T-compiler Meeting Agenda 2023-01-05

Announcements

• Reminder: if you see a PR/issue that seems like there might be legal implications due to copyright/IP/etc, please let the Core team know (or at least message @pnkfelix or @Wesley Wiser so we can pass it along).

Other WG meetings (calendar link)

• wg-async weekly at time:2023-01-05T12:00:00-05:00
• wg-rls-2.0 weekly sync-up at time:2023-01-09T10:00:00-05:00
• [Types team] Shallow subtyping weekly meeting at time:2023-01-09T10:00:00-05:00
• [wg-traits] GATs Sync at time:2023-01-09T16:00:00-05:00

MCPs/FCPs

• New MCPs (take a look, see if you like them!)
  • No new proposals this time.
• Old MCPs (not seconded, take a look)
  • “Add #[alias] attribute to allow symbol aliasing” compiler-team#526 (last review activity: 5 months ago)
  • " Promote i586-unknown-linux-gnu to Tier 2 with Host Tools " compiler-team#543 (last review activity: 4 months ago)
  • “Lower baseline expectations for i686 unix-like targets” compiler-team#548 (last review activity: about 12 days ago)
  • “configurable rustc timeout for compiletest tests” compiler-team#554 (last review activity: 3 months ago)
    • Stale MCP: candidate for closing soon
  • “Rustc Contributor Program Major Change Proposal” compiler-team#557 (last review activity: 2 months ago)
    • Stale MCP: candidate for closing soon (actually the initiative moved forward without MCP)
  • “Introduce a #[rustc::*] tool attribute namespace and provide a stable #[rustc::on_unimplemented] attribute” compiler-team#560 (last review activity: 2 months ago)
    • Replaced by T-lang RFC, maybe can be closed
  • “-Z instrument-xray option” compiler-team#561 (last review activity: 2 months ago)
  • “Upgrade *-linux-musl targets to musl 1.2” compiler-team#572 (last review activity: about 13 days ago)
  • “New tier-2 target for wasm32-wasi + threads” compiler-team#574 (last review activity: about 0 days ago)
  • “needs_drop as an auto trait” compiler-team#575 (last review activity: about 0 days ago)
• Pending FCP requests (check your boxes!)
  • “Add SEMICOLON_IN_EXPRESSIONS_FROM_MACROS to future-incompat report” rust#103418
  • “Add deployment-target –print flag for Apple targets” rust#105354
• Things in FCP (make sure you’re good with it)
  • No FCP requests this time.
• Accepted MCPs
  • “Move tests out of src” compiler-team#573
• Finalized FCPs (disposition merge)
  • “Tracking issue for the “efiapi” calling convention” rust#65815
  • “Make PROC_MACRO_DERIVE_RESOLUTION_FALLBACK a hard error” rust#84022
  • “Encode spans relative to the enclosing item – enable on nightly” rust#84762
  • “Stabilize #![feature(target_feature_11)]” rust#99767
  • “Make Sized coinductive, again” rust#100386
  • “Elaborate supertrait obligations when deducing closure signatures” rust#101834
  • “Don’t normalize in AstConv” rust#101947
  • “make const_err a hard error” rust#102091
  • “Stabilize default_alloc_error_handler” rust#102318
  • “make order_dependent_trait_objects show up in future-breakage reports” rust#102635
  • “Stabilize raw-dylib for non-x86” rust#102793
  • “Use token::Lit in ast::ExprKind::Lit.” rust#102944
  • “Remove drop order twist of && and || and make them associative” rust#103293
  • “Unreserve braced enum variants in value namespace” rust#103578
  • “Remove const eval limit and implement an exponential backoff lint instead” rust#103877
  • “Change bindings_with_variant_name to deny-by-default” rust#104154
  • “Stabilize native library modifier verbatim” rust#104360
  • “More deriving on packed structs” rust#104429
  • “Expand a style-guide principle: readability in plain text” rust#104506
  • “Find the right lower bound region in the scenario of partial order relations” rust#104765
  • “Stop promoting all the things” rust#105085
  • “Add IMPLIED_BOUNDS_ENTAILMENT lint” rust#105575

WG checkins

None

Backport nominations

T-compiler beta / T-compiler stable

• No beta nominations for T-compiler this time.
• :stable: “Don’t copy symbols from dylibs with -Zdylib-lto” rust#105800
  • backport nominated by @lqd
  • patch addresses (among others) rust#105637 a P-critical regression impacting bug reports on ICEs
  • beta backport previously approved (Zulip link)
    • stable backport was on hold to wait for some feedback once it lands on beta
    • or just let it ride and land on stable (Jan, 26th)

T-rustdoc beta / T-rustdoc stable

• No backport nominations for T-rustdoc this time.

:back: / :shrug: / :hand:

PRs S-waiting-on-team

T-compiler

• “Remove proc-macro back-compat hack for rental” rust#106060
  • Waiting for team but @Esteban Küber and @Aaron Hill already contributed some comments
• (other hidden issues in progress or waiting on other teams, details here)

Oldest PRs waiting for review

T-compiler

• “Add SEMICOLON_IN_EXPRESSIONS_FROM_MACROS to future-incompat report” rust#103418 (last review activity: 2 months ago)
  • The FCP needs some more ticks to progress. As mentioned in comment, T-lang can review but it’s not blocked on them
• “Parse unnamed struct and union fields” rust#99754 (last review activity: about 54 days ago)
  • cc: @cjgillot
• “TokenTreesReader now can find the correct mismatch delimiter pairs” rust#104012 (last review activity: about 54 days ago)
  • cc @Esteban Küber
• “Add Trusty OS as tier 3 target” rust#103895 (last review activity: about 50 days ago)
  • Does this proposal need an MCP? (Left a comment a little while ago)
• “Don’t drain-on-drop in DrainFilter impls of various collections.” rust#104455 (last review activity: about 50 days ago)
  • Some bits are under ./compiler (reviewable from T-compiler?)

Issues of Note

Short Summary

• 4 T-compiler P-critical issues
  • 3 of those are unassigned
• 48 T-compiler P-high issues
  • 18 of those are unassigned
• 2 P-critical, 0 P-high, 1 P-medium, 0 P-low regression-from-stable-to-beta
• 1 P-critical, 3 P-high, 3 P-medium, 0 P-low regression-from-stable-to-nightly
• 2 P-critical, 28 P-high, 89 P-medium, 8 P-low regression-from-stable-to-stable

P-critical

T-compiler

• “Seg fault in Rust 1.65.0 if I don’t create temporary variable” rust#105295
  • @Michael Goulet (compiler-errors) is following up on this with a lint (#105572) and then a new PR to promote it to an error (comment)
• “where_clauses_object_safety regression on a trait where it shouldn’t trigger” rust#106247
  • a revert PR landed with #106248
  • a fix PR by @Gary Guo is open at #106253, review assigned to @Michael Goulet (compiler-errors)
• “regression: anonymous lifetimes now unstable in impl trait” rust#106338
  • a crater run in latest beta found these regressions
  • bisection seems to point to #104048 cc: @cjgillot

T-types

• “regression: non-defining opaque type use in defining scope” rust#105826
  • regression in beta, some crates fail building
  • seems to point to #103491 cc @cjgillot (comment)
  • issue is I-types-nominated

T-rustdoc

• No P-critical issues for T-rustdoc this time.

P-high regressions

P-high beta regressions

• No P-high beta regressions this time.

Unassigned P-high nightly regressions

• “deref_target_place is not a deref projection” rust#105881
  • should be fixed by rust#105958 (reviewed and in progress)

Performance logs

triage logs for 2023-01-03

Fairly busy week with some massive performance improvements at the expense of some significant albeit smaller regressions. The main wins came in a long-standing PR from @cjgillot to enable encoding spans in metadata relative to their enclosing item. This causes more work in full compilation which causes some regressions up to 5% but can lead to very large wins in incremental compilation scenarios (up to ~70%). For example, the clap crate compiles 68% faster after a small 1 line change than it did previously.

Triage done by @rylev. Revision range: b38a6d..b43596

Summary:

1 Regressions, 1 Improvements, 4 Mixed; 1 of them in rollups 47 artifact comparisons made in total

Regressions

Rollup of 8 pull requests #106228 (Comparison Link)

• Kicked off a perf run on the only likely culprit. The regression is small enough that I don’t think it’s a big deal if we don’t figure this out.

  Allow trait method paths to satisfy const Fn bounds #106210 (Comparison Link)

• Massive amount of noise reversed in #106268. No idea where this noise comes from.

Improvements

fix comment for TokenCursor::desugar #106268 (Comparison Link)

• Massive amount of noise coming in #106210 reversed here. No idea where this noise comes from.

  Abolish QueryVTable in favour of more assoc items on QueryConfig #106307 (Comparison Link)

Mixed

Catch panics/unwinding in destruction of TLS values #105426 (Comparison Link)

• Very much unsure where the performance changes are coming from. Some show LLVM regressions but not all. LLVM also does not explain the improvements.

• Ask in the PR for thoughts.

  Use some more const_eval_select in pointer methods for compile times #106275 (Comparison Link)

• Author investigated why this might have caused a regression in regex and found nothing. The gains much outweigh the regressions so we should be fine keeping it.

  Encode spans relative to the enclosing item – enable on nightly #84762 (Comparison Link)

• There was an FCP to decide whether the trade off of some regressions for some huge incremental performance wins was worth it.

• This FCP passed and so the PR was merged.

  Reenable limited top-down MIR inlining #106364 (Comparison Link)

• As pointed out here this was effectively a performance wash with regressions and improvements canceling themselves out.

Nominated Issues

T-compiler

• No nominated issues for T-compiler this time.

RFC

• No nominated RFCs for T-compiler this time.

Next week’s WG checkins

• @_WG-self-profile by @mw and @Wesley Wiser