Stream: t-compiler

Topic: pre-meeting triage 2019-04-25 #54818


nagisa (Apr 25 2019 at 13:02, on Zulip):

Ok, lets start this. We have a number (14) of nominated issues.

nagisa (Apr 25 2019 at 13:03, on Zulip):

ICE while bootstrapping (./x.py build) #2 #60228

This is a metadata ICE. My fingers itch to make this a P-high, but need to find somebody to investigate it.

nagisa (Apr 25 2019 at 13:04, on Zulip):

Very plausible that this will not be easily reproducible and will likely fix itself for the reporter if they started a clean build as well.

nagisa (Apr 25 2019 at 13:06, on Zulip):

ICE on higher-trait bounds #60218

Bisected, @nikomatsakis’s universes PR got blamed. It has been a while since it landed, which suggests people do not encounter this often. Still P-high?

nagisa (Apr 25 2019 at 13:08, on Zulip):

Async fns that take structs with elided lifetime parameters are broken. #60203

Reading @Taylor Cramer’s comment it seems like the current behaviour is expected? It seems like this was nominated for T-lang either way, so leaving it be.

nagisa (Apr 25 2019 at 13:10, on Zulip):

cannot borrow as mutable because it is also borrowed as immutable (likely regression) #60136

P-high, as it is a regression to beta. Should fix within a month or so. Any takers?

nagisa (Apr 25 2019 at 13:13, on Zulip):

error: internal compiler error: unexpected region in query response: ReScope(Destruction(219)) #60127

Another ICE, another P-high? Needs an assignee to investigate.

nagisa (Apr 25 2019 at 13:15, on Zulip):

-Clinker-plugin-lto doesn't work without extra manual work #60059

@nikomatsakis nominated this for an actual discussion, will return to it later

nagisa (Apr 25 2019 at 13:17, on Zulip):

rustc incorrectly warn #[no_mangle] is not used when doing incremental build #60050

P-…medium, perhaps? Would be nice to see it fixed to have rock-stable incremental, but also does not block anything.

nagisa (Apr 25 2019 at 13:19, on Zulip):

Rust 1.34 generates significantly less debug information for libstd functions vs. Rust 1.33 #60020

No changes since the last time. IIRC last time we complained this is providing descriptions of the behaviour but not a reproducer. On those grounds will skip this this week as well.

nagisa (Apr 25 2019 at 13:21, on Zulip):

Change powerpc64 base CPU #59932

Seems like this just needs discussion from stakelholders in PPC support, none of which (other than me, I guess) anybody on T-compiler are? Inclined to denominate.

nagisa (Apr 25 2019 at 13:25, on Zulip):

Decouple nightly RLS from Clippy #59761

Skipping, as it appears to be mostly T-dev-tools thing.

nagisa (Apr 25 2019 at 13:28, on Zulip):

reached the type-length limit while instantiating #58952

Regression caused by changes to stdlib. Just like with recursion limit and stack limit in compiler… it seems like this is an inherent limitation in the compiler that we cannot resolve easily/quickly.

nagisa (Apr 25 2019 at 13:28, on Zulip):

Inclined to just skip.

nagisa (Apr 25 2019 at 13:29, on Zulip):

mir-opt tests extremely slow. #58485

They are slow, but what is the exact source of slowness?

nagisa (Apr 25 2019 at 13:30, on Zulip):

x.py in incremental mode still rebuilds stage0-rustc if stage0-std changed. #54712

Nominated for resource allocation, so will return to it during the proper meeting.

nagisa (Apr 25 2019 at 13:31, on Zulip):

Exponential compile-time and type_length_limit blowup when nesting closure wrappers #54540

Seems to be the same as reached the type-length limit while instantiating #58952 from before.

nagisa (Apr 25 2019 at 13:33, on Zulip):

The regressions have been looked at as part of inspection of the I-nominated issues.

nikomatsakis (Apr 25 2019 at 13:55, on Zulip):

ICE on higher-trait bounds #60218

Bisected, nikomatsakis’s universes PR got blamed. It has been a while since it landed, which suggests people do not encounter this often. Still P-high?

Seems P-high to me. Presently it's just nominated. I guess I will try to investigate tomorrow, let's assign to me for now.

Last update: Nov 20 2019 at 02:30UTC