Stream: t-compiler/wg-prioritization/alerts

Topic: I-prioritize #73980 ICE when using associated constant in ty


triagebot (Jul 03 2020 at 09:47, on Zulip):

@WG-prioritization/alerts issue #73980 has been requested for prioritization.

Procedure

Yuki Okushi (Jul 03 2020 at 15:17, on Zulip):

according to the godbolt, it regressed since 1.22.0. and seems it's related to const-eval and cyclic reference. P-high for it?

lcnr (Jul 03 2020 at 15:25, on Zulip):

Yeah, at least P-high imo, I would go for P-critical if it didn't went unnoticed since version 1.22

lcnr (Jul 03 2020 at 15:29, on Zulip):

Ahhh, nm... that's not really P-high. This is just a problem with using generics in array lengths which relies on lazy normalization

lcnr (Jul 03 2020 at 15:29, on Zulip):

Let me quickly get the relevants duplicate issues here

lcnr (Jul 03 2020 at 16:26, on Zulip):

I don't really think we should try to fix this before lazy_normalization_consts is stable :thinking:
So P-mediummaybe, don't know how to best handle this.

lcnr (Jul 03 2020 at 16:26, on Zulip):

Opened #74000 to separate lazy_normalization_consts from const_generics
will slightly adjust my priorities after looking at this for a bit

lcnr (Jul 03 2020 at 16:27, on Zulip):

But it might be a good idea to fix name resolution to at least prevent the ICE

Yuki Okushi (Jul 03 2020 at 16:27, on Zulip):

thanks for investigating! sounds good, no objections for that prio

triagebot (Jul 06 2020 at 15:31, on Zulip):

Issue #73980's prioritization request has been removed.

LeSeulArtichaut (Jul 06 2020 at 15:32, on Zulip):

(Assigned P-medium)

Last update: Apr 15 2021 at 02:45UTC