T-lang planning meeting agenda

• Meeting date: 2021-07-07

Attendance

• Team members: nikomatsakis, Josh, Felix
• Others: simulacrum

Meeting roles

• Action item scribe: simulacrum
• Note-taker:

Announcements and other updates

(Anyone can feel free to add things here, tag it with your username.)

• nikomatsakis: Josh and I plan to write up the new "initiative" process that we discussed as a series of edits to our lang-team web page. I am also going to make various proposals around the current set of active initiatives as we go.

Proposed meetings

• "Structural equality" lang-team#94
• "Never allow unwinding from Drop impls" lang-team#97
• "Lang team process, part 2" lang-team#104

• July 14 – no meeting (also, Felix has a conflict)

• July 21 – Lang team process, part 2

• July 28 – Structural equality (depending on Oli's scheduling constraints; may move to the 27th)

• August 4 – next planning meeting, Niko on PTO

• August 11 – Niko on PTO

• Action item: update the issues and announce the meetings

Notes

• For lang-team#97:
  • For this meeting to be a success, we would need some exploration of what the alternatives are and what kinds of backwards compatibility concerns exist in the wild.
    • Are people relying on the current behavior?
    • Can we estimate how many latent bugs exist because of the current behavior?
      • Anecdotally: Rayon for example found it too hard to manage and just aborts.
      • Libs team has found this painful in the std library.
    • Josh to summarize.
• For lang-team#94:
  • oli can prepare the write-up, nikomatsakis is response to ensure it gets done
  • Problem is how to deal with matching on constants, const generic equality, etc in a coherent way
  • Scheduled to July 21
• For lang-team#104:

Active initiatives

• Planning board

• Umbrella initiatives:
  *

"const-evaluation" lang-team#22

Link: https://github.com/rust-lang/lang-team/issues/22

• Comment
• Inline constants #76001 have not had any substantive progress
  • This is a blocker for const support in inline assembly
  • What does it take to make forward progress here?
  • What work has been done here?
  • Action item: follow up on the status and discuss next steps (nikomatsakis)

"async foundations" lang-team#33

Link: https://github.com/rust-lang/lang-team/issues/33

• Comment

"const-generics" lang-team#51

Link: https://github.com/rust-lang/lang-team/issues/51

• Update
• Vision doc in progress here
• Upcoming design meeting:
  • Structural equality

"project-safe-transmute" lang-team#21

Link: https://github.com/rust-lang/lang-team/issues/21

• Update

"Deref patterns" lang-team#88

Link: https://github.com/rust-lang/lang-team/issues/88

• Update
• nrc recently opened a bunch of threads and seems to have an interest

let-else statements

Link: https://github.com/rust-lang/rfcs/pull/3137

• Discussed in triage yesterday

macro metavariable

Link: https://github.com/rust-lang/rust/issues/83527

• Update
• Ping author regarding "vacation period"

"RFC 2229" lang-team#50

• Update
• Closure size, after optimization
  • Closure size, before optimization
• Maybe automate the capturing of data instead, rather than asking people to submit
  • Then we can use crater instead

"never type stabilization" lang-team#60

Link: https://github.com/rust-lang/lang-team/issues/60

• Update (1/2)
• Update (2/2)
• Plan to produce some sort of write-up about what we do
  • One question is to what extent we want to document the algorithm
  • Also, where? There isn't much to fit it into
• Action item: Mark to kickoff a conversation about the documentation aspect

"ffi-unwind" lang-team#19

Link: https://github.com/rust-lang/lang-team/issues/19

• Update

Tracking Issue for denying trailing semicolons in expression macro bodies #79813

Link: https://github.com/rust-lang/rust/issues/79813

• Update: Ongoing crater run

"inline-asm" lang-team#20

Link: https://github.com/rust-lang/lang-team/issues/20

• Update
• There is a subset that could be stabilized, modulo:
  • clobber support
  • namespacing decision being finalized

Tracking Issue for #[instruction_set] attribute (RFC 2867) #74727

Link: https://github.com/rust-lang/rust/issues/74727

• Hasn't been changes in a while
• Update from April 6 suggests:
  • The implementation works but
  • Generates suboptimal code owing to ungreat inlining
• Seems ready to stabilize if we wanted to
  • Minor bikeshed of #[repr(instruction_set = "...")] as an alternative spelling
• What would be the plan to get more feedback

Tracking issue for X.., ..X, and ..=X (#![feature(half_open_range_patterns)]) #67264

• Has a stabilization PR in FCP