Stream: t-compiler/wg-prioritization

Topic: pre-meeting triage 2020-05-21 #54818


Santiago Pastorino (May 20 2020 at 19:01, on Zulip):

@WG-prioritization let's start the pre-triage

Santiago Pastorino (May 20 2020 at 19:05, on Zulip):

The triage meeting will happen in #t-compiler/meetings > [weekly meeting] 2020-05-21 #54818

Santiago Pastorino (May 20 2020 at 19:05, on Zulip):

sorry for the delay I was adjusting some things

Santiago Pastorino (May 20 2020 at 19:06, on Zulip):

I have created the agenda we will be building during this meeting

Santiago Pastorino (May 20 2020 at 19:06, on Zulip):

I've rebumped a ton the script so let's get started

Santiago Pastorino (May 20 2020 at 19:07, on Zulip):
[santiago@galago triagebot (prioritization-meeting-automation)]$ GITHUB_API_TOKEN=<HIDDEN> cargo run --bin prioritization
   Compiling triagebot v0.1.0 (/home/santiago/src/oss/triagebot)
    Finished dev [unoptimized + debuginfo] target(s) in 3.45s
     Running `target/debug/prioritization`
## Unprioritized I-prioritize

1. [All unprioritized I-prioritize](https://github.com/rust-lang/rust/issues?q=is%3Aopen+is%3Aissue+-label%3AP-critical+-label%3AP-high+-label%3AP-medium+-label%3AP-low+label%3AI-prioritize)
    - Add T-compiler or libs-impl tag when it corresponds.
    - Do a sanity check on them.

### Issues

Empty

2. [T-compiler](https://github.com/rust-lang/rust/issues?q=is%3Aopen+is%3Aissue+label%3AT-compiler+-label%3AP-critical+-label%3AP-high+-label%3AP-medium+-label%3AP-low+label%3AI-prioritize)
    - Prioritize issues and remove nomination of the ones not worth discussing.
    - Tag regressions accordingly.
    - Ping appropriate people and/or [ICE-breakers](https://rustc-dev-guide.rust-lang.org/ice-breaker/about.html#tagging-an-issue-for-an-ice-breaker-group).

### Issues

Empty

3. [libs-impl](https://github.com/rust-lang/rust/issues?q=is%3Aopen+is%3Aissue+label%3Alibs-impl+-label%3AP-critical+-label%3AP-high+-label%3AP-medium+-label%3AP-low+label%3AI-prioritize)
    - Prioritize issues and remove nomination of the ones not worth discussing.
    - Tag regressions accordingly.
    - Ping appropriate people and/or [ICE-breakers](https://rustc-dev-guide.rust-lang.org/ice-breaker/about.html#tagging-an-issue-for-an-ice-breaker-group).

### Issues

Empty

Press a key to continue ...
Santiago Pastorino (May 20 2020 at 19:08, on Zulip):

pasted that to show the input but well as always everything is prioritized

Santiago Pastorino (May 20 2020 at 19:08, on Zulip):

gonna paste things straight here so they look better

Santiago Pastorino (May 20 2020 at 19:08, on Zulip):

let's move to the next part

Santiago Pastorino (May 20 2020 at 19:08, on Zulip):

Regressions

  1. Beta regressions without P-label

    - Prioritize.
    - Ping appropriate people and/or ICE-breakers.
    - Assign if possible; if it remains unassigned, add it to agenda so we can assign during the meeting.

Issues

Empty

  1. Nightly regressions without P-label

    - Prioritize.
    - Ping appropriate people and/or ICE-breakers.
    - Assign if possible; if it remains unassigned, add it to agenda so we can assign during the meeting.

Issues

Empty

  1. Stable regressions without P-label

    - Prioritize (once we have this under control).
    - Ping appropriate people and/or ICE-breakers.
    - Assign if possible; if it remains unassigned, add it to agenda so we can assign during the meeting.

Issues

Press a key to continue ...

Santiago Pastorino (May 20 2020 at 19:09, on Zulip):

this means no beta regressions without p-label and no nightly regressions without p-label and given that we are already automatically adding i-prioritize labels to regressions nightly and beta checks without p-labels are not worth anymore

Santiago Pastorino (May 20 2020 at 19:09, on Zulip):

the list by this new definition will always be empty

Santiago Pastorino (May 20 2020 at 19:10, on Zulip):

but we have a bunch of stable-to-stable unprioritized regressions as we always had

Santiago Pastorino (May 20 2020 at 19:10, on Zulip):

what I want to propose is to start slowly to prioritize all those, not in this meeting but async, until this list is zero

Santiago Pastorino (May 20 2020 at 19:10, on Zulip):

all new stable to stable regressions are going to be tagged with I-prioritize

Santiago Pastorino (May 20 2020 at 19:10, on Zulip):

so we can slowly move this number to zero

Santiago Pastorino (May 20 2020 at 19:10, on Zulip):

does that sounds good?

Santiago Pastorino (May 20 2020 at 19:11, on Zulip):

if feels good, I'd suggest we tag 3 or 4 as I-prioritize per week until this number is zero :)

Santiago Pastorino (May 20 2020 at 19:11, on Zulip):

if someone wants to tag 3 or 4 now would be great :)

Santiago Pastorino (May 20 2020 at 19:12, on Zulip):

let's continue meanwhile

Santiago Pastorino (May 20 2020 at 19:12, on Zulip):

let's go with nominations now

Santiago Pastorino (May 20 2020 at 19:12, on Zulip):

Stable nominations

Issues

Empty

Beta nominations

Issues

Santiago Pastorino (May 20 2020 at 19:13, on Zulip):

we have one beta nominations and zero stable nominations

Santiago Pastorino (May 20 2020 at 19:13, on Zulip):

beta nomination is rustdoc

Santiago Pastorino (May 20 2020 at 19:14, on Zulip):

I think we should remove from all our searches T-rustdoc, T-release and T-infra, just in case, can you confirm this is ok @pnkfelix ?

Santiago Pastorino (May 20 2020 at 19:14, on Zulip):

I think we have talked about this but I seem to keep forgetting

Santiago Pastorino (May 20 2020 at 19:14, on Zulip):

if I have confirmation I can modify script and leave these out forever :)

Santiago Pastorino (May 20 2020 at 19:14, on Zulip):

let's continue

Santiago Pastorino (May 20 2020 at 19:14, on Zulip):

I-nominated

  1. All I-nominated

    - Add T-compiler or libs-impl tag when it corresponds.
    - Do a sanity check on them.

Issues

Santiago Pastorino (May 20 2020 at 19:15, on Zulip):

remember that that's all i-nominated

Santiago Pastorino (May 20 2020 at 19:15, on Zulip):

let's see if there's something that needs to be tagged with T-compiler or libs-impl there

Santiago Pastorino (May 20 2020 at 19:16, on Zulip):

it seems like all the issues have teams assigned

Santiago Pastorino (May 20 2020 at 19:16, on Zulip):

let's move then

Santiago Pastorino (May 20 2020 at 19:16, on Zulip):
  1. I-nominated T-compiler

    - Remove leftovers from last meeting.
    - Do a sanity check on them.

Issues

Santiago Pastorino (May 20 2020 at 19:17, on Zulip):

actually one important thing

Santiago Pastorino (May 20 2020 at 19:17, on Zulip):

last meeting was cancelled so not only we don't have leftovers

Santiago Pastorino (May 20 2020 at 19:17, on Zulip):

but we also need to merge the two agendas

Santiago Pastorino (May 20 2020 at 19:18, on Zulip):

let's sanity check those ...

Santiago Pastorino (May 20 2020 at 19:19, on Zulip):

repr one keeps showing up but it's about t-lang really

Santiago Pastorino (May 20 2020 at 19:20, on Zulip):

the problem is that it's tagged with both t-lang and t-compiler

Santiago Pastorino (May 20 2020 at 19:20, on Zulip):

I also need to check if we've made a definition about this with @pnkfelix

Santiago Pastorino (May 20 2020 at 19:20, on Zulip):

let's continue

Santiago Pastorino (May 20 2020 at 19:20, on Zulip):
  1. I-nominated libs-impl

    - Remove leftovers from last meeting.
    - Do a sanity check on them.

Issues

Santiago Pastorino (May 20 2020 at 19:21, on Zulip):

again, there are no leftovers, let's sanity check

Santiago Pastorino (May 20 2020 at 19:21, on Zulip):

this is ok

Santiago Pastorino (May 20 2020 at 19:22, on Zulip):

let's move on

Santiago Pastorino (May 20 2020 at 19:22, on Zulip):

PR's waiting for on team

  1. All PR's waiting on team

    - Add T-compiler or libs-impl tag when it corresponds.
    - Do a sanity check on them.

Issues

  1. PR's waiting on T-compiler

    - Explicitly nominate any that you think may be able to be resolved quickly in triage meeting.

Issues

Empty

  1. PR's waiting on libs-impl

    - Explicitly nominate any that you think may be able to be resolved quickly in triage meeting.

Issues

Empty

Santiago Pastorino (May 20 2020 at 19:23, on Zulip):

it seems like there's only one PR waiting but on t-lang

Santiago Pastorino (May 20 2020 at 19:23, on Zulip):

checking just in case

Santiago Pastorino (May 20 2020 at 19:23, on Zulip):

yeah it seems like that

Santiago Pastorino (May 20 2020 at 19:24, on Zulip):

so, as you may have noted I've reorganized the meeting a bit different

Santiago Pastorino (May 20 2020 at 19:24, on Zulip):

and we are now done with the first part which I'd call preparation for the agenda :)

Santiago Pastorino (May 20 2020 at 19:25, on Zulip):

and we are now done with the first part which I'd call preparation for the agenda :)

Santiago Pastorino (May 20 2020 at 19:25, on Zulip):

next part the script dumps the agenda

LeSeulArtichaut (May 20 2020 at 19:28, on Zulip):

Btw @Santiago Pastorino did you put the script somewhere, e.g. on GitHub?

Santiago Pastorino (May 20 2020 at 19:28, on Zulip):

gonna push soon :)

Santiago Pastorino (May 20 2020 at 19:28, on Zulip):

I was having a little trouble, seems like something I've recently changed broke something related to the agenda

Santiago Pastorino (May 20 2020 at 19:28, on Zulip):

anyway, can fix it later

Santiago Pastorino (May 20 2020 at 19:29, on Zulip):

so the steps are:

Santiago Pastorino (May 20 2020 at 19:30, on Zulip):
  1. do work the work that will prepare everything to build an agenda (what we've just done)
Santiago Pastorino (May 20 2020 at 19:30, on Zulip):
  1. dump the agenda
Santiago Pastorino (May 20 2020 at 19:30, on Zulip):
  1. read how packed the agenda looks like and check if we can add more stuff
Santiago Pastorino (May 20 2020 at 19:30, on Zulip):

I couldn't properly dump the agenda

Santiago Pastorino (May 20 2020 at 19:30, on Zulip):

but

Santiago Pastorino (May 20 2020 at 19:31, on Zulip):

we can 100% sure tell that the agenda will be packed because we have what we had for last meeting plus this one :)

Santiago Pastorino (May 20 2020 at 19:31, on Zulip):

anyway, let's go through the steps just in case we find something outstanding

Santiago Pastorino (May 20 2020 at 19:34, on Zulip):

so next part would be ...

Santiago Pastorino (May 20 2020 at 19:34, on Zulip):

Nominate issues

Check how packed the agenda looks like and if there's room for more, consider the following ...

Santiago Pastorino (May 20 2020 at 19:35, on Zulip):

I don't see anything outstanding

Santiago Pastorino (May 20 2020 at 19:35, on Zulip):

given how packed the agenda is ...

Santiago Pastorino (May 20 2020 at 19:35, on Zulip):

Announcements

Santiago Pastorino (May 20 2020 at 19:35, on Zulip):

there's a design meeting on friday

Santiago Pastorino (May 20 2020 at 19:36, on Zulip):

that's all I think

Santiago Pastorino (May 20 2020 at 19:36, on Zulip):

Toolstate

Santiago Pastorino (May 20 2020 at 19:36, on Zulip):

clippy, miri, rls and rustfmt are broken, cc @DPC about this

Santiago Pastorino (May 20 2020 at 19:37, on Zulip):

would be great to think about this step, automating it in some way or something but I don't see us doing something useful here :)

Santiago Pastorino (May 20 2020 at 19:37, on Zulip):

let's move on

Santiago Pastorino (May 20 2020 at 19:37, on Zulip):

Performance regressions

DPC (May 20 2020 at 19:38, on Zulip):

release is on 2020-06-04 and I don't see any prs fixing the broken ones. Updating the respective teams

Wesley Wiser (May 20 2020 at 19:39, on Zulip):

The opt incremental regression is expected https://github.com/rust-lang/rust/pull/71840#issuecomment-625919572

Santiago Pastorino (May 20 2020 at 19:39, on Zulip):

about regressions I was going to say, I don't see anything "new"

Santiago Pastorino (May 20 2020 at 19:40, on Zulip):

there's some stuff we've already talked about

Santiago Pastorino (May 20 2020 at 19:40, on Zulip):

so I think this is more or less all

Santiago Pastorino (May 20 2020 at 19:40, on Zulip):

a couple of things before closing up

Santiago Pastorino (May 20 2020 at 19:41, on Zulip):

unsure how many people that do not understand this process is reading and trying to help with this effort but please send me a private message and I'd love to explain step by step

Santiago Pastorino (May 20 2020 at 19:42, on Zulip):

also, I'm going to build the agenda right after the meeting

Santiago Pastorino (May 20 2020 at 19:42, on Zulip):

going to work in the script and publish it somewhere, as a next step would be nice to polish it up a bit and integrating it with zulip bots

Santiago Pastorino (May 20 2020 at 19:43, on Zulip):

will share here once I have the agenda and the script, script won't be today :)

Santiago Pastorino (May 20 2020 at 19:43, on Zulip):

and last thing, if everyone agrees I'm going to change this meeting again to be a 1h meeting starting 1h later

Santiago Pastorino (May 20 2020 at 19:43, on Zulip):

because it's not 2hs long anymore fortunately :)

Santiago Pastorino (May 20 2020 at 19:44, on Zulip):

Santiago Pastorino said:

and last thing, if everyone agrees I'm going to change this meeting again to be a 1h meeting starting 1h later

@WG-prioritization :point_up:

Santiago Pastorino (May 20 2020 at 19:44, on Zulip):

that's all I guess, :wave:

DPC (May 20 2020 at 19:44, on Zulip):

so 20.00 UTC?

pnkfelix (May 20 2020 at 20:09, on Zulip):

sorry for my absence, let me try to follow up on the back scroll now

pnkfelix (May 20 2020 at 20:12, on Zulip):

Santiago Pastorino said:

I think we should remove from all our searches T-rustdoc, T-release and T-infra, just in case, can you confirm this is ok pnkfelix ?

we can probably remove T-release and T-infra. In general the whole "add T-compiler when it corresponds" was probably meant for issues that had no other team label. But given that we have agreed to take on beta-backport approval duties for T-libs, I do wonder if we implicitly have the same responsibility for T-rustdoc .... is there even an actual rustdoc team at this point?

Santiago Pastorino (May 20 2020 at 20:14, on Zulip):

unsure but this issue may be important then ...

Santiago Pastorino (May 20 2020 at 20:15, on Zulip):

@pnkfelix https://rust-lang.zulipchat.com/#narrow/stream/227806-t-compiler.2Fwg-prioritization/topic/I-prioritize.20.2372340.20Regression.3A.20.60.5BBody.3A.3Aempty.5D.60.20cannot.20be.20re

pnkfelix (May 20 2020 at 20:17, on Zulip):

Santiago Pastorino said:

repr one keeps showing up but it's about t-lang really

lang team is finally done with that topic, I think. I left a comment on issue #27060 and am removing the nomination.

pnkfelix (May 20 2020 at 20:24, on Zulip):

pnkfelix said:

But given that we have agreed to take on beta-backport approval duties for T-libs, I do wonder if we implicitly have the same responsibility for T-rustdoc .... is there even an actual rustdoc team at this point?

Okay I checked in discord. There is a skeleton crew of people working to fix rustdoc when they can, but we should not rely on them for things like approving backports or evaluating regressions

Santiago Pastorino (May 21 2020 at 13:16, on Zulip):

@WG-prioritization yesterday I had to leave quickly and forgot to mention that the agenda was ready

Santiago Pastorino (May 21 2020 at 13:16, on Zulip):

https://hackmd.io/OL02gcLSR96Qrym6-n5qmA?view

Santiago Pastorino (May 21 2020 at 13:16, on Zulip):

in case people want to take a look at it

Santiago Pastorino (May 21 2020 at 13:16, on Zulip):

would be nice to explain why nominations are there

Santiago Pastorino (May 21 2020 at 13:17, on Zulip):

going to do that now so we have this ready before the meeting

Last update: Jun 05 2020 at 23:10UTC