Change the return type of `Ty::kind` from `&TyKind` to `TyKind`. #126069

nnethercote · 2024-06-06T10:17:05Z

This is valid because TyKind impls Copy, and makes Ty consistent with Predicate and Region, both of which have kind methods that return a non-reference.

The change removes more than one thousand & and * sigils, while requiring the addition of just five! It's clearly moving with the grain of the existing code.

Almost all of the removed sigils fit one of the following three patterns.

Remove the dereference in the match expression:

match *ty.kind() { ... } ->
match ty.kind() { ... }
Remove the dereference in the match pattern:

match ty.kind() { &ty::Foo(foo) => { ... foo ... } ->
match ty.kind() { ty::Foo(foo) => { ... foo ... }
Remove the derefernce in the match arm:

match ty.kind() { ty::Foo(foo) => { ... *foo ... } ->
match ty.kind() { ty::Foo(foo) => { ... foo ... }

A couple of other methods had their signatures changed.

TypeErrCtxt::cmp_fn_sig: & removed from two args.
EncodableWithShorthand::variant: & removed from return type.

r? @ghost

rustbot · 2024-06-06T10:17:11Z

Some changes occurred in compiler/rustc_codegen_cranelift

cc @bjorn3

Some changes occurred in compiler/rustc_codegen_gcc

cc @antoyo, @GuillaumeGomez

Some changes occurred to the CTFE / Miri engine

cc @rust-lang/miri

Some changes occurred to MIR optimizations

cc @rust-lang/wg-mir-opt

changes to the core type system

cc @compiler-errors, @lcnr

Some changes occurred in match lowering

cc @Nadrieril

The Miri subtree was changed

cc @rust-lang/miri

changes to the core type system

cc @compiler-errors, @lcnr

Some changes occurred to the core trait solver

cc @rust-lang/initiative-trait-system-refactor

Some changes occurred in rustc_ty_utils::consts.rs

cc @BoxyUwU

Some changes occurred in match checking

cc @Nadrieril

changes to the core type system

cc @compiler-errors, @lcnr

Some changes occurred in src/tools/clippy

cc @rust-lang/clippy

Some changes occurred in exhaustiveness checking

cc @Nadrieril

HIR ty lowering was modified

cc @fmease

Some changes occurred in compiler/rustc_sanitizers

cc @rust-lang/project-exploit-mitigations, @rcvalle

Some changes occurred in need_type_info.rs

cc @lcnr

lcnr · 2024-06-06T10:23:05Z

this needs perf. I remember trying this previously and getting a performance regression

RalfJung · 2024-06-06T10:23:48Z

The commit message still describes the change the wrong way around (lcnr fixed it in the PR title). I got quite confused due to that when reading the notification email. :)

nnethercote · 2024-06-06T10:27:57Z

I really wish it was possible to mark a PR as a draft while creating it and not notifying everybody until it's in a non-draft state.

lcnr · 2024-06-06T10:32:32Z

I really wish it was possible to mark a PR as a draft while creating it and not notifying everybody until it's in a non-draft state.

That should be possible 🤔 or do you mean something else?

nnethercote · 2024-06-06T11:33:05Z

@bors try @rust-timer queue

bors · 2024-06-06T11:34:18Z

⌛ Trying commit 25e854a with merge b3b9397...

Change the return type of `Ty::kind` from `&TyKind` to `TyKind`. This is valid because `TyKind` impls `Copy`, and makes `Ty` consistent with `Predicate` and `Region`, both of which have `kind` methods that return a non-reference. The change removes more than one thousand `&` and `*` sigils, while requiring the addition of just five! It's clearly moving with the grain of the existing code. Almost all of the removed sigils fit one of the following three patterns. - Remove the dereference in the match expression: `match *ty.kind() { ... }` -> `match ty.kind() { ... }` - Remove the dereference in the match pattern: `match ty.kind() { &ty::Foo(foo) => { ... foo ... }` -> `match ty.kind() { ty::Foo(foo) => { ... foo ... }` - Remove the derefernce in the match arm: `match ty.kind() { ty::Foo(foo) => { ... *foo ... }` -> `match ty.kind() { ty::Foo(foo) => { ... foo ... }` A couple of other methods had their signatures changed. - `TypeErrCtxt::cmp_fn_sig`: `&` removed from two args. - `EncodableWithShorthand::variant`: `&` removed from return type. r? `@ghost`

bors · 2024-06-06T13:10:44Z

☀️ Try build successful - checks-actions
Build commit: b3b9397 (b3b93972dc3a27bf8b67074e3e737a4a686bd669)

rust-timer · 2024-06-06T15:20:58Z

Finished benchmarking commit (b3b9397): comparison URL.

Overall result: ❌ regressions - ACTION NEEDED

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please fix the regressions and do another perf run. If the next run shows neutral or positive results, the label will be automatically removed.

@bors rollup=never
@rustbot label: -S-waiting-on-perf +perf-regression

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.4%	[0.2%, 0.9%]	45
Regressions ❌ (secondary)	0.7%	[0.4%, 1.2%]	36
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.4%	[0.2%, 0.9%]	45

Max RSS (memory usage)

This benchmark run did not return any relevant results for this metric.

Cycles

Results (secondary 3.2%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	3.2%	[3.1%, 3.4%]	3
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 672.761s -> 673.707s (0.14%)
Artifact size: 319.40 MiB -> 319.45 MiB (0.02%)

mati865 · 2024-06-06T16:47:33Z

I really wish it was possible to mark a PR as a draft while creating it and not notifying everybody until it's in a non-draft state.

That should be possible 🤔 or do you mean something else?

I think you can create easily make draft PR but rustbot won't distinguish it from "ready PR", or at least that used to be the case.

lcnr · 2024-08-14T06:46:45Z

going through my list of assigned PRs today.

@rustbot author

This is valid because `TyKind` impls `Copy`, and makes `Ty` consistent with `Predicate` and `Region`, both of which have `kind` methods that return a non-reference. The change removes more than one thousand `&` and `*` sigils, while requiring the addition of just five! It's clearly moving with the grain of the existing code. Almost all of the removed sigils fit one of the following three patterns. - Remove the dereference in the match expression: `match *ty.kind() { ... }` -> `match ty.kind() { ... }` - Remove the dereference in the match pattern: `match ty.kind() { &ty::Foo(foo) => { ... foo ... }` -> `match ty.kind() { ty::Foo(foo) => { ... foo ... }` - Remove the derefernce in the match arm: `match ty.kind() { ty::Foo(foo) => { ... *foo ... }` -> `match ty.kind() { ty::Foo(foo) => { ... foo ... }` A couple of other methods had their signatures changed. - `TypeErrCtxt::cmp_fn_sig`: `&` removed from two args. - `EncodableWithShorthand::variant`: `&` removed from return type.

nnethercote · 2024-08-14T09:58:51Z

@bors try @rust-timer queue

Change the return type of `Ty::kind` from `&TyKind` to `TyKind`. This is valid because `TyKind` impls `Copy`, and makes `Ty` consistent with `Predicate` and `Region`, both of which have `kind` methods that return a non-reference. The change removes more than one thousand `&` and `*` sigils, while requiring the addition of just five! It's clearly moving with the grain of the existing code. Almost all of the removed sigils fit one of the following three patterns. - Remove the dereference in the match expression: `match *ty.kind() { ... }` -> `match ty.kind() { ... }` - Remove the dereference in the match pattern: `match ty.kind() { &ty::Foo(foo) => { ... foo ... }` -> `match ty.kind() { ty::Foo(foo) => { ... foo ... }` - Remove the derefernce in the match arm: `match ty.kind() { ty::Foo(foo) => { ... *foo ... }` -> `match ty.kind() { ty::Foo(foo) => { ... foo ... }` A couple of other methods had their signatures changed. - `TypeErrCtxt::cmp_fn_sig`: `&` removed from two args. - `EncodableWithShorthand::variant`: `&` removed from return type. r? `@ghost`

bors · 2024-08-14T10:00:08Z

⌛ Trying commit bd3667d with merge 342c2c0...

bors · 2024-08-14T11:58:13Z

☀️ Try build successful - checks-actions
Build commit: 342c2c0 (342c2c0d4f0bb38271cf24e102ffd76bc2e9bca5)

rust-timer · 2024-08-14T13:13:22Z

Finished benchmarking commit (342c2c0): comparison URL.

Overall result: ❌✅ regressions and improvements - ACTION NEEDED

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please fix the regressions and do another perf run. If the next run shows neutral or positive results, the label will be automatically removed.

@bors rollup=never
@rustbot label: -S-waiting-on-perf +perf-regression

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.5%	[0.3%, 0.7%]	45
Regressions ❌ (secondary)	0.9%	[0.3%, 1.9%]	50
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.4%	[-0.4%, -0.2%]	12
All ❌✅ (primary)	0.5%	[0.3%, 0.7%]	45

Max RSS (memory usage)

Results (secondary -3.1%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.1%	[-3.1%, -3.1%]	1
All ❌✅ (primary)	-	-	0

Cycles

This benchmark run did not return any relevant results for this metric.

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 751.474s -> 753.088s (0.21%)
Artifact size: 341.45 MiB -> 341.40 MiB (-0.01%)

nnethercote · 2024-08-14T23:58:04Z

I had hoped that shrinking TyKind in #128812 might have helped with the regression, but it didn't make much difference. I have filed #129110 to add a comment so that nobody else spends time trying this again in the future.

…, r=jieyouxu Add a comment explaining the return type of `Ty::kind`. At least we'll get a useful comment out of rust-lang#126069 :) r? `@lcnr`

…, r=jieyouxu Add a comment explaining the return type of `Ty::kind`. At least we'll get a useful comment out of rust-lang#126069 :) r? ``@lcnr``

…, r=jieyouxu Add a comment explaining the return type of `Ty::kind`. At least we'll get a useful comment out of rust-lang#126069 :) r? ```@lcnr```

…, r=jieyouxu Add a comment explaining the return type of `Ty::kind`. At least we'll get a useful comment out of rust-lang#126069 :) r? ````@lcnr````

Rollup merge of rust-lang#129110 - nnethercote:Ty-kind-ret-ty-comment, r=jieyouxu Add a comment explaining the return type of `Ty::kind`. At least we'll get a useful comment out of rust-lang#126069 :) r? ````@lcnr````

nnethercote marked this pull request as draft June 6, 2024 10:17