bors [Sun, 6 Oct 2019 16:32:46 +0000 (16:32 +0000)]
Auto merge of #64564 - jonas-schievink:cowardly-default, r=nikomatsakis
Deny specializing items not in the parent impl
Part of https://github.com/rust-lang/rust/issues/29661 (https://github.com/rust-lang/rfcs/pull/2532). At least sort of?
This was discussed in https://github.com/rust-lang/rust/pull/61812#discussion_r300504114 and is needed for that PR to make progress (fixing an unsoundness).
One annoyance with doing this is that it sometimes requires users to copy-paste a provided trait method into an impl just to mark it `default` (ie. there is no syntax to forward this impl method to the provided trait method).
Previously, we were using an `FxHashMap` to collect module re-exports.
However, re-exports end up getting serialized into crate metadata, which
means that metadata generation was non-deterministic. This resulted in
spurious error messages changes (e.g. PR #64906) due to pretty-printing
implicitly depending on the order of re-exports when computing the
proper path to show to the user.
See #65042 for a long-term strategy to detect this kind of issue
bors [Sun, 6 Oct 2019 04:59:16 +0000 (04:59 +0000)]
Auto merge of #65152 - tmandry:rollup-btn4a01, r=tmandry
Rollup of 18 pull requests
This contains changes from all the successful runs that bors marked as timed out, plus a revert of #63649 which appears to be the immediate cause of the timeouts.
Successful merges:
- #64708 (Stabilize `Option::as_deref` and `Option::as_deref_mut`)
- #64728 (Stabilize UdpSocket::peer_addr)
- #64765 (std: Reduce checks for `feature = "backtrace"`)
- #64909 (When encountering chained operators use heuristics to recover from bad turbofish)
- #65011 (Do not ICE when dereferencing non-Copy raw pointer)
- #65064 (permit asyncawait-ondeck to be added by anyone)
- #65066 ([const-prop] Fix ICE when trying to eval polymorphic promoted MIR)
- #65100 (Replace GeneratorSubsts with SubstsRef)
- #65105 (Split out some passes from librustc)
- #65106 (Allow unused attributes to avoid incremental bug)
- #65113 (Fix lonely backtick)
- #65116 (Remove unneeded visit_statement definition)
- #65118 (Update the documented default of -Z mutable-noalias)
- #65123 (Account for macro invocation in `let mut $pat` diagnostic.)
- #65124 (Replace some instances of `as *[const | mut] _` with `.cast()`)
- #65126 (Fix typo on `now()` comments)
- #65130 (lint: extern non-exhaustive types are improper)
- #65151 (Revert #63649 - "Upgrade Emscripten targets to use upstream LLVM backend")
Tyler Mandry [Sun, 6 Oct 2019 04:55:13 +0000 (21:55 -0700)]
Rollup merge of #65151 - tmandry:revert-emscripten-upgrade, r=tmandry
Revert #63649 - "Upgrade Emscripten targets to use upstream LLVM backend"
This change caused the runtime of the linux-asmjs builder to nearly double from 2+ hours to about 4 hours, which happens to be the bors timeout. (It made it in barely under 4 hours when it was merged.) This is causing timeouts on all new changes.
Tyler Mandry [Sun, 6 Oct 2019 04:55:11 +0000 (21:55 -0700)]
Rollup merge of #65130 - davidtwco:rfc-2008-improper-ctypes, r=petrochenkov
lint: extern non-exhaustive types are improper
This PR makes the `improper_ctype` lint trigger for non-exhaustive types when those types aren't defined in the current crate, as per [this comment](https://github.com/rust-lang/rust/issues/44109#issuecomment-537583344).
Tyler Mandry [Sun, 6 Oct 2019 04:55:00 +0000 (21:55 -0700)]
Rollup merge of #65105 - Mark-Simulacrum:split-librustc, r=nikomatsakis
Split out some passes from librustc
This is just moving them out to librustc_passes -- I've not measured compile time or run time. I don't expect any significant impact, but this seems prudent regardless.
Tyler Mandry [Sun, 6 Oct 2019 04:54:51 +0000 (21:54 -0700)]
Rollup merge of #64765 - alexcrichton:less-check-backtrace, r=sfackler
std: Reduce checks for `feature = "backtrace"`
This is a stylistic change to libstd to reduce the number of checks of
`feature = "backtrace"` now that we unconditionally depend on the
`backtrace` crate and rely on it having an empty implementation.
otherwise.
Aaron Hill [Thu, 3 Oct 2019 04:31:21 +0000 (00:31 -0400)]
Make re-export collection deterministic
Previously, we were using an `FxHashMap` to collect module re-exports.
However, re-exports end up getting serialized into crate metadata, which
means that metadata generation was non-deterministic. This resulted in
spurious error messages changes (e.g. PR #64906) due to pretty-printing
implicitly depending on the order of re-exports when computing the
proper path to show to the user.
See #65042 for a long-term strategy to detect this kind of issue
bors [Fri, 4 Oct 2019 20:33:02 +0000 (20:33 +0000)]
Auto merge of #63649 - tlively:emscripten-upstream-upgrade, r=alexcrichton
Upgrade Emscripten targets to use upstream LLVM backend
- Refactors the Emscripten target spec to share code with other wasm
targets.
- Replaces the incorrect wasm32 C call ABI with the old asmjs
version, which is correct for both wasm32 and JS.
- Updates the varargs ABI used by Emscripten and deletes the old one.
- Removes the obsolete wasm32-experimental-emscripten target.
- Temporarily makes Emscripten targets use panic=abort by default
because supporting unwinding will require an LLVM patch.
bors [Fri, 4 Oct 2019 12:17:13 +0000 (12:17 +0000)]
Auto merge of #64994 - GuillaumeGomez:fix-rustdoc-display-js-disabled, r=Mark-Simulacrum
Fix rustdoc display with js disabled
Fixes #64988.
Currently, all sections are collapsed when the page is loading, and then is displayed once done. However, if js is disabled, they never get expanded. Therefore, they need to be shown by default.
Thomas Lively [Sat, 17 Aug 2019 05:08:01 +0000 (22:08 -0700)]
Upgrade Emscripten targets to use upstream LLVM backend
- Refactors the Emscripten target spec to share code with other wasm
targets.
- Replaces the incorrect wasm32 C call ABI with the old asmjs
version, which is correct for both wasm32 and JS.
- Updates the varargs ABI used by Emscripten and deletes the old one.
- Removes the obsolete wasm32-experimental-emscripten target.
- Temporarily makes Emscripten targets use panic=abort by default
because supporting unwinding will require an LLVM patch.
Introduce a special case in `IntRange::from_const`.
The `if let Some(val) = value.try_eval_bits(...)` branch in `from_const()` is
very hot for the `unicode_normalization` benchmark.
This commit introduces a special-case alternative for scalars that avoids
`try_eval_bits()` and all the functions it calls (`Const::eval()`,
`ConstValue::try_to_bits()`, `ConstValue::try_to_scalar()`, and
`Scalar::to_bits()`), instead extracting the result immediately.
The type and value checking done by `Scalar::to_bits()` is replicated by moving
it into a new function `Scalar::check_raw()` and using that new function in the
special case.
PR #64673 introduced some special-case handling of scalar types in
`Const::try_eval_bits()`. This handling is now moved out of that function into
the new `IntRange::integral_size_and_signed_bias` function.
This commit reduces the instruction count for
`unicode_normalization-check-clean` by about 10%.
Rollup merge of #65026 - petrochenkov:ice1, r=eddyb
metadata: Some crate loading cleanup
So, my goal was to fix caching of loaded crates which is broken and causes ICEs like #56935 or #64450.
While investigating I found that the code is pretty messy and likes to confuse various things that look similar but are actually different.
This PR does some initial cleanup in that area, I hope to get to the caching itself a bit later.
Rollup merge of #64817 - csmoe:closure, r=nikomatsakis
Replace ClosureSubsts with SubstsRef
Addresses https://github.com/rust-lang/rust/issues/42340 part 3
https://github.com/rust-lang/rust/pull/59312 might benefit from this clean up.
r? @nikomatsakis
Tyler Mandry [Thu, 3 Oct 2019 23:25:39 +0000 (16:25 -0700)]
Rollup merge of #64690 - petrochenkov:mixed, r=dtolnay
proc_macro API: Expose `macro_rules` hygiene
Proc macros do not have direct access to our oldest and most stable hygiene kind - `macro_rules` hygiene.
To emulate it macro authors have to go through two steps - first generate a temporary `macro_rules` item (using a derive, at least until https://github.com/rust-lang/rust/pull/64035 is merged), then generate a macro call to that item. Popular crates like [proc_macro_hack](https://crates.io/crates/proc-macro-hack) use this trick to generate hygienic identifiers from proc macros.
I'd say that these workarounds with nested macro definitions have more chances to hit some corner cases in our hygiene system, in which we don't have full confidence.
So, let's provide a direct access to `macro_rules` hygiene instead.
This PR does that by adding a new method `Span::mixed_site` (bikeshedding is welcome) in addition to existing `Span::call_site` (stable) and `Span::def_site` (unstable).
Identifiers with this span resolve at def-site in for local variables, labels and `$crate`, and resolve at call-site for everything else, i.e. exactly like identifiers produced by `macro_rules`.
This API addition opens the way to stabilizing proc macros in expression positions (https://github.com/rust-lang/rust/issues/54727), for which use of call-site hygiene or workarounds with temporary items would be quite unfortunate.
(`macro_rules` expanded in expression position, on the other hand, are stable since 1.0 and widely used.)
Tyler Mandry [Thu, 3 Oct 2019 23:25:37 +0000 (16:25 -0700)]
Rollup merge of #64675 - Centril:deprecate-plugin, r=oli-obk
Deprecate `#![plugin]` & `#[plugin_registrar]`
This PR deprecates `#![plugin]` and `#[plugin_registrar]`.
~A removal deadline is set: 1.44.0. This will be in 9 months from now and should give everyone who is still relying on the feature ample time to rid themselves of this dependency.~
It was only used for retreiving edition, which was a bug.
In case of dual proc macros the edition should be taken from the target crate version, like any other metadata.
projects / rust.git / log