sure, ideally we would have a nice diagnostic.

oli (Oct 11 2018 at 14:25, on Zulip):

there's a nice diagnostic happening, and then we ICE

pnkfelix (Oct 11 2018 at 14:26, on Zulip):

anyway, how should we prioritize this?

simulacrum (Oct 11 2018 at 14:26, on Zulip):

P-low, not a regression IMO

oli (Oct 11 2018 at 14:26, on Zulip):

probably just turn it into a delay_span_bug, I'll write mentoring instructions

pnkfelix (Oct 11 2018 at 14:26, on Zulip):

okay, thanks @Oli

pnkfelix (Oct 11 2018 at 14:28, on Zulip):

next: "[1.30 beta] Test suite of the jemalloc-ctl crate is failing" #54478

pnkfelix (Oct 11 2018 at 14:28, on Zulip):

there's some comments from @Alex Crichton explaining what happened here

pnkfelix (Oct 11 2018 at 14:29, on Zulip):

but the short answer is that rustdoc has for years turned on -C prefer_dynamic

pnkfelix (Oct 11 2018 at 14:29, on Zulip):

and other changes we have made cause that to have unreliable behavior with respect to global allocator registration

pnkfelix (Oct 11 2018 at 14:30, on Zulip):

The main suggested "fix" is to turn off the -C prefer-dynamic in rustdoc

pnkfelix (Oct 11 2018 at 14:30, on Zulip):

but we also are not sure whether such a fix would make sense to backport to beta

pnkfelix (Oct 11 2018 at 14:30, on Zulip):

(we're currently awaiting the results of a crater run for such a change)

pnkfelix (Oct 11 2018 at 14:31, on Zulip):

I guess my point in saying all this is: Feel free to weigh in on the Issue (#54478) or on PR #54939 if you have an opinion about this

Pietro Albini (Oct 11 2018 at 14:31, on Zulip):

(that's going to be a great crater run, it's already at 108% progress...)

pnkfelix (Oct 11 2018 at 14:32, on Zulip):

In particular, given the current trajectory, we may well just end up breaking the crates here

pnkfelix (Oct 11 2018 at 14:32, on Zulip):

because the risks of doing anything else are too great...

pnkfelix (Oct 11 2018 at 14:32, on Zulip):

so I'll just move along

pnkfelix (Oct 11 2018 at 14:32, on Zulip):

next: "[1.30 beta] Compiler hangs when compiling primal crate for armv7-apple-ios" #54627

pnkfelix (Oct 11 2018 at 14:33, on Zulip):

@nagisa already commented with a status report here

pnkfelix (Oct 11 2018 at 14:33, on Zulip):

@nagisa if it turns out you cannot address this during the weekend, try to send up a flag to the rest of us

pnkfelix (Oct 11 2018 at 14:34, on Zulip):

but realistically, this isn't going to get fixed by us, right?

pnkfelix (Oct 11 2018 at 14:34, on Zulip):

unless we come up with a local LLVM patch that we're willing to backport, tout suite ?

nagisa (Oct 11 2018 at 14:34, on Zulip):

correct

pnkfelix (Oct 11 2018 at 14:35, on Zulip):

So... the Milestone of 1.30 is just not realistic then, right?

nagisa (Oct 11 2018 at 14:35, on Zulip):

well, it is possible that we might be able to work around this by generating LLVM IR some other way to not trigger the problematic path in LLVM

pnkfelix (Oct 11 2018 at 14:35, on Zulip):

@Pietro Albini you just added that to that milestone!

nagisa (Oct 11 2018 at 14:35, on Zulip):

but that sounds unworkable to me

Pietro Albini (Oct 11 2018 at 14:35, on Zulip):

just a note, please assign all the beta regressions to the project (and ideally the milestone) for that release

pnkfelix (Oct 11 2018 at 14:35, on Zulip):

Hmm okay

Pietro Albini (Oct 11 2018 at 14:35, on Zulip):

@pnkfelix oh, wasn't looking at the chat :/

pnkfelix (Oct 11 2018 at 14:35, on Zulip):

@Pietro Albini but what's the point if we aren't going to hit it?

pnkfelix (Oct 11 2018 at 14:36, on Zulip):

oh, you said "ideally"

pnkfelix (Oct 11 2018 at 14:36, on Zulip):

I guess that's where we have the leeway to not have it assigned to the milestone? If we determine that its not feasible to resolve it by that time?

Pietro Albini (Oct 11 2018 at 14:36, on Zulip):

@pnkfelix yeah

pnkfelix (Oct 11 2018 at 14:36, on Zulip):

Okay lets maybe not take up meeting time now debating this matter

pnkfelix (Oct 11 2018 at 14:37, on Zulip):

@nagisa , if you look at this and confirm that its not realistic to resolve ourselves, please do remove from milestone. But we'll leave it on there for now since we simply don't know yet.

Pietro Albini (Oct 11 2018 at 14:37, on Zulip):

[removed the milestone]

nagisa (Oct 11 2018 at 14:37, on Zulip):

Okay.

pnkfelix (Oct 11 2018 at 14:37, on Zulip):

next: "Regression from stable: pointer to usize conversion no longer compiles" #54709

pnkfelix (Oct 11 2018 at 14:38, on Zulip):

This is tagged with T-lang

pnkfelix (Oct 11 2018 at 14:39, on Zulip):

I cannot immediately tell whether there is something on #54709 for T-compiler to immediately address. It seems like its first blocked on a decision from T-lang about ... something ...

pnkfelix (Oct 11 2018 at 14:39, on Zulip):

namely @RalfJ posed a I-needs-decision matter here

pnkfelix (Oct 11 2018 at 14:40, on Zulip):

I again am struck wondering whether we can reasonably address this by October 25th

pnkfelix (Oct 11 2018 at 14:40, on Zulip):

I guess I'll leave it on the 1.30 milestone for now

pnkfelix (Oct 11 2018 at 14:40, on Zulip):

and wait to see what T-lang says

pnkfelix (Oct 11 2018 at 14:41, on Zulip):

that's all the stable-to-beta regressions

pnkfelix (Oct 11 2018 at 14:41, on Zulip):

as for stable-to-nightly regressions

pnkfelix (Oct 11 2018 at 14:41, on Zulip):

there are two

pnkfelix (Oct 11 2018 at 14:41, on Zulip):

first: "Rustc panics on nightly with crate interpolate" #54654

pnkfelix (Oct 11 2018 at 14:42, on Zulip):

We must not have gotten to this list last week

pnkfelix (Oct 11 2018 at 14:42, on Zulip):

seems like it needs investigation.

pnkfelix (Oct 11 2018 at 14:42, on Zulip):

I'll assign it to myself to do some initial investigation

pnkfelix (Oct 11 2018 at 14:43, on Zulip):

second: "Compiler assertion failed: tcx.migrate_borrowck() when failing to move #[thread_local] static vars." #54797

pnkfelix (Oct 11 2018 at 14:44, on Zulip):

I assigned this to myself; its clearly an NLL issue

pnkfelix (Oct 11 2018 at 14:44, on Zulip):

okay, next agenda item: Waiting for our team

pnkfelix (Oct 11 2018 at 14:45, on Zulip):

only one thing here: "Correct alignment of atomic types and (re)add Atomic{I,U}128" #53514

pnkfelix (Oct 11 2018 at 14:46, on Zulip):

Does anyone here want to take point on this?

nagisa (Oct 11 2018 at 14:46, on Zulip):

I have submitted a patch to LLVM back in the past

nagisa (Oct 11 2018 at 14:47, on Zulip):

but it didn’t fly, because Clang is written in some particular way that didn’t take well to those changes

nagisa (Oct 11 2018 at 14:47, on Zulip):

if we want to fix this on our side, what we need to do is to remove the sanity checking that checks our data-layouts against LLVM’s own and override the datalayouts for all targets to set the alignment to the correct one for 128-bit integers

nagisa (Oct 11 2018 at 14:49, on Zulip):

This should be fairly easy to fix, we just need to decide if it is alright for us to not check against LLVM’s own data-layouts anymore and just supply our own.

pnkfelix (Oct 11 2018 at 14:49, on Zulip):

@nagisa by "fix this", I take it you mean "if we want to re-add #[repr(transparent)] to the atomic types here?

pnkfelix (Oct 11 2018 at 14:49, on Zulip):

or do you also mean changing ("correcting" (?)) the alignments of the relevant atomic types?

nagisa (Oct 11 2018 at 14:49, on Zulip):

(oh, I see, I was thinking of i128 specifically)

nagisa (Oct 11 2018 at 14:50, on Zulip):

this is the other issue

nagisa (Oct 11 2018 at 14:50, on Zulip):

ignore me heh

pnkfelix (Oct 11 2018 at 14:50, on Zulip):

Okay I don't think we're going to resolve this question right here right now

pnkfelix (Oct 11 2018 at 14:50, on Zulip):

But I think we should try to plan to respond in some way

nikomatsakis (Oct 11 2018 at 14:50, on Zulip):

yeah, I was just reading the backscroll, lots of complex stuff

nikomatsakis (Oct 11 2018 at 14:51, on Zulip):

tfw you see comments from yourself and don't remember writing them at all

pnkfelix (Oct 11 2018 at 14:51, on Zulip):

Normally I'd ask @nagisa to be the local expert here and try to summarize the whole matter for the rest of us for next weeks' meeting

pnkfelix (Oct 11 2018 at 14:51, on Zulip):

so that we could make a decision then

nikomatsakis (Oct 11 2018 at 14:51, on Zulip):

obviously it's good to decide

nikomatsakis (Oct 11 2018 at 14:51, on Zulip):

but is there time pressure here?

pnkfelix (Oct 11 2018 at 14:51, on Zulip):

Hmm I guess not ... not anymore?

nikomatsakis (Oct 11 2018 at 14:51, on Zulip):

what is even the decision, exactly?

nikomatsakis (Oct 11 2018 at 14:52, on Zulip):

seems like LLVM has certain requirements

pnkfelix (Oct 11 2018 at 14:52, on Zulip):

https://github.com/rust-lang/rust/pull/53514#issuecomment-426012818

nagisa (Oct 11 2018 at 14:52, on Zulip):

I don’t think there’s time pressure, but people definitely do want to use those types.

pnkfelix (Oct 11 2018 at 14:52, on Zulip):

It seems, if nothing else

pnkfelix (Oct 11 2018 at 14:52, on Zulip):

we should probably assign this to someone other than @Taylor Cramer ?

pnkfelix (Oct 11 2018 at 14:53, on Zulip):

since it is marked as blocked by our team?

nikomatsakis (Oct 11 2018 at 14:53, on Zulip):

One topic was this:

Unclear what alignment we document as required for *mut T for the *mut T as &AtomicT conversion to be valid

but maybe the answer is to make it defined on a case-by-case basis and leave e.g. u64 undefined? or say, "it is true if these alignment restrictions are met, and maybe otherwise"?

nikomatsakis (Oct 11 2018 at 14:53, on Zulip):

I guess I'll leave those questions on thread

pnkfelix (Oct 11 2018 at 14:54, on Zulip):

@nagisa do you think you'd have time to make a summary comment?

pnkfelix (Oct 11 2018 at 14:54, on Zulip):

if you don't, that's fine; I can assign to self to try to do

pnkfelix (Oct 11 2018 at 14:54, on Zulip):

but I just figure you'd both do a better job and it would proably be a lot faster

nagisa (Oct 11 2018 at 14:54, on Zulip):

Maybe, this, I think is very related to a more basic issue https://github.com/rust-lang/rust/issues/54341

nagisa (Oct 11 2018 at 14:55, on Zulip):

Fixing this one should at least allow making AtomicI/U128 work and be transparent by default without needing to do anything else.

nagisa (Oct 11 2018 at 14:55, on Zulip):

(at least on platforms where i128 is indeed aligned to 128 bits)

pnkfelix (Oct 11 2018 at 14:55, on Zulip):

Okay, I'm going to assign this to @nagisa to provide us all with a summary of our status and what our options are

nagisa (Oct 11 2018 at 14:55, on Zulip):

Okay.

pnkfelix (Oct 11 2018 at 14:56, on Zulip):

... and the I-nominated T-compiler is ... :scared:

pnkfelix (Oct 11 2018 at 14:57, on Zulip):

None of these are high priority, luckily. :smile:

nikomatsakis (Oct 11 2018 at 14:57, on Zulip):

3 minutes? No problemo

pnkfelix (Oct 11 2018 at 14:57, on Zulip):

(except for the ones we've already looked at)

pnkfelix (Oct 11 2018 at 14:58, on Zulip):

The things tagged with A-rust-2018-preview might be worth looking at though

nikomatsakis (Oct 11 2018 at 14:58, on Zulip):

I nominated https://github.com/rust-lang/rust/issues/52098 for prioritization

nikomatsakis (Oct 11 2018 at 14:58, on Zulip):

and in particular to ask whether it is an RC2 blocker

nikomatsakis (Oct 11 2018 at 14:58, on Zulip):

(speaking of comments I do not remember writing)

nikomatsakis (Oct 11 2018 at 14:58, on Zulip):

that said, it might be fixed by now in #54689

nikomatsakis (Oct 11 2018 at 14:58, on Zulip):

oh, no, that was a dup

nagisa (Oct 11 2018 at 14:58, on Zulip):

https://github.com/rust-lang/rust/pull/54592 is complete, it is just waiting for minor changes from the author and can land.

nagisa (Oct 11 2018 at 14:59, on Zulip):

can unnominate

nikomatsakis (Oct 11 2018 at 14:59, on Zulip):

I'll just make #52098 P-high and assign to myself

nikomatsakis (Oct 11 2018 at 14:59, on Zulip):

it's the same area of code as this other '_-related thing

nikomatsakis (Oct 11 2018 at 15:00, on Zulip):

/me gotta run, bbs

pnkfelix (Oct 11 2018 at 15:00, on Zulip):

okay. and "[Rust 2018] rustdoc doesn't link "pub use whatever_crate;"" #52509 is more a rustdc bug than a T-compiler one, I think...

oli (Oct 11 2018 at 15:01, on Zulip):

if you haven't already, please look at https://github.com/rust-lang/rust/pull/53821#issuecomment-427053495 and check your boxes if you agree, the PR has been ready for a while, we just never got to it in a meeting

pnkfelix (Oct 11 2018 at 15:01, on Zulip):

@nikomatsakis added T-compiler though. Too bad they ran off. :smile:

pnkfelix (Oct 11 2018 at 15:02, on Zulip):

Okay that's all folks! Thanks for attending!

nagisa (Oct 11 2018 at 15:02, on Zulip):

o/

Last update: Nov 22 2019 at 05:50UTC