Stream: t-compiler/wg-nll

Topic: weekly meeting July 31


pnkfelix (Jul 31 2018 at 19:30, on Zulip):

hi @everyone

simulacrum (Jul 31 2018 at 19:31, on Zulip):

o/

davidtwco (Jul 31 2018 at 19:31, on Zulip):

o/

pnkfelix (Jul 31 2018 at 19:32, on Zulip):

As usual we have the NLL triage paper as a place for everyone to jot down their status

pnkfelix (Jul 31 2018 at 19:32, on Zulip):

I would like to call everyone's attention to my last bullet item: want everyone here to weigh in on buckets in https://github.com/rust-lang/rust/issues/52663

pnkfelix (Jul 31 2018 at 19:33, on Zulip):

namely, I tried to be pretty exhaustive in enumerating all the differences,

Santiago Pastorino (Jul 31 2018 at 19:34, on Zulip):

hi :)

pnkfelix (Jul 31 2018 at 19:34, on Zulip):

even differences that there is a pretty good argument that the difference is fine or at least acceptable for now

pnkfelix (Jul 31 2018 at 19:34, on Zulip):

but anyway, I figure I could let other people weigh in on which differences seem to actually matter

pnkfelix (Jul 31 2018 at 19:35, on Zulip):

(and of course, if anyone wants to have a go at fixing any of them, that's also an option.

pnkfelix (Jul 31 2018 at 19:35, on Zulip):

So @nikomatsakis is on PTO this week

pnkfelix (Jul 31 2018 at 19:36, on Zulip):

the big news (at least in my opinion) is that PR #52681 landed

pnkfelix (Jul 31 2018 at 19:36, on Zulip):

and it included integration into the 2008 2018 edition. So, we're committed at this point. :)

Pietro Albini (Jul 31 2018 at 19:37, on Zulip):

it got integrated this late?

davidtwco (Jul 31 2018 at 19:37, on Zulip):

I like the 2008 edition.

pnkfelix (Jul 31 2018 at 19:37, on Zulip):

it got integrated this late?

well, it was in the compiler

pnkfelix (Jul 31 2018 at 19:37, on Zulip):

but it wasn't toggled on by the 2008 2018 flag

Santiago Pastorino (Jul 31 2018 at 19:38, on Zulip):

Felix still didn't realize :P

pnkfelix (Jul 31 2018 at 19:38, on Zulip):

and it was mainly waiting on the migration mode (that same PR) that downgrades errors to warnings if AST-borrowck accepts the code

pnkfelix (Jul 31 2018 at 19:38, on Zulip):

so, yes, you are right to be somewhat shocked that we're turning it on this late in the game

Santiago Pastorino (Jul 31 2018 at 19:39, on Zulip):

Felix, we are in 2018 :P

Pietro Albini (Jul 31 2018 at 19:39, on Zulip):

10 years late is a bit too much though

pnkfelix (Jul 31 2018 at 19:39, on Zulip):

oh

Santiago Pastorino (Jul 31 2018 at 19:39, on Zulip):

:D

pnkfelix (Jul 31 2018 at 19:40, on Zulip):

anyway

pnkfelix (Jul 31 2018 at 19:40, on Zulip):

What's next!

pnkfelix (Jul 31 2018 at 19:40, on Zulip):

I already mentioned the diagnostic survey

pnkfelix (Jul 31 2018 at 19:40, on Zulip):

but my real take away from that effort was: I think we are actually in a pretty good spot

pnkfelix (Jul 31 2018 at 19:40, on Zulip):

We do need to triage that list of buckets, figure out which things to prioritize for the Release Candidate

pnkfelix (Jul 31 2018 at 19:41, on Zulip):

but overall I was pretty happy with our diagnostics

pnkfelix (Jul 31 2018 at 19:41, on Zulip):

How are we doing on performance?

pnkfelix (Jul 31 2018 at 19:41, on Zulip):

(I honestly have no idea.)

davidtwco (Jul 31 2018 at 19:42, on Zulip):

@lqd probably has the best idea, I'm not too sure myself.

Matthew Jasper (Jul 31 2018 at 19:42, on Zulip):

Well, const eval got slower, so we now look better. :rolling_on_the_floor_laughing:

pnkfelix (Jul 31 2018 at 19:43, on Zulip):

Yeah I don't see any massive difference when I look at https://perf.rust-lang.org/nll-dashboard.html ; we're still really bad on html5ever and tuple-stress

lqd (Jul 31 2018 at 19:44, on Zulip):

We seem to be okay I think, up to date crate versions run faster than the versions on perf.rlo

pnkfelix (Jul 31 2018 at 19:44, on Zulip):

and also people running the full compiler won't notice the impact as much as people running cargo check, right?

lqd (Jul 31 2018 at 19:44, on Zulip):

And we have some wip work for html5ever, but will likely hit he dataflow/moveck slowness hitting tuple-stress next

pnkfelix (Jul 31 2018 at 19:45, on Zulip):

(deleted)

lqd (Jul 31 2018 at 19:45, on Zulip):

true

pnkfelix (Jul 31 2018 at 19:46, on Zulip):

Hey @Matthew Jasper since you have been looking at stuff with closure captures

lqd (Jul 31 2018 at 19:46, on Zulip):

felix you might have ideas on the tuple-stress profile in the "perf.rlo benchmarks" topic

davidtwco (Jul 31 2018 at 19:46, on Zulip):

We should make sure there's a way to keep track of who is doing any of the diagnostic buckets so that we don't duplicate any effort.

pnkfelix (Jul 31 2018 at 19:47, on Zulip):

@Matthew Jasper could you take a look at the list in NLL lacks special case handling of closures and make sure all of those have appropriate issues?

lqd (Jul 31 2018 at 19:48, on Zulip):

(oh niko started looking at the crater results and there will be new bugs to be filed)

pnkfelix (Jul 31 2018 at 19:48, on Zulip):

@David Wood yeah; assigning people in issues would be one step

pnkfelix (Jul 31 2018 at 19:48, on Zulip):

@David Wood if something doesn't get an issue because its "too minor", then they could at least add themselves to the table?

pnkfelix (Jul 31 2018 at 19:48, on Zulip):

or I guess you all don't have edit privleges for the table

davidtwco (Jul 31 2018 at 19:49, on Zulip):

Yeah, a column on the table is probably sufficient. I think anyone in the WG-compiler-nll group can edit the table.

pnkfelix (Jul 31 2018 at 19:49, on Zulip):

@lqd okay. I'm not 100% sure I'll have time to look at it too closely because there are some non NLL compiler issues that I think I need to attack this week

pnkfelix (Jul 31 2018 at 19:50, on Zulip):

@lqd but I'll try to remember to at least look at it

lqd (Jul 31 2018 at 19:51, on Zulip):

Cool, could be worthwhile eventually (ie the liveness + static fast path for html5ever won't be enough most likely)

pnkfelix (Jul 31 2018 at 19:53, on Zulip):

I'm trying to figure out if there's anything else besides diagnostics and performance to talk about

pnkfelix (Jul 31 2018 at 19:53, on Zulip):

oh

pnkfelix (Jul 31 2018 at 19:53, on Zulip):

ICE's of course

Matthew Jasper (Jul 31 2018 at 19:54, on Zulip):

I think that we've got most of them now

pnkfelix (Jul 31 2018 at 19:54, on Zulip):

okay. I just noticed #52792

pnkfelix (Jul 31 2018 at 19:55, on Zulip):

(but I think other people on the compiler team who know more about the generator support code are going to be dealing with that.)

pnkfelix (Jul 31 2018 at 19:55, on Zulip):

but the main thing is

pnkfelix (Jul 31 2018 at 19:55, on Zulip):

that probably ICE's need to take highest priority now

pnkfelix (Jul 31 2018 at 19:56, on Zulip):

now that we're turned on in the 2008 edition

lqd (Jul 31 2018 at 19:56, on Zulip):

all the ones we hit in the crater runs are fixed now I think

Matthew Jasper (Jul 31 2018 at 19:56, on Zulip):

I think #52792 is independent of NLL

pnkfelix (Jul 31 2018 at 19:58, on Zulip):

Oh there was one other thing I wanted to get people's temperature on

pnkfelix (Jul 31 2018 at 19:59, on Zulip):

Back in May, PR #50593 turned off the location-dependent outlives relation

pnkfelix (Jul 31 2018 at 19:59, on Zulip):

and that, as expected, broke #51526

pnkfelix (Jul 31 2018 at 19:59, on Zulip):

the thing I've been wondering is ..

pnkfelix (Jul 31 2018 at 20:00, on Zulip):

should we put back in the location-dependent outlives relation, under a debug-flag

pnkfelix (Jul 31 2018 at 20:00, on Zulip):

(or other feature-gated opt-in)

lqd (Jul 31 2018 at 20:00, on Zulip):

oh right the thing Simon talked about yesterday

pnkfelix (Jul 31 2018 at 20:00, on Zulip):

as a way to still test the more feature-full NLL

davidtwco (Jul 31 2018 at 20:00, on Zulip):

What ever happened with the type annotations issue?

pnkfelix (Jul 31 2018 at 20:00, on Zulip):

without going all the way to Polonius

pnkfelix (Jul 31 2018 at 20:00, on Zulip):

@David Wood I think that issue is still open

pnkfelix (Jul 31 2018 at 20:00, on Zulip):

... I was treating it as a diagnostic issue

pnkfelix (Jul 31 2018 at 20:01, on Zulip):

but now that you mention it, its not purely diagnostic, is it

pnkfelix (Jul 31 2018 at 20:01, on Zulip):

this is issue #47184

pnkfelix (Jul 31 2018 at 20:01, on Zulip):

so yeah, we do want to resolve that in time for the Release Candidate

pnkfelix (Jul 31 2018 at 20:02, on Zulip):

and ... i don't think we have any solid plan for resolving it. Not currently.

pnkfelix (Jul 31 2018 at 20:04, on Zulip):

I did have some code hacked up to try to support let (a, b): (T, U); checking for #47184

pnkfelix (Jul 31 2018 at 20:04, on Zulip):

but I don't remember where I put it. And it didn't quite work. And I didn't like it.

pnkfelix (Jul 31 2018 at 20:05, on Zulip):

I should at least go see if I can find that branch again.

pnkfelix (Jul 31 2018 at 20:09, on Zulip):

okay well our time is up

pnkfelix (Jul 31 2018 at 20:09, on Zulip):

Thanks @everyone for attending

lqd (Jul 31 2018 at 20:10, on Zulip):

(Location dependent outlives could be nice under another flag :) \o

Last update: Nov 21 2019 at 23:45UTC