Stream: t-compiler

Topic: 1.36 regressions

Pietro Albini (Jun 18 2019 at 20:25, on Zulip):

@nikomatsakis @pnkfelix there are 14 open regression issues, 12 of them marked T-compiler

Pietro Albini (Jun 18 2019 at 20:25, on Zulip):

and the release is in a bit more than 2 weeks :/

Pietro Albini (Jun 18 2019 at 20:26, on Zulip):

pnkfelix (Jun 18 2019 at 20:36, on Zulip):

I’ll try to go through these on my own tomorrow, in order to figure out which ones need to be subjects at the T-compiler meeting on Thursday

pnkfelix (Jun 18 2019 at 20:36, on Zulip):

(Luckily I am not at the All-Hands so I can actually commit to such a task)

centril (Jun 18 2019 at 21:34, on Zulip):

I'm going to just close outright because I don't see the language team accepting this back.

davidtwco (Jun 18 2019 at 22:00, on Zulip):

Is that the correct link? It goes to a months old PR for me.

centril (Jun 18 2019 at 22:32, on Zulip):

@davidtwco oops; I meant

pnkfelix (Jun 20 2019 at 10:24, on Zulip):

I just realized I should probably be logging my actions on these issues somewhere

pnkfelix (Jun 20 2019 at 10:24, on Zulip):

so lets see

pnkfelix (Jun 20 2019 at 10:25, on Zulip):

"ICE while bootstrapping (./ build)" #59945 -- I closed this yesterday as "not enough information to reproduce nor act."

pnkfelix (Jun 20 2019 at 10:27, on Zulip):

"regression: overflow evaluating, Send/Sync?" #61472 -- I confirmed that fuzzy-pickles and conch-runtime were resolved on nightly at approx same time that PR #61754 landed. I have also nominated PR #61754 for a beta backport

pnkfelix (Jun 20 2019 at 10:32, on Zulip):

"regression: conflicting implementation for &_" #61473 -- closed as "fixed by upstream crate patches", as in, cargo update should suffice. (Maybe crater should include this as something it automatically attempts?)

pnkfelix (Jun 20 2019 at 10:40, on Zulip):

"regression: conflicting implementations for From<&_>" #61474 -- closed as "fixed by upstream crate update". But in this case, I don't think cargo update suffices for the fix, because the upstream update was not a patch release.

pnkfelix (Jun 20 2019 at 12:23, on Zulip):

"regression: trait no longer implemented" #61563 -- posted a lot a comments in the issue. Basically this was a subtle issue of a previously untested doc example that started being tested and thus exposed code that does not compile. Filed an issue with the originating crate and closed as not-a-bug.

pnkfelix (Jun 20 2019 at 13:09, on Zulip):

"regression: type annotations required cannot resolve std::borrow::Cow<'_, str>: std::borrow::Borrow<_>" #61476 -- closed as duplicate of #60958

pnkfelix (Jun 21 2019 at 14:51, on Zulip):

"Panic with Layout mismatch when copying! from macro" #61663 -- very very weird; see comments on issue. Bisected injection down to PR #59560. assigned to @Matthew Jasper and self. Removed nomination tag.

oli (Jun 21 2019 at 14:54, on Zulip):

I'm reading it ever since you wrote which PR caused it (I reviewed it after all)

oli (Jun 21 2019 at 14:55, on Zulip):

The only entry point I have so far is that enum discriminant computations are always confusing to me since everything is isize until actually used

Last update: May 24 2020 at 23:30UTC