New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Mb/36812 ich function interfaces #36974

Merged

bors merged 6 commits into rust-lang:master from MathieuBordere:mb/36812_ICHFunctionInterfaces

Oct 6, 2016

MathieuBordere commented Oct 5, 2016

r? @michaelwoerister

This PR contains fixes for #36812 and #36914

Mathieu Borderé added 3 commits

October 5, 2016 06:46


          ICH - Add test cases for function interfaces

9081c00


          ICH - Include omitted elements in inc. comp. hash rust-lang#36914

e051eb3


          Adjustments due to naming changes in Ty_ and PatKind structs

14fe7ce

rust-highfive assigned michaelwoerister

Collaborator

rust-highfive commented Oct 5, 2016

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @michaelwoerister (or someone else) soon.

If any changes to this PR are deemed necessary, please add them as extra commits. This ensures that the reviewer can see what has changed since they last reviewed the code. Due to the way GitHub handles out-of-date commits, this should also make it reasonably obvious what issues have or haven't been addressed. Large or tricky changes may require several passes of review and changes.

Please see the contribution instructions for more information.

Member

michaelwoerister commented Oct 5, 2016

Thank you so much! I'll take a look at this ASAP.

michaelwoerister reviewed

View reviewed changes

Member

michaelwoerister left a comment

This is excellent work! I like how you included SawTyComponent even though we didn't explicitly talk about it. I think this has already caught quite a few bugs before we had to find out about them the hard way.

I've left a few comments about things that need to change. But this almost ready to merge.

src/test/incremental/hashes/function_interfaces.rs

+              // Change Parameter Order ------------------------------------------------------
+              #[cfg(cfail1)]
+              fn order_of_parameters(p1: i32, p2: i32) {}

Member

michaelwoerister Oct 5, 2016

Could you give p2 a different type than p1?

src/test/incremental/hashes/function_interfaces.rs

+              // Return Impl Trait -----------------------------------------------------------
+              #[cfg(cfail1)]
+              fn return_impl_trait() -> i32 {

Member

michaelwoerister Oct 5, 2016

I'm wondering if it would make more sense to have it change from impl Clone to impl Copy instead? It's probably best to add a separate test case for that.

src/test/incremental/hashes/function_interfaces.rs

+              // Second Lifetime Bound -------------------------------------------------------
+              #[cfg(cfail1)]
+              fn second_lifetime_bound<'a, T: 'a>() {}

Member

michaelwoerister Oct 5, 2016

Can you add 'b also to the cfail1 version so that we are really just testing for the addition of a bound (as opposed to a bound and a new lifetime parameter)?

src/test/incremental/hashes/function_interfaces.rs

+              // Inline Never ----------------------------------------------------------------
+              #[cfg(cfail1)]
+              fn inline_never() {}

Member

michaelwoerister Oct 5, 2016

Can you add an #[inline(always)] attribute to the cfail1 version? I'd like this to test whether changing the kind of inline hint is detected.

src/test/incremental/hashes/function_interfaces.rs

+              #[rustc_clean(label="Hir", cfg="cfail3")]
+              #[rustc_metadata_dirty(cfg="cfail2")]
+              #[rustc_metadata_clean(cfg="cfail3")]
+              #[linkage]

Member

michaelwoerister Oct 5, 2016

I'm surprised this compiles :)
Can you change that to #[linkage="weak_odr"] so we actually specify an explicit linkage?

src/test/incremental/hashes/function_interfaces.rs

+              // Lifetime Bound --------------------------------------------------------------
+              #[cfg(cfail1)]
+              fn lifetime_bound<T>() {}

Member

michaelwoerister Oct 5, 2016

Can you add 'a to the cfail1 version too? (See a similar case with the second lifetime bound)

src/librustc_incremental/calculate_svh/svh_visitor.rs

+                  SawTyArray,
+                  SawTyPtr(Mutability),
+                  SawTyRptr(Mutability),
+                  SawTyBareFn,

Member

michaelwoerister Oct 5, 2016

The BareFnTy contained in TyBareFn also contains an unsafety and an abi field that should be added here.

Author

MathieuBordere commented Oct 5, 2016

Thanks, will incorporate your remarks.

Mathieu Borderé added 2 commits

October 5, 2016 23:17


          ICH: update saw_ty for TyBareFn; Update tests for functioninterfaces

4b5a9a3


          Merge branch 'master' into mb/36812_ICHFunctionInterfaces

6f64296

michaelwoerister reviewed

View reviewed changes

src/librustc_incremental/calculate_svh/svh_visitor.rs

@@ @@ -366,7 +366,10 @@ fn saw_ty(node: &Ty_) -> SawTyComponent { @@
                     TyArray(..) => SawTyArray,
                     TyPtr(ref mty) => SawTyPtr(mty.mutbl),
                     TyRptr(_, ref mty) => SawTyRptr(mty.mutbl),
-                    TyBareFn(..) => SawTyBareFn,
+                    TyBareFn(ref barefnty) => {
+                        let ref fnty = *barefnty;

Member

michaelwoerister Oct 5, 2016

This seems redundant. Can't you just write SawTyBareFn(barefnty.unsafety, barefnty.abi) in the line below? Should do the same thing.

Author

MathieuBordere Oct 6, 2016

You're right. I'm very new to Rust, still a lot to learn :)


          ICH: Remove obsolete binding in saw_ty

0e40dbb

Member

michaelwoerister commented Oct 6, 2016

Thanks, I think all my comments have been addressed.

Member

michaelwoerister commented Oct 6, 2016

Contributor

bors commented Oct 6, 2016

📌 Commit 0e40dbb has been approved by michaelwoerister

sophiajt mentioned this pull request

Rollup of 15 pull requests #37002

Merged

sophiajt pushed a commit to sophiajt/rust that referenced this pull request


          Rollup merge of rust-lang#36974 - MathieuBordere:mb/36812_ICHFunction…

2c8c0de

…Interfaces, r=michaelwoerister

Mb/36812 ich function interfaces

r? @michaelwoerister

This PR contains fixes for rust-lang#36812 and rust-lang#36914

bors added a commit that referenced this pull request


          Auto merge of #37002 - jonathandturner:rollup, r=jonathandturner

ad19c32

Rollup of 15 pull requests

- Successful merges: #36726, #36832, #36909, #36930, #36932, #36957, #36959, #36960, #36962, #36965, #36966, #36967, #36972, #36974, #36977
- Failed merges:

bors merged commit 0e40dbb into rust-lang:master

Member

michaelwoerister commented Oct 7, 2016

🎉

Author

MathieuBordere commented Oct 7, 2016

Feels nice to be able to contribute, up to the next! Thanks!

MathieuBordere deleted the mb/36812_ICHFunctionInterfaces branch

October 7, 2016 19:22

michaelwoerister mentioned this pull request

ICH: Enable some cases in trait definition hashing. #37072

Merged

GuillaumeGomez added a commit to GuillaumeGomez/rust that referenced this pull request


          Rollup merge of rust-lang#37072 - michaelwoerister:enable-trait-ich-t…

bfbd0a5

…ests, r=nikomatsakis

ICH: Enable some cases in trait definition hashing.

Enable some test cases originally written by @eulerdisk. The tests can be enabled now because @MathieuBordere has fixed the underlying problem in rust-lang#36974.

r? @nikomatsakis

This was referenced Oct 18, 2016

ICH: Function unsafety does not affect incr. comp. hash #36914

Closed

incr.comp.: Create Test Case for Incr. Comp. Hash for function interfaces #36812

Closed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet