Skip to content

enforce WF conditions after generalizing - #41716

Merged
bors merged 3 commits into
rust-lang:masterfrom
nikomatsakis:issue-41677
May 12, 2017
Merged

enforce WF conditions after generalizing#41716
bors merged 3 commits into
rust-lang:masterfrom
nikomatsakis:issue-41677

Conversation

@nikomatsakis

@nikomatsakisnikomatsakis commented May 2, 2017

Copy link
Copy Markdown
Contributor

Add a WF(T') obligation after generalizing T to T', if T' contains an unconstrained type variable in a bivariant context.

Fixes#41677.

Beta nominating -- regression.

r? @arielb1

@nikomatsakisnikomatsakis added beta-nominated Nominated for backporting to the compiler in the beta channel. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels May 2, 2017
@aidanhsaidanhs added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label May 3, 2017
@bors

bors commented May 6, 2017

Copy link
Copy Markdown
Collaborator

☔ The latest upstream changes (presumably #41773) made this pull request unmergeable. Please resolve the merge conflicts.

@Mark-SimulacrumMark-Simulacrum 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 May 7, 2017
@arielb1

arielb1 commented May 7, 2017

Copy link
Copy Markdown
Contributor

Got back from holidays. Had a fun week.

This looks like a rather brutal fix. As for concrete worries, I was worried that we might create wf obligations "inside" an higher-ranked binder, and that would cause trouble, but I don't think we ever do subtyping in a higher-ranked context right now?

@nikomatsakis

nikomatsakis commented May 9, 2017

Copy link
Copy Markdown
ContributorAuthor

This looks like a rather brutal fix.

Brutal as in "brute force"?

As for concrete worries, I was worried that we might create wf obligations "inside" an higher-ranked binder, and that would cause trouble, but I don't think we ever do subtyping in a higher-ranked context right now?

I don't think that would be a problem. It's not that we don't do subtyping in a higher-ranked context. Rather, it's that we protect against skolemized regions "escaping" into instantiated type variables. If we did instantiate some type variable with something that includes skolemized regions --- or with region variables that wind up related to skolemized variables --- that would be detected during the leak-check. Since we only ever create the WF relations with the value from an instantiated variable, this implies that these relations cannot reference skolemized regions.

(As an aside, I am, though, actively planning how to overhaul that system right now to make it more robust, roughly in the manner that we talked about.)

@arielb1

arielb1 commented May 9, 2017

Copy link
Copy Markdown
Contributor

This is basically a hack around bivariance not being implemented properly, but let's go for it (r+ modulo conflict fix).

This was referenced May 9, 2017
Closed
Closed
@nikomatsakis
replace the type generalizer with one based on variance
a950c37
@nikomatsakis
add a WF obligation if a type variable appears in bivariant position
a4151ff
@nikomatsakis
correct various error messages that changed
2490ee5
The new messages seem universally better. I think these result because
we recognize that we are in an invariant context more often.
@nikomatsakis

nikomatsakis commented May 11, 2017

Copy link
Copy Markdown
ContributorAuthor

@bors r=arielb1

@bors

bors commented May 11, 2017

Copy link
Copy Markdown
Collaborator

📌 Commit 2490ee5 has been approved by arielb1

Mark-Simulacrum added a commit to Mark-Simulacrum/rust that referenced this pull request May 12, 2017
@Mark-Simulacrum
Rollup merge ofrust-lang#41716- nikomatsakis:issue-41677, r=arielb1
5b5c045
enforce WF conditions after generalizing
Add a `WF(T')` obligation after generalizing `T` to `T'`, if `T'` contains an unconstrained type variable in a bivariant context.
Fixesrust-lang#41677.
Beta nominating -- regression.
r? @arielb1
@Mark-SimulacrumMark-Simulacrum mentioned this pull request May 12, 2017
Closed
Mark-Simulacrum added a commit to Mark-Simulacrum/rust that referenced this pull request May 12, 2017
@Mark-Simulacrum
Rollup merge ofrust-lang#41716- nikomatsakis:issue-41677, r=arielb1
c8b1559
enforce WF conditions after generalizing
Add a `WF(T')` obligation after generalizing `T` to `T'`, if `T'` contains an unconstrained type variable in a bivariant context.
Fixesrust-lang#41677.
Beta nominating -- regression.
r? @arielb1
@Mark-SimulacrumMark-Simulacrum mentioned this pull request May 12, 2017
Closed
@bors

bors commented May 12, 2017

Copy link
Copy Markdown
Collaborator

⌛ Testing commit 2490ee5 with merge 141e8a6...

bors added a commit that referenced this pull request May 12, 2017
@bors
Auto merge of#41716- nikomatsakis:issue-41677, r=arielb1
141e8a6
enforce WF conditions after generalizing
Add a `WF(T')` obligation after generalizing `T` to `T'`, if `T'` contains an unconstrained type variable in a bivariant context.
Fixes#41677.
Beta nominating -- regression.
r? @arielb1
@bors

bors commented May 12, 2017

Copy link
Copy Markdown
Collaborator

☀️ Test successful - status-appveyor, status-travis
Approved by: arielb1
Pushing 141e8a6 to master...

@bors
bors merged commit 2490ee5 into rust-lang:masterMay 12, 2017
@brson

brson commented May 23, 2017

Copy link
Copy Markdown
Contributor

Doesn't cherry-pick cleanly cc @nikomatsakis

@nikomatsakisnikomatsakis mentioned this pull request May 23, 2017
Merged
@nikomatsakisnikomatsakis added the beta-accepted Accepted for backporting to the compiler in the beta channel. label May 23, 2017
@nikomatsakis

nikomatsakis commented May 23, 2017

Copy link
Copy Markdown
ContributorAuthor

Accepting for backport. cc @rust-lang/compiler

bors added a commit that referenced this pull request May 24, 2017
@bors
Auto merge of#42182- nikomatsakis:beta, r=brson
2df18ed
Beta backports
Backports of:
- #41913 (or some of it)
- #41937
- #41716
- #41563
@nikomatsakisnikomatsakis removed the beta-nominated Nominated for backporting to the compiler in the beta channel. label May 25, 2017
This was referenced Jun 8, 2017
Closed
Closed
Sign up for freeto join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

@arielb1arielb1

Labels

beta-acceptedAccepted for backporting to the compiler in the beta channel.S-waiting-on-authorStatus: This is awaiting some action (such as code changes or more information) from the author.T-compilerRelevant to the compiler team, which will review and decide on the PR/issue.

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

6 participants

@nikomatsakis@bors@arielb1@brson@aidanhs@Mark-Simulacrum