Stream: t-compiler

Topic: #54818 weekly meeting 2019-01-10


nagisa (Jan 10 2019 at 13:49, on Zulip):

@nikomatsakis want me to drive again? I might not have much time for prep.

nikomatsakis (Jan 10 2019 at 14:15, on Zulip):

@nagisa good timing, I was literally just sitting down to ping you

nikomatsakis (Jan 10 2019 at 14:16, on Zulip):

I'm happy for you to drive, but if you're pressed for time I don't mind doing it. I could also just do a bit of prep if you like

nagisa (Jan 10 2019 at 14:16, on Zulip):

I’ll leave it up to you this time around then.

nikomatsakis (Jan 10 2019 at 14:16, on Zulip):

Oh, I see my computer just delivered a notification for an older message :) well, you beat me :)

nikomatsakis (Jan 10 2019 at 14:16, on Zulip):

OK

nagisa (Jan 10 2019 at 14:17, on Zulip):

Oh, I see my computer just delivered a notification for an older message :slight_smile: well, you beat me :slight_smile:

I changed the topic to add the issue number, which must be why your notification arrived now.

nikomatsakis (Jan 10 2019 at 14:21, on Zulip):

# Pre-meeting review

nikomatsakis (Jan 10 2019 at 14:21, on Zulip):

## P-high

nikomatsakis (Jan 10 2019 at 14:21, on Zulip):

### [NLL] Bad higher ranked subtype error #57374

nikomatsakis (Jan 10 2019 at 14:22, on Zulip):

We discussed this last night at the NLL meeting. These errors are not exposed to users yet, but they will be when we end migration mode. @lqd is going to take a stab at fixing it, I am going to try and mentor.

nikomatsakis (Jan 10 2019 at 14:22, on Zulip):

### Confusing error message associated with universe transition: "one type is more general than the other" #57362

nikomatsakis (Jan 10 2019 at 14:22, on Zulip):

This is assigned to me, but I haven't done anything yet. Related to the previous issue so maybe I'll try to address both at the same time.

nikomatsakis (Jan 10 2019 at 14:23, on Zulip):

### error: internal compiler error: src/librustc/ty/relate.rs:753: impossible case reached: can't relate: (T,) with Lifetime('_#1r) #57156

nikomatsakis (Jan 10 2019 at 14:23, on Zulip):

Has pending fix in #57355. (Thanks @Ariel Ben-Yehuda!) That PR is beta-nominated.

nikomatsakis (Jan 10 2019 at 14:24, on Zulip):

### PhantomData fields in repr(C) structs change ABI on aarch64 #56877

We had some discussion about the best solution and it seems like we're nearing a proposal. I took a stab at prototyping some code but didn't really get as far as testing it etc.

nikomatsakis (Jan 10 2019 at 14:24, on Zulip):

### Nightly: thread '<unnamed>' panicked at 'couldn't compile the test', src/librustdoc/test.rs:326:13 #56867

nikomatsakis (Jan 10 2019 at 14:25, on Zulip):

We thought we had a workaround in the form of https://github.com/crossbeam-rs/crossbeam/pull/268, but it seems like we did not, or at least not fully. Not sure of the current status. @nagisa is assigned. I'll add this to the list to revisit during the meeting for an update.

nikomatsakis (Jan 10 2019 at 14:26, on Zulip):

### prohibit "two-phase borrows" with existing borrows? #56254

nikomatsakis (Jan 10 2019 at 14:27, on Zulip):

This is assigned to me. I think we decided to try and adopt a conservative stance for now (which means more errors, though they will begin as warnings). @Matthew Jasper started preparing a PR to change how we do the lowering of matches which is a necessary first step but ran into some complications. They and I need to figure out how to resolve.

nikomatsakis (Jan 10 2019 at 14:27, on Zulip):

### regression: stack overflow on macosx with xcode 6.4 #55471

nikomatsakis (Jan 10 2019 at 14:29, on Zulip):

Seems like this was finally fixed "for good" in #56813 (Thanks @Oli and @nagisa!) but there remain some questions of whether to backport. I will add to the list to discuss whether to close, backport, or what.

nikomatsakis (Jan 10 2019 at 14:29, on Zulip):

### [nll] _ patterns should not count as borrows #53114

nikomatsakis (Jan 10 2019 at 14:30, on Zulip):

I assigned to myself but failed to follow up with @Matthew Jasper -- seems likely though that the right action is to close, so not super high pressure.

nikomatsakis (Jan 10 2019 at 14:30, on Zulip):

### NLL / MIR-borrowck regression in getopts #48225

I forgot to nominate this for discussion in the NLL meeting. I think the right action is just to close. I will nominate for the NLL meeting next week though.

nikomatsakis (Jan 10 2019 at 14:31, on Zulip):

### two-phase-borrows need a specification #46901

@pnkfelix has been out, no progress.

nikomatsakis (Jan 10 2019 at 14:32, on Zulip):

### figure out how to integrate constants and the MIR type checker #46702

Assigned to me, but no progress. @Matthew Jasper has some pending PRs in this area though, should perhaps reassign to them.

nikomatsakis (Jan 10 2019 at 14:39, on Zulip):

## Beta nominations

nikomatsakis (Jan 10 2019 at 14:47, on Zulip):

Listed in the doc

nikomatsakis (Jan 10 2019 at 14:47, on Zulip):

## Stable nominations

nikomatsakis (Jan 10 2019 at 14:47, on Zulip):

None

nikomatsakis (Jan 10 2019 at 14:47, on Zulip):

## Stable-to-beta regressions

nikomatsakis (Jan 10 2019 at 14:47, on Zulip):

Nothing new for T-compiler, only T-rustdoc

nikomatsakis (Jan 10 2019 at 14:47, on Zulip):

## Stable-to-nightly regressions

nikomatsakis (Jan 10 2019 at 14:48, on Zulip):

Same, nothing new for T-compiler

nikomatsakis (Jan 10 2019 at 14:48, on Zulip):

## Waiting for our team

nikomatsakis (Jan 10 2019 at 14:49, on Zulip):

### Add future incompatibility lint for #[link] without arguments #57139

nikomatsakis (Jan 10 2019 at 14:49, on Zulip):

### Automatically open an issue when a tool breaks #56951

nikomatsakis (Jan 10 2019 at 14:49, on Zulip):

## Nominated

nikomatsakis (Jan 10 2019 at 14:52, on Zulip):

Hello @T-compiler / @T-compiler/meeting! Meeting in this topic in ~10 minutes. I've already gone through most of the agenda and created this HackMd document where we can the condensed stuff where feedback seems useful:

https://hackmd.io/hLbW5ZxTQ56C8ngRk0r6Zw#

nikomatsakis (Jan 10 2019 at 14:53, on Zulip):

(You can also view my log further up...)

nikomatsakis (Jan 10 2019 at 15:01, on Zulip):

Hello @T-compiler / @T-compiler/meeting =) Triage meeting

nikomatsakis (Jan 10 2019 at 15:01, on Zulip):

:wave:

nikomatsakis (Jan 10 2019 at 15:02, on Zulip):

Based on the HackMD document above, we can skip directly to the "items of interest"

nikomatsakis (Jan 10 2019 at 15:02, on Zulip):

seems like a quiet week

nikomatsakis (Jan 10 2019 at 15:03, on Zulip):

Should we get started? (I'm debating if we should give a few more minutes for folks to show up)

nikomatsakis (Jan 10 2019 at 15:04, on Zulip):

OK, first thing:

nikomatsakis (Jan 10 2019 at 15:04, on Zulip):

Nightly: thread ‘<unnamed>’ panicked at ‘couldn’t compile the test’, src/librustdoc/test.rs:326:13 #56867

nagisa (Jan 10 2019 at 15:04, on Zulip):

seems like I have neglected to write up my thoughts on this after last meeting

nikomatsakis (Jan 10 2019 at 15:04, on Zulip):

Seems like the crossbeam issue "resolved itself"

nikomatsakis (Jan 10 2019 at 15:04, on Zulip):

There is some other potential issue that we cannot reproduce

nikomatsakis (Jan 10 2019 at 15:04, on Zulip):

I'm guessing @nagisa you're inclined to close at this point?

nagisa (Jan 10 2019 at 15:05, on Zulip):

Seems like the crossbeam issue "resolved itself"

I’m not sure this is true.

nagisa (Jan 10 2019 at 15:05, on Zulip):

that is, I’ve never tried reproducing again

nikomatsakis (Jan 10 2019 at 15:05, on Zulip):

Always suspicious

nikomatsakis (Jan 10 2019 at 15:05, on Zulip):

Well, anyway, I guess the summary for purposes of this meeting is "still in hand", right?

nikomatsakis (Jan 10 2019 at 15:05, on Zulip):

i.e., you are assigned, you're making progress, don't need anybody else to be assigned?

nagisa (Jan 10 2019 at 15:05, on Zulip):

I would be inclined to close the original issue if I never receive a minimized sample that reproduces for me

nikomatsakis (Jan 10 2019 at 15:06, on Zulip):

OK

nagisa (Jan 10 2019 at 15:06, on Zulip):

I’ll probably do just that if I cannot reproduce the crossbeam issue either.

nikomatsakis (Jan 10 2019 at 15:06, on Zulip):

maybe should we ping sstangl explicitly then

nikomatsakis (Jan 10 2019 at 15:06, on Zulip):

in case they -- like so many of us -- are behind on GH notifications :)

nagisa (Jan 10 2019 at 15:07, on Zulip):

Fair :slight_smile:

nikomatsakis (Jan 10 2019 at 15:08, on Zulip):

Left a message

nikomatsakis (Jan 10 2019 at 15:08, on Zulip):

Feel free to correct if I misrepresented anything

nikomatsakis (Jan 10 2019 at 15:08, on Zulip):

### regression: stack overflow on macosx with xcode 6.4 #55471

nikomatsakis (Jan 10 2019 at 15:08, on Zulip):

Seems like this was finally fixed “for good” in #56813 (Thanks @Oli and @nagisa!) but there remain some questions of whether to backport. I will add to the list to discuss whether to close, backport, or what.

nikomatsakis (Jan 10 2019 at 15:08, on Zulip):

Thoughts @Oli , @nagisa ?

nagisa (Jan 10 2019 at 15:09, on Zulip):

I have a feeling that we slipped past all the backport deadlines and there is very little reason to backport anything here

nikomatsakis (Jan 10 2019 at 15:09, on Zulip):

Seems plausible. We do have a beta-nominated PR so maybe we'll just discuss then?

nikomatsakis (Jan 10 2019 at 15:10, on Zulip):

( @Pietro Albini might be able to tell us re: deadlines. )

nikomatsakis (Jan 10 2019 at 15:10, on Zulip):

OK, next up:

# Beta Nominations

nikomatsakis (Jan 10 2019 at 15:10, on Zulip):

## Delay gensym creation for “underscore items” (use foo as _/const _) until name resolution #56392

nikomatsakis (Jan 10 2019 at 15:11, on Zulip):

(presuming there is still some point to backporting at this moment)

nikomatsakis (Jan 10 2019 at 15:11, on Zulip):

Rust 1.32 stable will be released on Thu Jan 17 2019.

Vadim Petrochenkov (Jan 10 2019 at 15:11, on Zulip):

Only one commit from it (currently rolled up into https://github.com/rust-lang/rust/pull/57483)

nikomatsakis (Jan 10 2019 at 15:11, on Zulip):

seems like there is still time

nikomatsakis (Jan 10 2019 at 15:11, on Zulip):

Only one commit from it (currently rolled up into https://github.com/rust-lang/rust/pull/57483)

ok

nikomatsakis (Jan 10 2019 at 15:11, on Zulip):

I guess we'll do the vote thing

nikomatsakis (Jan 10 2019 at 15:12, on Zulip):

Please signal your opinion =)

:back: == backport
:x: == no

nikomatsakis (Jan 10 2019 at 15:12, on Zulip):

Maybe we should just be voting on #57483 in its entirety

mw (Jan 10 2019 at 15:12, on Zulip):

That's a pretty big PR for a backport

nikomatsakis (Jan 10 2019 at 15:12, on Zulip):

@mw which are you referring to

mw (Jan 10 2019 at 15:13, on Zulip):

#57483

nikomatsakis (Jan 10 2019 at 15:13, on Zulip):

it's 3 backports I guess

nikomatsakis (Jan 10 2019 at 15:13, on Zulip):

I'm not sure how to handle this procedurally :)

nikomatsakis (Jan 10 2019 at 15:14, on Zulip):

but I think it's a good idea to close compatibility problems if we can

mw (Jan 10 2019 at 15:14, on Zulip):

although 2/3 of it are updated tests

nagisa (Jan 10 2019 at 15:14, on Zulip):

note that this appears to backport the stabilisation of uniform paths

nagisa (Jan 10 2019 at 15:15, on Zulip):

Zulip 500-ing on me

Vadim Petrochenkov (Jan 10 2019 at 15:15, on Zulip):

This is an unfortunate analogue of https://github.com/rust-lang/rust/pull/55884, but for the next release.

nagisa (Jan 10 2019 at 15:15, on Zulip):

That’s where the large chunk of changes is coming from.

mw (Jan 10 2019 at 15:16, on Zulip):

(btw, :heart: for all the work your doing, @Vadim Petrochenkov!)

nikomatsakis (Jan 10 2019 at 15:16, on Zulip):

@Vadim Petrochenkov what is the incentive to backport?

Vadim Petrochenkov (Jan 10 2019 at 15:17, on Zulip):

(Zulip, what are you doing.)

Vadim Petrochenkov (Jan 10 2019 at 15:22, on Zulip):

2) Anchored paths are confusing, issues like https://www.reddit.com/r/rust/comments/aeh8dn/rust_modules_are_not_working/ are commonly reported.

nikomatsakis (Jan 10 2019 at 15:22, on Zulip):

er, to backport the stabilization specifically

nikomatsakis (Jan 10 2019 at 15:22, on Zulip):

IIRC, that PR includes some bugfixes as well, is that part of it?

nikomatsakis (Jan 10 2019 at 15:22, on Zulip):

(Zulip seems to be working now, at least for me)

nikomatsakis (Jan 10 2019 at 15:22, on Zulip):

@T-compiler/meeting I realize that sending messages ove Zulip to inform you that Zulip is working again is silly, but if you looked away from the tab, and it reloaded, it seems to work now =)

nikomatsakis (Jan 10 2019 at 15:22, on Zulip):

So, let's step back a second. @Vadim Petrochenkov has a PR with 3 distinct backports:

Vadim Petrochenkov (Jan 10 2019 at 15:23, on Zulip):

what is the incentive to backport?

See the list of issues closed by https://github.com/rust-lang/rust/pull/56759 for examples.

nikomatsakis (Jan 10 2019 at 15:23, on Zulip):

I think personally that the first two seem like obvious candidates for a backport (small fixes, regressions)

Vadim Petrochenkov (Jan 10 2019 at 15:23, on Zulip):

Two categories mainly:

nikomatsakis (Jan 10 2019 at 15:23, on Zulip):

I'm not sure about #56759

Vadim Petrochenkov (Jan 10 2019 at 15:23, on Zulip):

1) rustfix hits the uniform_paths future gates during migration to 2018.

Vadim Petrochenkov (Jan 10 2019 at 15:23, on Zulip):

E.g. extern crate json -> use json -> feature error, json is a macro_use'd macro in addition to the crate.

Vadim Petrochenkov (Jan 10 2019 at 15:23, on Zulip):

2) Anchored paths are confusing, issues like https://www.reddit.com/r/rust/comments/aeh8dn/rust_modules_are_not_working/ are commonly reported.

nagisa (Jan 10 2019 at 15:23, on Zulip):

ping pong echo~o?

Pietro Albini (Jan 10 2019 at 15:23, on Zulip):

btw, for the backports timeline

Pietro Albini (Jan 10 2019 at 15:24, on Zulip):

hard deadline is on Monday for the promotion

nikomatsakis (Jan 10 2019 at 15:24, on Zulip):

@Vadim Petrochenkov ok, thanks for the clarifications

Pietro Albini (Jan 10 2019 at 15:24, on Zulip):

any question in particular?

nikomatsakis (Jan 10 2019 at 15:24, on Zulip):

@Pietro Albini that was it

nikomatsakis (Jan 10 2019 at 15:25, on Zulip):

OK, so the question then is whether to backport #57483, which encompasses the PRs I listed previously.

nikomatsakis (Jan 10 2019 at 15:26, on Zulip):

@Vadim Petrochenkov mentioned that this is good because it fixes a rustfix migration problem and because, in general, the changes are good and eliminate a source of some confusion.

nikomatsakis (Jan 10 2019 at 15:26, on Zulip):

the counter argument is that the PR is large-ish (although iirc from reviewing it's a fairly rote transformation) and that we don't typically backport stabilizations

nagisa (Jan 10 2019 at 15:27, on Zulip):

I would gladly backport the first 2

Pietro Albini (Jan 10 2019 at 15:28, on Zulip):

FYI, the release team discussed this yesterday in our meeting, and opinions were mixed -- we'd prefer not but it's sort of ok if you find critical to backport the stabilization

nikomatsakis (Jan 10 2019 at 15:28, on Zulip):

@Vadim Petrochenkov do you know how often the rustfix issues etc arise?

nikomatsakis (Jan 10 2019 at 15:29, on Zulip):

(I guess the same applies to the confusion; it seems unfortunate, but then our paths have always been sort of confusing, and the fix is coming.)

Vadim Petrochenkov (Jan 10 2019 at 15:29, on Zulip):

I find it important enough.
Boats and aturon kind of abandoned the feature now, but they initially wanted the stabilization to happen in 1.31 for the edition release.

nikomatsakis (Jan 10 2019 at 15:30, on Zulip):

Certainly, that would've been nice

mw (Jan 10 2019 at 15:31, on Zulip):

I have a hard time assessing the risks here.

Vadim Petrochenkov (Jan 10 2019 at 15:32, on Zulip):

do you know how often the rustfix issues etc arise?

At least multiple of them were reported during the last release cycle - https://github.com/rust-lang/rust/issues/56326 https://github.com/rust-lang/rust/issues/56398 https://github.com/rust-lang/rust/issues/57252 https://github.com/rust-lang/rust/issues/57422

nikomatsakis (Jan 10 2019 at 15:32, on Zulip):

/me looks

Vadim Petrochenkov (Jan 10 2019 at 15:32, on Zulip):

I have a hard time assessing the risks here.

FWIW, a crater run is currently in progress for https://github.com/rust-lang/rust/pull/57483

mw (Jan 10 2019 at 15:33, on Zulip):

That's certainly a good idea

mw (Jan 10 2019 at 15:33, on Zulip):

will it be finished in time?

nikomatsakis (Jan 10 2019 at 15:33, on Zulip):

I wnated to ask about that

mw (Jan 10 2019 at 15:33, on Zulip):

started 12h ago

nikomatsakis (Jan 10 2019 at 15:33, on Zulip):

I feel on the fence, but I'm inclined to give some weight to @Vadim Petrochenkov's opinion here, since they've been carrying this code through.

Vadim Petrochenkov (Jan 10 2019 at 15:34, on Zulip):

(Crater should finish tomorrow.)

nikomatsakis (Jan 10 2019 at 15:35, on Zulip):

We could do a conditional approval based on crater results, potentially

Vadim Petrochenkov (Jan 10 2019 at 15:35, on Zulip):

We could do a conditional approval based on crater results, potentially

That's what release team decided yesterday as well.

mw (Jan 10 2019 at 15:36, on Zulip):

I wouldn't object to that

Wesley Wiser (Jan 10 2019 at 15:36, on Zulip):

That seems like a very pragmatic approach to me :thumbs_up:

nikomatsakis (Jan 10 2019 at 15:37, on Zulip):

OK

nikomatsakis (Jan 10 2019 at 15:39, on Zulip):

comment

mw (Jan 10 2019 at 15:40, on Zulip):

:tada:

nikomatsakis (Jan 10 2019 at 15:40, on Zulip):

ok, next up

nikomatsakis (Jan 10 2019 at 15:40, on Zulip):

## Always run rustc in a thread #56813

nikomatsakis (Jan 10 2019 at 15:40, on Zulip):

This was the PR that @nagisa mentioned earlier with we discussed the issue around stack overflow

nikomatsakis (Jan 10 2019 at 15:40, on Zulip):

(regression: stack overflow on macosx with xcode 6.4 #55471)

nagisa (Jan 10 2019 at 15:41, on Zulip):

First lets figure out if the regression this fixes is not in stable yet :slight_smile: The regression itself doesn’t appear too commonly encountered by people, so this does not fix that many of the users.

nikomatsakis (Jan 10 2019 at 15:41, on Zulip):

Seems like a pretty harmless PR though

nikomatsakis (Jan 10 2019 at 15:41, on Zulip):

it's basically revering us back to the older behavior, correct?

nagisa (Jan 10 2019 at 15:41, on Zulip):

Sure.

nagisa (Jan 10 2019 at 15:41, on Zulip):

I’m just operating under an assumption that not backporting has lower cost than backporting.

nikomatsakis (Jan 10 2019 at 15:41, on Zulip):

I don't quite understand @nagisa what this means:

First lets figure out if the regression this fixes is not in stable yet

nikomatsakis (Jan 10 2019 at 15:42, on Zulip):

I’m just operating under an assumption that not backporting has lower cost than backporting.

well it's strictly less work I guess

nagisa (Jan 10 2019 at 15:42, on Zulip):

@nikomatsakis although the original regression is tagged regression-from-stable-to-beta, it has been tagged that way since october 31 last year

nagisa (Jan 10 2019 at 15:43, on Zulip):

which would suggest this is already a stable-to-stable regression.

nikomatsakis (Jan 10 2019 at 15:43, on Zulip):

OK so you're saying "maybe this is in stable already and it's not that big a deal it can just ride the trains"

nagisa (Jan 10 2019 at 15:43, on Zulip):

right.

nikomatsakis (Jan 10 2019 at 15:44, on Zulip):

well, can we figure that out? which issue was it?

nagisa (Jan 10 2019 at 15:44, on Zulip):

the problem of course is that it is super non-trivial to reproduce (in fact neither I nor @Oli even attempted to do so, we intuited the fix)

nikomatsakis (Jan 10 2019 at 15:44, on Zulip):

#55471?

nagisa (Jan 10 2019 at 15:44, on Zulip):

yeah this is the issue

nagisa (Jan 10 2019 at 15:45, on Zulip):

also for how long this issue has been open, there has been awfully few "me too" comments or emotes :slight_smile:

nikomatsakis (Jan 10 2019 at 15:45, on Zulip):

the merge commit causing the problem was 5758c2dd14fd29caf7c7bb2123eb6b23443b9233

mw (Jan 10 2019 at 15:46, on Zulip):

yeah, doesn't seem all that urgent

nagisa (Jan 10 2019 at 15:46, on Zulip):

Right, but the stack overflow itself could’ve started appearing later on if e.g. miri showed up and started creating deep callstacks :slight_smile:

nikomatsakis (Jan 10 2019 at 15:46, on Zulip):

that commit is in stable

nikomatsakis (Jan 10 2019 at 15:46, on Zulip):

ok, seems overall low priority

nikomatsakis (Jan 10 2019 at 15:46, on Zulip):

let's vote

nikomatsakis (Jan 10 2019 at 15:46, on Zulip):

Yes! (add a :+1: )

nikomatsakis (Jan 10 2019 at 15:46, on Zulip):

No! (add a :+1: )

mw (Jan 10 2019 at 15:47, on Zulip):

:D

nikomatsakis (Jan 10 2019 at 15:48, on Zulip):

next up: use the correct supertrait substitution in object_ty_for_trait #57355

nikomatsakis (Jan 10 2019 at 15:50, on Zulip):

the patch is actually pretty small, but looks a bit bigger owing to debug! and things

nagisa (Jan 10 2019 at 15:50, on Zulip):

One of the similar PRs ended up breaking something in the recent past

nagisa (Jan 10 2019 at 15:50, on Zulip):

(similar as in similar area of code)

nikomatsakis (Jan 10 2019 at 15:51, on Zulip):

I presume you're not referring to the PR that caused the ICE which this is fixing

nikomatsakis (Jan 10 2019 at 15:51, on Zulip):

=)

nagisa (Jan 10 2019 at 15:51, on Zulip):

I might be :slight_smile:

nikomatsakis (Jan 10 2019 at 15:51, on Zulip):

I guess I'm inclined to backport: it's a beta regression, small patch, seems to fit the criteria. One nit though is that it has not landed yet

mw (Jan 10 2019 at 15:51, on Zulip):

this hasn't even landed on nightly yet

nikomatsakis (Jan 10 2019 at 15:51, on Zulip):

So maybe we wait

nikomatsakis (Jan 10 2019 at 15:51, on Zulip):

but that means the regression will hit stable

nikomatsakis (Jan 10 2019 at 15:51, on Zulip):

(better the devil we know, perhaps, though..?)

nagisa (Jan 10 2019 at 15:52, on Zulip):

easy solution

nikomatsakis (Jan 10 2019 at 15:52, on Zulip):

(I guess we could prep a backport and see what happens over next few days, merge on sunday or something if all is going well)

nikomatsakis (Jan 10 2019 at 15:52, on Zulip):

I would expect a clean backport

nagisa (Jan 10 2019 at 15:52, on Zulip):

@bors p=42 backport-accepted conditional on this landing first

nikomatsakis (Jan 10 2019 at 15:53, on Zulip):

yeah, I mean, it's already p=1 and next in the queue

nikomatsakis (Jan 10 2019 at 15:53, on Zulip):

but we can p=22

nikomatsakis (Jan 10 2019 at 15:53, on Zulip):

and try to stay on it

nikomatsakis (Jan 10 2019 at 15:53, on Zulip):

so that it stays that way

nikomatsakis (Jan 10 2019 at 15:53, on Zulip):

it's still not going to be a lot of "airtime"

nikomatsakis (Jan 10 2019 at 15:54, on Zulip):

(still, I say we do it)

mw (Jan 10 2019 at 15:54, on Zulip):

bug reports come in pretty quickly, usually

nikomatsakis (Jan 10 2019 at 15:54, on Zulip):

I guess it depends a bit on how obscure the underlying regression is

nikomatsakis (Jan 10 2019 at 15:54, on Zulip):

so shall we prep the "presuming it has landed and everything looks ok thus far..." backport?

mw (Jan 10 2019 at 15:55, on Zulip):

let's make sure it lands on nightly asap and backport of nothing breaks till monday

nikomatsakis (Jan 10 2019 at 15:55, on Zulip):

ok

nikomatsakis (Jan 10 2019 at 15:56, on Zulip):

comment

nikomatsakis (Jan 10 2019 at 15:56, on Zulip):

ok, that took way longer than anticipated

nikomatsakis (Jan 10 2019 at 15:56, on Zulip):

we have only a few minutes left, so let me raise a question instead of opening a new topic:

nikomatsakis (Jan 10 2019 at 15:57, on Zulip):

well, first, let me advertise that these issues have checklists, and you might want to put your approval or concerns:

nikomatsakis (Jan 10 2019 at 15:57, on Zulip):

then I'd like to ask: for some of these nominated issues, it feels like discussing them at triage is very hard, what should we do?

nikomatsakis (Jan 10 2019 at 15:58, on Zulip):

we could schedule a separate meeting

nikomatsakis (Jan 10 2019 at 15:58, on Zulip):

or use next steering meeting next week

nikomatsakis (Jan 10 2019 at 15:58, on Zulip):

examples:

nagisa (Jan 10 2019 at 15:58, on Zulip):

I find that doing a pre-pass for P-high issues saves a lot of the time and we sometimes get to the nominated ones :slight_smile:

nikomatsakis (Jan 10 2019 at 15:58, on Zulip):
nagisa (Jan 10 2019 at 15:58, on Zulip):

today was impossible because of technical difficulties among other things

nikomatsakis (Jan 10 2019 at 15:59, on Zulip):

right

nikomatsakis (Jan 10 2019 at 15:59, on Zulip):

on that list, I think that #57214 is probably a good meaty technical topic to look at during the All Hands

nikomatsakis (Jan 10 2019 at 16:00, on Zulip):

I'm not sure when is a great time to discuss #57016, but I hate to leave @Alexander Regueiro hanging forever. Maybe the answer is that we should try to finish the discussion async and maybe have an FCP motion.

nikomatsakis (Jan 10 2019 at 16:01, on Zulip):

(Oh, on a related note, I'm inclined to close https://github.com/rust-lang/rust/issues/56375, which @Vadim Petrochenkov says is "working as expected" -- perhaps I should FCP it though, seems like more of a T-lang thing than anything.)

nikomatsakis (Jan 10 2019 at 16:03, on Zulip):

ok, well, let's see if we can get some more motion on #57016 happening async then.

nikomatsakis (Jan 10 2019 at 16:03, on Zulip):

Thanks all! :heart:

Matthew Jasper (Jan 10 2019 at 16:21, on Zulip):

### figure out how to integrate constants and the MIR type checker #46702

Assigned to me, but no progress. @Matthew Jasper has some pending PRs in this area though, should perhaps reassign to them.

I have a fix for the cases in my comment on that issue locally, I'll try to get a PR up soon.

Alexander Regueiro (Jan 10 2019 at 16:35, on Zulip):

Could work...

Last update: Nov 22 2019 at 04:30UTC