Matthew Jasper [Sun, 11 Aug 2019 10:48:22 +0000 (11:48 +0100)]
Stop emulating cross-crate hygiene with gensyms
Most `Ident`s are serialized as `InternedString`s the exceptions are:
* Reexports
* Attributes
* Idents in macro definitions
Using gensyms helped reexports emulate hygiene. However, the actual item
wouldn't have a gensymmed name so would be usable cross-crate. So
removing this case until we have proper cross-crate hygiene seems
sensible.
Codegen attributes (`inline`, `export_name`) are resolved by their
`Symbol`. This meant that opaque macro-expanded codegen attributes could
cause linker errors. This prevented making built-in derives hygienic.
Rollup merge of #63641 - DieracDelta:master, r=eddyb
add git keyword to submodule comments in config.example.toml
I searched over config.example.toml file looking for a place to disable git submodules from being updated, and missed the two options related to this because they did not include the keyword git. This pr simply adds git to the relevant comments so hopefully others won't also miss that these options exist.
Rollup merge of #63558 - jgalenson:move-remap, r=alexcrichton
Remap paths for proc-macro crates.
The remap-debuginfo config option remaps paths in most crates, but it does not apply to proc-macros, so they are still non-reproducible. This patch fixes that.
I'm not completely sure if this is the best way to do this, but to get reproducible builds we need librustc_macros to be built with --remap-path-prefix. I was previously modifying Cargo to pass that argument to all child crates, so this seems simpler and more correct.
I did not add a test since there do not seem to be any existing tests for RUSTC_DEBUGINFO_MAP.
Rollup merge of #63548 - eddyb:unicode-demangling, r=alexcrichton
Update rustc-demangle to 0.1.16.
Includes https://github.com/alexcrichton/rustc-demangle/pull/29 and https://github.com/alexcrichton/rustc-demangle/pull/30.
You can see the effects of the former in the testcase changes.
Rollup merge of #63149 - petrochenkov:lazypop2, r=eddyb
resolve: Populate external modules in more automatic and lazy way
So, resolve had this function `populate_module_if_necessary` for loading module children from other crates from metadata.
I never really understood when it should've been called and when not.
This PR removes the function and loads the module children automatically on the first access instead.
bors [Fri, 16 Aug 2019 18:35:17 +0000 (18:35 +0000)]
Auto merge of #63640 - Centril:rollup-yeb8o66, r=Centril
Rollup of 10 pull requests
Successful merges:
- #60492 (Add custom nth_back for Chain)
- #61780 (Finalize the error type for `try_reserve`)
- #63495 ( Remove redundant `ty` fields from `mir::Constant` and `hair::pattern::PatternRange`.)
- #63525 (Make sure that all file loading happens via SourceMap)
- #63595 (add sparc64-unknown-openbsd target)
- #63604 (Some update for vxWorks)
- #63613 (Hygienize use of built-in macros in the standard library)
- #63632 (A couple of comment fixes.)
- #63634 (ci: properly set the job name in CPU stats)
- #63636 (ci: move linkcheck from mingw-2 to mingw-1)
Move some code into `build_reduced_graph.rs` to keep `BuildReducedGraphVisitor` it private
Also move the def collector call to `build_reduced_graph.rs`, it belongs there.
Rollup merge of #63636 - pietroalbini:ci-mingw, r=alexcrichton
ci: move linkcheck from mingw-2 to mingw-1
Running UI tests now takes a huge amount of time on mingw builders
(between 40 and 50 minutes), with mingw-1 builders taking even an hour
less to finish than mingw-2. This PR moves linkcheck from mingw-2 to
mingw-1, removing between 10 and 20 minutes of runtime on the -2
builders.
Rollup merge of #63604 - Wind-River:master, r=alexcrichton
Some update for vxWorks
1. support crt-static
2. change armv7_wrs_vxworks to armv7_wrs_vxworks_eabihf.
3. change vx-cxx to wr-c++, vx-ar to wr-ar and vx-run to wr-run.
4. code cleanup
Rollup merge of #63595 - semarie:openbsd-sparc64, r=alexcrichton
add sparc64-unknown-openbsd target
on OpenBSD, some architectures relies on libc++ (from LLVM) and some
others on libestdc++ (particular version of libstdc++ from GCC).
sparc64-unknown-openbsd needs libestdc++ and libgcc (as x86_64 some
years ago). Reintroduce the support of them for openbsd, only for
sparc64 arch. Some others architectures on OpenBSD could use them too.
Rollup merge of #61780 - SimonSapin:container-error, r=Amanieu
Finalize the error type for `try_reserve`
See tracking issue comments from https://github.com/rust-lang/rust/issues/48043#issuecomment-500828346.
It is now:
```rust
/// The error type for `try_reserve` methods.
#[derive(Clone, PartialEq, Eq, Debug)]
#[unstable(feature = "try_reserve", reason = "new API", issue="48043")]
pub enum TryReserveError {
/// Error due to the computed capacity exceeding the collection's maximum
/// (usually `isize::MAX` bytes).
CapacityOverflow,
/// The memory allocator returned an error
AllocError {
/// The layout of allocation request that failed
layout: Layout,
#[doc(hidden)]
#[unstable(feature = "container_error_extra", issue = "0", reason = "\
Enable exposing the allocator’s custom error value \
if an associated type is added in the future: \
https://github.com/rust-lang/wg-allocators/issues/23")]
non_exhaustive: (),
},
}
* A `Layout` is included. Firefox wants to log the size of failed allocations. If this were not part of the return value of e.g. `HashMap::try_reserve`, users would only be able to estimate based on `HashMap::capacity` and assumptions about the allocation strategy of `HashMap`.
* There’s a dummy field that can stay unstable when `try_reserve` and the rest of this enum are stabilized. This forces non-exhaustive matching ~(https://github.com/rust-lang/rust/issues/44109 is not implemented yet for variants)~ and allows adding another field in the future if we want to expose custom error values from the allocator. See https://github.com/rust-lang/wg-allocators/issues/23.
- If the `Alloc` trait is stabilized without an associated error type and with a zero-size `AllocErr` type, we can simply remove this dummy field.
- If an associated type is added, we can add a default type parameter to `ContainerError` and a generic field to the `AllocError` variant.
* ~Moved from the `collections` module to the `alloc` module, and replaced `Collection` in the enum name with `Container`. The wold collection implies a multiplicity of items which is not relevant to this type. For example we may want to use this error type in a future `Box::try_new` method.~
- Renamed to `TryReserveError`, after the methods that involve this type: https://github.com/rust-lang/rust/pull/61780#issuecomment-501392487
* Replaced `Err` with `Error` in the enum and variant names. There is more precedent for this in https://doc.rust-lang.org/std/error/trait.Error.html#implementors, `AllocErr` and `LayoutErr` are the odd ones.
* ~Dropped `Alloc` in the enum name. `ContainerAllocError` with a mouthful, and being in the `alloc` module already provides the same indication.~
Simon Sapin [Wed, 12 Jun 2019 16:47:58 +0000 (18:47 +0200)]
Add the Layout of the failed allocation to TryReserveError::AllocError
… and add a separately-unstable field to force non-exhaustive matching
(`#[non_exhaustive]` is no implemented yet on enum variants)
so that we have the option to later expose the allocator’s error value.
CC https://github.com/rust-lang/wg-allocators/issues/23
Pietro Albini [Fri, 16 Aug 2019 15:02:01 +0000 (17:02 +0200)]
ci: move linkcheck from mingw-2 to mingw-1
Running UI tests now takes a huge amount of time on mingw builders
(between 40 and 50 minutes), with mingw-1 builders taking even an hour
less to finish than mingw-2. This PR moves linkcheck from mingw-2 to
mingw-1, removing between 10 and 20 minutes of runtime on the -2
builders.
bors [Fri, 16 Aug 2019 14:41:13 +0000 (14:41 +0000)]
Auto merge of #63470 - Mark-Simulacrum:rustc-depdep, r=alexcrichton
Utilize -Zbinary-dep-depinfo in rustbuild
We no longer utilize stamp-file mtimes at all inside rustbuild, and a future PR may be able to entirely eliminate them by eagerly copying to the appropriate sysroot. The only mtime-based dependency tracking left is for documentation because we lie to Cargo about the rustdoc binary, so Cargo does not track changes to the real binary, and codegen-backends because binary-dep-depinfo does not emit that information into the depfiles.
Both of these are fixable in the longer term but this existing patch gives us the following benefits:
* We no longer delete Cargo target directories manually within a stage. Cross-stage, changes to codegen backends will still clear out target directories. This means that incremental state persists across individual steps (e.g., rebuilding libstd does not clear out librustc incremental state). Fixes #54712.
* Dependency tracking across steps within a given stage is now fully precise. We will not clear out all codegen backend dependencies due to changes in librustc_driver, for example, only deleting the final librustc_codegen_llvm crate. Fixes #54008, fixes #50481.
* We properly track codegen backends as a dependency (equivalent to rustc) across changes. Fixes #53284, and fixes #52719.
* Cross-stage dependency tracking of crates is also much more accurate and reliable. Most likely fixes #49979 (but no reproduction steps in that issue). Fixes #59105.
bors [Fri, 16 Aug 2019 06:53:13 +0000 (06:53 +0000)]
Auto merge of #63627 - Centril:rollup-v8i7x5i, r=Centril
Rollup of 7 pull requests
Successful merges:
- #62593 (Group all ABI tests.)
- #63173 (Use libunwind from llvm-project submodule for musl targets)
- #63535 (Continue refactoring resolve and hygiene)
- #63539 (Suggest Rust 2018 on `<expr>.await` with no such field)
- #63584 (libcore: more cleanups using `#![feature(associated_type_bounds)]`)
- #63612 (Do not suggest `try_into` for base types inside of macro expansions)
- #63615 (Fix typo in DoubleEndedIterator::nth_back doc)
Rollup merge of #63539 - Centril:2015.await, r=oli-obk
Suggest Rust 2018 on `<expr>.await` with no such field
When type checking a field projection (`fn check_field`) to `<expr>.await` where `<expr>: τ` and `τ` is not a primitive type, suggest switching to Rust 2018. E.g.
```
error[E0609]: no field `await` on type `std::pin::Pin<&mut dyn std::future::Future<Output = ()>>`
--> $DIR/suggest-switching-edition-on-await.rs:31:7
|
LL | x.await;
| ^^^^^ unknown field
|
= note: to `.await` a `Future`, switch to Rust 2018
= help: set `edition = "2018"` in `Cargo.toml`
= note: for more on editions, read https://doc.rust-lang.org/edition-guide
```
This PR also performs some preparatory cleanups in `fn check_field`; the last 2 commits are where the suggestion is introduced and tested respectively.
Rollup merge of #63535 - petrochenkov:expndata, r=matthewjasper
Continue refactoring resolve and hygiene
The general goal is addressing FIXMEs from the previous PRs.
Merging similar data structures (+ prerequisites for such merging), accounting for the fact that all `ExpnId`s have associated data in `HygieneData` now (less `Option`s).
Also, some renaming.
This should be the last renaming session in this area, I think.
resolve: Do not "normalize away" trait/enum modules prematurely
The previous approach was brittle - what would happen if `ParentScope` wasn't created by `invoc_parent_scope`?
That's exactly the case for various uses of `ParentScope` in diagnostics and in built-in attribute validation.
resolve: Move macro resolution traces from `Module`s to `Resolver`
Traces already contain module info without that.
It's easy to forget to call `finalize_*` on a module.
In particular, macros enum and trait modules weren't finalized.
By happy accident macros weren't placed into those modules until now.
Sébastien Marie [Thu, 15 Aug 2019 13:34:23 +0000 (15:34 +0200)]
add sparc64-unknown-openbsd target
on OpenBSD, some architectures relies on libc++ (from LLVM) and some
others on libestdc++ (particular version of libstdc++ from GCC).
sparc64-unknown-openbsd needs libestdc++ and libgcc (as x86_64 some
years ago). Reintroduce the support of them for openbsd, only for
sparc64 arch. Some others architectures on OpenBSD could use them too.
Rollup merge of #63467 - terhechte:support-ios-catalyst-macabi-target-triple, r=estebank
Add Catalyst (iOS apps running on macOS) target
This is a first attempt of adding support for the new [Apple Catalyst](https://developer.apple.com/ipad-apps-for-mac/) target (i.e. running iOS apps on macOS). Currently, `rustc` supports the iOS and iOS simulator targets for iOS:
- iOS: ARM cpu, iOS SDK, linked agains the iOS ABI
- Simulator: X86_64 cpu, iOS SDK, linked against the iOS ABI
Apple Catalyst will add an additional target:
- Macabi: X86_64 CPU, iOS SDK, linked again the macOS ABI.
Note, it the actual SDK is the also the macOS 10.15 SDK, but the symbols are the iOS SDK symbols as they were added to macOS with 10.15.
I've collected additional information via links in the open question sections below. This is way out of my comfort zone so please excuse whatever errors I may have made.
# Open Questions:
## Clang Version
It seems to me that `macabi` has not been merged into `clang` yet, I don't know whether that is a requirement rustc to compile, or if it is sufficient if the Clang that is used on a developers system is the correct one supporting macabi (that comes with current Xcode)
## Hardcoded iOS version
`swift-llvm` actually used [x86_64-apple-ios13.0-macabi](https://github.com/apple/swift-llvm/commit/3f1fd4f46a8e82cee6fffb284ee46daf8ddf66ae) as the target triple which hard-codes the current iOS version. A post on stackoverflow [points out that `MIN_IOS_VERSION` and `MIN_OSX_VERSION` should be used when compiling C code for clang (`-target x86_64-apple-ios${MIN_IOS_VERSION}-macabi`)](https://stackoverflow.com/questions/56487645/how-to-compile-a-3rd-party-library-to-be-used-with-uikit-for-mac-catalyst). However, I wasn't entirely sure how to do that in this PR. Pointers welcome.
## Data Layout
I'm probably using the wrong data-layout. I don't know whether it should be the macOS version or the iOS version. This is probably easier to answer for somebody who understands these things much better than me. I just copied the iOS Simulator X86_64 version as it seems to be (based on what I understand) that Catalyst is just the simulator target build against a different SDK.
# Current State
1. I got it to compile
2. I could successfully compile a `macabi` `libcore` via `cargo build --target x86_64-apple-ios-macabi`
I'm not sure what needs to be done next. Supposedly I need to compile everything into a toolchain somehow that I can then test via `rustup` to make sure that a binary compiled against the toolchain also works with Catalyst. [I read this article, but I'm still lost](https://www.reddit.com/r/rust/comments/5ag60z/how_do_i_bootstrap_rust_to_crosscompile_for_a_new/d9gicr2/) and would love pointers what to do next here.
# Additional Information
- [Commit adding Catalyst support to the Swift Clang Fork](https://github.com/CocoaPods/CocoaPods/issues/8877)
- [Compiling C to Catalyst Discussion](https://github.com/CocoaPods/CocoaPods/issues/8877)
- [CocoaPods Discussion on Adding Catalyst support](https://github.com/CocoaPods/CocoaPods/issues/8877)
With https://github.com/rust-lang/miri/pull/872, Miri only retags "bare" references, not those nested in compound types. This adjust `Retag` statement generation to don't emit retags if they are definitely not a bare reference.
I also expanded the mir-opt test to cover the `Retag` in the drop shim, which had previously not been tested.
Rollup merge of #63155 - mfkl:uwp-msvc, r=alexcrichton
Add UWP MSVC targets
Hi,
- The README URI change is the correct one for VS2019 community edition, which I suspect most people would use. Doesn't _need_ to be merged though.
- This https://github.com/rust-lang/rust/commit/5e6619edd1a3b5c3f85438166d4d32af49f800fd fixes the UWP build (msvc or not, doesn't matter). I suspect it broke with recent changes unnoticed because no CI.
- Store lib location is found through the VCToolsInstallDir env variable. The end of the path is currently for the VS2019 store lib locations only.
- I could not test the aarch64_uwp_windows_msvc target because the rust build script does not currently support arm64 msvc AFAIU.
Kevin Per [Thu, 11 Jul 2019 17:40:09 +0000 (19:40 +0200)]
Fixing broken tests #62401
The grouping led to a lot of `mv`. Therefore, some relative paths were
wrong. In this commit the dependent files were also moved so that the paths
work again.