Allow let bindings and destructuring in constants and const fn #49172

oli-obk · 2018-03-19T13:46:27Z

eddyb · 2018-03-24T16:16:02Z

src/librustc_mir/transform/qualify_consts.rs

@@ -120,7 +120,7 @@ struct Qualifier<'a, 'gcx: 'a+'tcx, 'tcx: 'a> {
    rpo: ReversePostorder<'a, 'tcx>,
    tcx: TyCtxt<'a, 'gcx, 'tcx>,
    param_env: ty::ParamEnv<'tcx>,
-    temp_qualif: IndexVec<Local, Option<Qualif>>,
+    local_qualif: IndexVec<Local, Option<Qualif>>,
    return_qualif: Option<Qualif>,


Can you move this into local_qualif?

eddyb · 2018-03-24T16:16:57Z

src/librustc_mir/transform/qualify_consts.rs

@@ -189,8 +189,16 @@ impl<'a, 'tcx> Qualifier<'a, 'tcx, 'tcx> {

    /// Error about extra statements in a constant.
    fn statement_like(&mut self) {
+        if self.tcx.sess.features_untracked().const_let {
+            return;


This seems bad, because statement_like is used for a lot of errors - it should maybe be split into several functions?

eddyb · 2018-03-24T16:19:47Z

src/librustc_mir/transform/qualify_consts.rs

+                                   self.tcx.sess.features_untracked().const_let => {
+                debug!("store to var {:?}", index);
+                match self.local_qualif[index] {
+                    Some(ref mut qualif) => *qualif &= self.qualif,


I think you want to replace the qualif always (e.g. let mut x = None; x = Some(TypeThatNeedsDrop);).
Also, if you have this special case, why do you need to do anything about statement_like?

eddyb · 2018-03-24T16:24:23Z

src/librustc_mir/transform/qualify_consts.rs

@@ -678,6 +700,10 @@ impl<'a, 'tcx> Visitor<'tcx> for Qualifier<'a, 'tcx, 'tcx> {
                    self.add(Qualif::STATIC_REF);
                }

+                if self.mode == Mode::ConstFn && self.tcx.sess.features_untracked().const_let {
+                    return;


This is bad because it breaks promotion in const fn. Also, you want mutable borrows to locals outside of const fn too, right? I think we should orthogonally allow &mut foo in regular static, so only const needs a special-case at the outermost level (and promotion also needs to keep disallowing &mut NotZST).

eddyb · 2018-03-24T16:44:22Z

@oli-obk If you want to get this merged quickly, I'd suggest dropping the borrowing changes.

cc @nikomatsakis A bit of a conundrum: static X: &'static mut Foo = &mut Foo {...}; would be fine to allow because static providing immutable access means it's effectively only &Foo.
However, const Y: &'static mut Foo = &mut Foo {...}; can't be allowed (unless Foo is ZST) because the const semantics would result in copies of a &mut Foo (only valid for ZSTs).

But we do want to allow const Z: Bar = Bar::new(&mut Foo {...}); with a temporary lifetime for the &mut Foo reference, so the "top-level escaping mutable reference" is relevant here.
So I'm wondering if, for const specifically, we want to adjust the temporary lifetime rules, such that &mut expr expressions aren't special-cased (same as if they were written id(&mut expr)).
That means const ARRAY: &'static mut [T; 0] = &mut []; would still work, but only through promotion (which has the same rules, IIRC, as what we allow today at the-top level, except around drops, which shouldn't be an issue for [T; 0] types, the only ones we ever allowed).

oli-obk · 2018-03-30T10:38:47Z

I removed the borrowing changes

shepmaster · 2018-04-07T19:50:58Z

Ping from triage, @eddyb !

oli-obk · 2018-04-08T20:35:01Z

This is very low priority atm. We could close it until after the next beta cut.

pietroalbini · 2018-04-16T10:31:01Z

Ping from triage @eddyb! This PR needs your review.

pietroalbini · 2018-04-23T11:57:28Z

Ping from triage! Can @eddyb or someone else from @rust-lang/compiler review this?

eddyb · 2018-04-24T17:22:10Z

src/librustc_mir/transform/qualify_consts.rs

@@ -1058,6 +1078,11 @@ This does not pose a problem by itself because they can't be accessed directly."
                // Avoid a generic error for other uses of arguments.
                if self.qualif.intersects(Qualif::FN_ARGUMENT) {
                    let decl = &self.mir.local_decls[index];
+                    if decl.source_info.span.allows_unstable() {


Isn't this condition flipped? EDIT: this appears to be the same for all allows_unstable.

eddyb · 2018-04-24T17:24:48Z

src/librustc_mir/transform/qualify_consts.rs

@@ -355,10 +365,13 @@ impl<'a, 'tcx> Qualifier<'a, 'tcx, 'tcx> {
                TerminatorKind::FalseUnwind { .. } => None,

                TerminatorKind::Return => {
+                    if self.tcx.sess.features_untracked().const_let {
+                        break;


I would instead put an if around all the code below, just to make sure it's explicit.

oli-obk · 2018-04-24T17:30:21Z

Note to self. more tests:

// ok
const X: FakeNeedsDrop = { let x = FakeNeedsDrop; x };
// error
const Y: FakeNeedsDrop = { let mut x = FakeNeedsDrop; x = FakeNeedsDrop; x };
// error
const Z: () = { let mut x = None; x = Some(FakeNeedsDrop); }

shepmaster · 2018-05-06T21:25:28Z

Ping from triage @oli-obk ! Any time to work on this in the near future?

pietroalbini · 2018-05-14T09:54:51Z

Ping from triage @eddyb! This PR needs your review.

oli-obk · 2018-05-14T09:58:00Z

#50603 takes priority over this PR, because it fixes a soundness hole (there will be conflicts, so we're waiting for #50603 to get merged first)

kennytm · 2018-05-19T17:05:13Z

#50603 has been merged.

oli-obk · 2018-05-19T18:04:23Z

amazingly there are no conflicts whatsoever. @eddyb this is ready for another review

eddyb · 2018-05-21T17:49:04Z

src/librustc_mir/transform/qualify_consts.rs

            }
            Place::Local(index) if self.mir.local_kind(index) == LocalKind::ReturnPointer => {
                debug!("store to return place {:?}", index);
-                store(&mut self.return_qualif)
+                store(&mut self.local_qualif[RETURN_PLACE])


Can you combine these two match arms?

eddyb · 2018-05-21T17:53:03Z

r=me with the nit fixed

oli-obk · 2018-05-22T12:04:36Z

@bors r=eddyb

bors · 2018-05-22T12:04:37Z

📌 Commit 2483c81 has been approved by eddyb

bors · 2018-05-22T12:57:13Z

⌛ Testing commit 2483c81 with merge 9f80ea3...

@eddyb

Allow let bindings and destructuring in constants and const fn r? @eddyb cc #48821

bors · 2018-05-22T15:16:16Z

☀️ Test successful - status-appveyor, status-travis
Approved by: eddyb
Pushing 9f80ea3 to master...

pietroalbini · 2018-05-24T16:33:42Z

This PR needs to be backported to beta in order to backport #50909.
Cause: #51037 (comment)

cc @rust-lang/compiler

oli-obk · 2018-05-24T18:40:43Z

i can create a version of #50909 that doesn't depend on this, it's just a debug statement that conflicts

pietroalbini · 2018-05-24T18:46:24Z

@oli-obk that would be awesome!

rust-highfive assigned eddyb Mar 19, 2018

rust-highfive added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Mar 19, 2018

oli-obk force-pushed the const_let branch from e0f18a8 to 38f5918 Compare March 20, 2018 10:19

eddyb reviewed Mar 24, 2018

View reviewed changes

oli-obk force-pushed the const_let branch from 38f5918 to 7d3f922 Compare March 27, 2018 19:16

eddyb reviewed Apr 24, 2018

View reviewed changes

pietroalbini added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Apr 30, 2018

oli-obk mentioned this pull request May 2, 2018

RFC: Keyword unreservations (pure, sizeof, alignof, offsetof) rust-lang/rfcs#2421

Merged

oli-obk force-pushed the const_let branch from 6d7d240 to 05ffe70 Compare May 8, 2018 11:59

This comment has been minimized.

Sign in to view

pietroalbini added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels May 14, 2018

pietroalbini added S-blocked Status: Marked as blocked ❌ on something else such as an RFC or other implementation work. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels May 14, 2018

kennytm added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-blocked Status: Marked as blocked ❌ on something else such as an RFC or other implementation work. labels May 19, 2018

eddyb reviewed May 21, 2018

View reviewed changes

oli-obk added 5 commits May 22, 2018 10:54

Allow let bindings in const fn and constants

9872160

Be more explicit about what's skipped

d7bf358

Report let bindings and statements as unstable

8a5eb68

Remove now unused error code

cfb6816

Fix a typo in a comment

61c58c2

oli-obk mentioned this pull request May 22, 2018

Make AllocId decoding thread-safe #50957

Closed

Deduplicate match arms

2483c81

oli-obk force-pushed the const_let branch from e72d03a to 2483c81 Compare May 22, 2018 11:56

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels May 22, 2018

bors added a commit that referenced this pull request May 22, 2018

Auto merge of #49172 - oli-obk:const_let, r=eddyb

9f80ea3

Allow let bindings and destructuring in constants and const fn r? @eddyb cc #48821

bors merged commit 2483c81 into rust-lang:master May 22, 2018

pietroalbini added the beta-nominated Nominated for backporting to the compiler in the beta channel. label May 24, 2018

pietroalbini mentioned this pull request May 24, 2018

[beta] Process backports #51037

Merged

oli-obk deleted the const_let branch May 24, 2018 18:38

pietroalbini removed the beta-nominated Nominated for backporting to the compiler in the beta channel. label May 24, 2018

oli-obk mentioned this pull request Nov 22, 2018

Fix various aspects around let bindings inside const functions #56160

Merged

Centril mentioned this pull request Dec 30, 2018

Stabilize let bindings and destructuring in constants and const fn #57175

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Allow let bindings and destructuring in constants and const fn #49172

Allow let bindings and destructuring in constants and const fn #49172

oli-obk commented Mar 19, 2018

eddyb Mar 24, 2018

eddyb Mar 24, 2018

eddyb Mar 24, 2018

eddyb Mar 24, 2018

eddyb commented Mar 24, 2018

oli-obk commented Mar 30, 2018

shepmaster commented Apr 7, 2018

oli-obk commented Apr 8, 2018

pietroalbini commented Apr 16, 2018

pietroalbini commented Apr 23, 2018

eddyb Apr 24, 2018 •

edited

Loading

eddyb Apr 24, 2018

oli-obk commented Apr 24, 2018 •

edited

Loading

shepmaster commented May 6, 2018

This comment has been minimized.

pietroalbini commented May 14, 2018

oli-obk commented May 14, 2018

kennytm commented May 19, 2018

oli-obk commented May 19, 2018

eddyb May 21, 2018

eddyb commented May 21, 2018

oli-obk commented May 22, 2018

bors commented May 22, 2018

bors commented May 22, 2018

bors commented May 22, 2018

pietroalbini commented May 24, 2018

oli-obk commented May 24, 2018

pietroalbini commented May 24, 2018

Allow let bindings and destructuring in constants and const fn #49172

Allow let bindings and destructuring in constants and const fn #49172

Conversation

oli-obk commented Mar 19, 2018

eddyb Mar 24, 2018

Choose a reason for hiding this comment

eddyb Mar 24, 2018

Choose a reason for hiding this comment

eddyb Mar 24, 2018

Choose a reason for hiding this comment

eddyb Mar 24, 2018

Choose a reason for hiding this comment

eddyb commented Mar 24, 2018

oli-obk commented Mar 30, 2018

shepmaster commented Apr 7, 2018

oli-obk commented Apr 8, 2018

pietroalbini commented Apr 16, 2018

pietroalbini commented Apr 23, 2018

eddyb Apr 24, 2018 • edited Loading

Choose a reason for hiding this comment

eddyb Apr 24, 2018

Choose a reason for hiding this comment

oli-obk commented Apr 24, 2018 • edited Loading

shepmaster commented May 6, 2018

This comment has been minimized.

pietroalbini commented May 14, 2018

oli-obk commented May 14, 2018

kennytm commented May 19, 2018

oli-obk commented May 19, 2018

eddyb May 21, 2018

Choose a reason for hiding this comment

eddyb commented May 21, 2018

oli-obk commented May 22, 2018

bors commented May 22, 2018

bors commented May 22, 2018

bors commented May 22, 2018

pietroalbini commented May 24, 2018

oli-obk commented May 24, 2018

pietroalbini commented May 24, 2018

eddyb Apr 24, 2018 •

edited

Loading

oli-obk commented Apr 24, 2018 •

edited

Loading