Commit Graph

73892 Commits

Author SHA1 Message Date
Alex Crichton
8dd36af9cb Rollup merge of #47626 - eddyb:one-less-unwrap, r=nagisa
rustc_trans: remove an unwrap by replacing a bool with Result.

Prompted by @shepmaster.
r? @nagisa
2018-01-25 13:49:47 -08:00
Alex Crichton
4856f07bb5 Rollup merge of #47618 - mrhota:dw_at_noreturn, r=michaelwoerister
Teach rustc about DW_AT_noreturn and a few more DIFlags

We achieve two small things with this PR:
1. We provide definitions for a few additional llvm debuginfo flags
1. We _use_ one of these new flags, `FlagNoReturn`, and add it to debuginfo for functions with the never return type (`!`).
2018-01-25 13:49:46 -08:00
Marco A L Barbosa
c199cb5370 Make 3 run-pass tests works on android (aarch64 and x86) 2018-01-25 18:55:11 -02:00
Alex Crichton
502de01ff4 rustc: SIMD types use pointers in Rust's ABI
This commit changes the ABI of SIMD types in the "Rust" ABI to unconditionally
be passed via pointers instead of being passed as immediates. This should fix a
longstanding issue, #44367, where SIMD-using programs ended up showing very odd
behavior at runtime because the ABI between functions was mismatched.

As a bit of a recap, this is sort of an LLVM bug and sort of an LLVM feature
(today's behavior). LLVM will generate code for a function solely looking at the
function it's generating, including calls to other functions. Let's then say
you've got something that looks like:

```llvm
define void @foo() { ; no target features enabled
  call void @bar(<i64 x 4> zeroinitializer)
  ret void
}

define void @bar(<i64 x 4>) #0 { ; enables the AVX feature
  ...
}
```

LLVM will codegen the call to `bar` *without* using AVX registers becauase `foo`
doesn't have access to these registers. Instead it's generated with emulation
that uses two 128-bit registers. The `bar` function, on the other hand, will
expect its argument in an AVX register (as it has AVX enabled). This means we've
got a codegen problem!

Comments on #44367 have some more contexutal information but the crux of the
issue is that if we want SIMD to work in general we'll need to ensure that
whenever a function calls another they ABI of the arguments being passed is in
agreement.

One possible solution to this would be to insert "shim functions" where whenever
a `target_feature` mismatch is detected the compiler inserts a shim function
where you pass arguments via memory to the shim and then the shim loads the
values and calls the target function (where the shim and the target have the
same target features enabled). This unfortunately is quite nontrivial to
implement in rustc today (especially when accounting for function pointers and
such).

This commit takes a different solution, *always* passing SIMD arguments through
memory instead of passing as immediates. This strategy solves the problem at the
LLVM layer because the ABI between two functions never uses SIMD registers. This
also shouldn't be a hit to performance because SIMD performance is thought to
often rely on inlining anyway, where a `call` instruction, even if using SIMD
registers, would be disastrous to performance regardless. LLVM should then be
more than capable of fixing all our memory usage to use registers instead after
enough inlining has been performed.

Note that there's a few caveats to this commit though:

* The "platform intrinsic" ABI is omitted from "always pass via memory". This
  ABI is used to define intrinsics like `simd_shuffle4` where LLVM and rustc
  need to have the arguments as an immediate.

* Additionally this commit does *not* fix the `extern` ("C") ABI. This means
  that the bug in #44367 can still happen when using non-Rust-ABI functions. My
  hope is that before stabilization we can ban and/or warn about SIMD types in
  these functions (as AFAIK there's not much motivation to belong there anyway),
  but I'll leave that for a later commit and if this is merged I'll file a
  follow-up issue.

All in all this...

Closes #44367
2018-01-25 12:05:24 -08:00
Esteban Küber
2b73733496 Add description to field and method 2018-01-25 11:40:33 -08:00
Esteban Küber
6006b67e6e Fix regression: account for impl methods in arg count mismatch error 2018-01-25 10:49:28 -08:00
Alex Crichton
9852b23657 Rollup merge of #47600 - varkor:empty-never-array, r=eddyb
Fix type inhabitedness check for arrays

Arrays of uninhabited types were considered to also be uninhabited if
their length had not been evaluated, causing unsoundness.

Fixes #47563.

r? @eddyb
2018-01-25 12:48:52 -06:00
Alex Crichton
31f1aa5706 Rollup merge of #47529 - nikomatsakis:impl-trait-issue-38064, r=cramertj
track recursion limit when expanding existential impl trait

r? @cramertj
2018-01-25 12:48:51 -06:00
Alex Crichton
98b375483c Rollup merge of #47502 - petrochenkov:label, r=eddyb
AST/HIR: Add a separate structure for labels
2018-01-25 12:48:49 -06:00
Alex Crichton
304885d959 Rollup merge of #47460 - Mark-Simulacrum:bootstrap-check, r=alexcrichton
Add ./x.py check src/{libstd,libtest,librustc}

This currently only supports a limited subset of the full compilation,
but is likely 90% of what people will want and is possible without
building a full compiler (and also building LLVM). In theory, this means
that contributors who don't want to build LLVM now have an easy way to
compile locally, though running tests won't work.
2018-01-25 12:48:48 -06:00
Alex Crichton
1bdef2f8fc Rollup merge of #47453 - pftbest:nointas, r=alexcrichton
Fix no_integrated_as option to work with new codegen architecture.

Old implementation called the assembler once per crate, but we need to call
it for each object file instead, because a single crate can now have more
than one object file.

This patch fixes issue #45836 (Can't compile core for msp430 in release mode)

This change can be tested on x86_64 using
```sh
export RUSTFLAGS="-C no_integrated_as -C save_temps"
```

r? @alexcrichton
cc @japaric
2018-01-25 12:48:47 -06:00
Alex Crichton
6eb1430547 Rollup merge of #47439 - eddyb:issue-45662, r=nagisa
rustc_trans: ignore trailing padding larger than 8 bytes.

Fixes #45662 by ignoring a missing second register component, as it could be entirely padding.
2018-01-25 12:48:46 -06:00
Alex Crichton
fe2fb24228 Rollup merge of #47437 - eddyb:issue-38763, r=nagisa
rustc_trans: take into account primitives larger than 8 bytes.

Fixes #38763 by marking all "eightbytes" covered by a primitive appropriately, not just the first.
2018-01-25 12:48:45 -06:00
Alex Crichton
3b2403306f Rollup merge of #47415 - varkor:cgu-partition-heuristic, r=michaelwoerister
Add CGU size heuristic for partitioning

This addresses the concern of #47316 by estimating CGU size based on
the size of its MIR. Looking at the size estimate differences for a
small selection of crates, this heuristic produces different orderings,
which should more accurately reflect optimisation time. (Fixes #47316.)

r? @michaelwoerister
2018-01-25 12:48:44 -06:00
Oliver Middleton
77bc826f90 Fix rustdoc-js test 2018-01-25 17:18:52 +00:00
bors
9fd7da904b Auto merge of #47740 - GuillaumeGomez:rollup, r=GuillaumeGomez
Rollup of 10 pull requests

- Successful merges: #47534, #47609, #47679, #47691, #47700, #47702, #47717, #47721, #47726, #47729
- Failed merges:
2018-01-25 15:27:18 +00:00
Guillaume Gomez
89ff1226e5 Rollup merge of #47729 - alexcrichton:update-cargo, r=sfackler
Update Cargo submodule to master

Just a routine update
2018-01-25 15:52:19 +01:00
Guillaume Gomez
58e56cca56 Rollup merge of #47726 - pietroalbini:fix-nested-empty-groups-span, r=petrochenkov
Fix spans in unused import lint for nested groups

This fixes an inconsistency for empty nested groups, and adds a test for all the possible cases of the lint.

```
warning: unused imports: `*`, `Foo`, `baz::{}`, `foobar::*`
  --> test.rs:16:11
   |
16 | use foo::{Foo, bar::{baz::{}, foobar::*}, *};
   |           ^^^        ^^^^^^^  ^^^^^^^^^   ^
   |
   = note: #[warn(unused_imports)] on by default

warning: unused import: `*`
  --> test.rs:17:24
   |
17 | use foo::bar::baz::{*, *};
   |                        ^

warning: unused import: `use foo::{};`
  --> test.rs:18:1
   |
18 | use foo::{};
   | ^^^^^^^^^^^^
```

cc #44494
2018-01-25 15:52:18 +01:00
Guillaume Gomez
6f2a0c6f68 Rollup merge of #47721 - GuillaumeGomez:experimental-color, r=QuietMisdreavus
Fix experimental text display on default theme

r? @QuietMisdreavus
2018-01-25 15:52:17 +01:00
Guillaume Gomez
f769b9de02 Rollup merge of #47717 - evelynmitchell:47716-doc-fix, r=steveklabnik
fix for documentation error issue 47716

Fix #47716
2018-01-25 15:52:16 +01:00
Guillaume Gomez
a809da3fee Rollup merge of #47702 - etaoins:fix-into-cast-paren-precedence, r=petrochenkov
Fix into() cast paren check precedence

As discussed in #47699 the logic for determining if an expression needs parenthesis when suggesting an `.into()` cast is incorrect. Two broken examples from nightly are:

```
error[E0308]: mismatched types
 --> main.rs:4:10
  |
4 |     test(foo as i8);
  |          ^^^^^^^^^ expected i32, found i8
help: you can cast an `i8` to `i32`, which will sign-extend the source value
  |
4 |     test(foo as i8.into());
  |
```

```
error[E0308]: mismatched types
 --> main.rs:4:10
  |
4 |     test(*foo);
  |          ^^^^ expected i32, found i8
help: you can cast an `i8` to `i32`, which will sign-extend the source value
  |
4 |     test(*foo.into());
  |
```

As suggested by @petrochenkov switch the precedence check to `PREC_POSTFIX`. This catches both `as` and unary operators. Fixes #47699.

r? @petrochenkov
2018-01-25 15:52:15 +01:00
Guillaume Gomez
3a863a3f45 Rollup merge of #47700 - EdSchouten:cc104, r=kennytm
Remove workarounds for cc 1.0.3.

Now that the Rust codebase depends on cc 1.0.4, there is no longer any
need to specify a compiler for CloudABI manually. Cargo will
automatically call into the right compiler executable.
2018-01-25 15:52:15 +01:00
Guillaume Gomez
df55cee0e5 Rollup merge of #47691 - estebank:unknown-lang-item-sp, r=rkruppe
Point at unknown lang item attribute
2018-01-25 15:52:14 +01:00
Guillaume Gomez
bbdc5b9637 Rollup merge of #47679 - etaoins:remove-redundant-backtrace-hint, r=estebank
Remove broken redundant backtrace hint

When the compiler driver panics it attempts to show a hint about using `RUST_BACKTRACE`. However, the logic is currently reversed to the hint is only shown if `RUST_BACKTRACE` is **already**   set:

```shell
> RUST_BACKTRACE=1 rustc /dev/null --crate-type proc-macro
error: internal compiler error: unexpected panic
...
note: run with `RUST_BACKTRACE=1` for a backtrace

thread 'rustc' panicked at 'attempt to subtract with overflow', librustc_errors/emitter.rs:287:49
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.

> RUST_BACKTRACE=0 rustc /dev/null --crate-type proc-macro
error: internal compiler error: unexpected panic
...
thread 'rustc' panicked at 'attempt to subtract with overflow', librustc_errors/emitter.rs:287:49
note: Run with `RUST_BACKTRACE=1` for a backtrace.
```

As the `panic` itself already has a working `RUST_BACKTRACE` hint just remove the broken duplicate hint entirely.
2018-01-25 15:52:13 +01:00
Guillaume Gomez
83603b8294 Rollup merge of #47609 - ritiek:test-mutating-references, r=nikomatsakis
NLL test for mutating &mut references

As mentioned in #46361.

cc @nikomatsakis?
2018-01-25 15:52:12 +01:00
Guillaume Gomez
e26aac5a9f Rollup merge of #47534 - estebank:suggest-public-traits, r=petrochenkov
On missing method do not suggest private traits

When encountering a method call for an ADT that doesn't have any
implementation of it, we search for traits that could be implemented
that do have that method. Filter out private non-local traits that would
not be able to be implemented.

This doesn't account for public traits that are in a private scope, but
works as a first approximation and is a more correct behavior than the
current one.

Fix #45781.
2018-01-25 15:52:11 +01:00
bors
4cf26f8a13 Auto merge of #47686 - GuillaumeGomez:theme-fixes, r=QuietMisdreavus
Few fixes for multiple themes support feature

r? @QuietMisdreavus

Fixes #47695.
2018-01-25 06:12:06 +00:00
Manish Goregaokar
08ca4fd135 Add tests 2018-01-25 10:11:25 +05:30
Manish Goregaokar
c26f8877c7 Handle variants 2018-01-25 10:11:14 +05:30
Alex Crichton
15899b0c12 Update Cargo submodule to master
Just a routine update
2018-01-24 20:02:16 -08:00
bors
247835aacb Auto merge of #47374 - topecongiro:issue-47096, r=nikomatsakis
Simplify irrefutable slice patterns

Closes #47096.
2018-01-25 03:20:54 +00:00
Alex Crichton
54462f5705 Update compiler-builtins submodule
No sense of urgency, just wanted to make sure we don't lag too far behind!
2018-01-24 18:26:27 -08:00
bors
a0a9007f8d Auto merge of #47006 - bitshifter:stabilize-repr-align, r=eddyb
Stabilized `#[repr(align(x))]` attribute (RFC 1358)

Stabilzed `#[repr(align(x))]` with attr_literal syntax as proposed by @eddyb https://github.com/rust-lang/rust/issues/33626#issuecomment-348467804
2018-01-25 00:26:17 +00:00
Alex Crichton
e9a64996be llvm6: Update FreeBSD images to Ubuntu 18.04
Looks like the clang with 16.04 fails to compile LLVM 6, but it looks like clang
in 18.04 can indeed compile LLVM 6.
2018-01-24 15:34:41 -08:00
Pietro Albini
0847ac0265
Fix wrong span for nested empty groups 2018-01-24 23:46:02 +01:00
Guillaume Gomez
e1e991d3d2 Fix experimental text display on default theme 2018-01-24 19:15:41 +01:00
Marco A L Barbosa
5c5d9ddf64 Add powerpc to run-pass/conditional-compile-arch.rs 2018-01-24 15:16:18 -02:00
evelynmitchell
05652d2ae3
fix for documentation error issue 47716 2018-01-24 09:25:46 -07:00
Alex Crichton
63b3168448 llvm6: Don't clone LLVM modules on wasm
The comment for why cloning exists doesn't actually apply for wasm today and
apparently cloning is causing subtle bugs in LLVM, so let's just avoid it
altogether. More specifically after we emit the assembly for the wasm target we
don't actually use the module again, so there's no need to keep both around.

This seemed to be causing some scary verifier assertions in LLVM which seemed to
be uncovered by presumably (?) buggy behavior. Let's just avoid it for now and
make the wasm target slightly more lean in the process.
2018-01-24 07:18:02 -08:00
Alex Crichton
caedb36f08 llvm6: Different return value for writeArchive
Updated in llvm-mirror/llvm@203c90ba this function now just returns an `Error`,
so this updates the C++ bindings accordingly
2018-01-24 07:18:02 -08:00
Alex Crichton
b6fe1127e4 llvm6: Remove MIPS64 archive variant
It looks like LLVM also removed it in llvm-mirror/llvm@f45adc29d in favor of the
name "GNU64". This was added in the thought that we'd need such a variant when
adding mips64 support but we ended up not needing it! For now let's just
removing the various support on the Rust side of things.
2018-01-24 07:18:02 -08:00
Alex Crichton
9eeecd2ada llvm6: Tweak fast math intrinsics
Looks like they did some refactoring of flags in the backend and this should
catch us up! The "unsafe algebra" boolean has been split into a number of
boolean flags for various operations, and this updates to use the `setFast`
function which should hopefully have the same behavior as before.

This was updated in llvm-mirror/llvm@00e900afd
2018-01-24 07:18:02 -08:00
Alex Crichton
2a7ed74d96 llvm6: Missing include for LLVM 6 in PassWrapper.cpp
Just bog-standard compile error fixed by adding some new header files
2018-01-24 07:18:01 -08:00
Alex Crichton
03f86ae982 llvm6: CodeModel::{JIT,}Default no longer exists
LLVM has since removed the `CodeModel::Default` enum value in favor of an
`Optional` implementationg throughout LLVM. Let's mirror the same change in Rust
and update the various bindings we call accordingly.

Removed in llvm-mirror/llvm@9aafb854c
2018-01-24 07:18:01 -08:00
Pietro Albini
5faba281ad
Fix ICE when use trees have multiple empty nested groups 2018-01-24 13:13:39 +01:00
Manish Goregaokar
a4d36928fd Make it work for traits 2018-01-24 17:19:18 +05:30
Guillaume Gomez
e78f1392b7 Few fixes for multiple themes support feature 2018-01-24 12:13:52 +01:00
John Kåre Alsaker
62afc43a84 Make use of the implemented red/green algorithm for variance 2018-01-24 12:07:37 +01:00
Manish Goregaokar
5a89f40116 Handle methods 2018-01-24 16:34:47 +05:30
bors
a0dcecff90 Auto merge of #47300 - remexre:duration-constructors-as-const-fns, r=alexcrichton
Makes the constructors of Duration const fns.

This affects `Duration::new`, `Duration::from_secs`, `Duration::from_millis`, `Duration::from_micros`, and `Duration::from_nanos`.
2018-01-24 10:39:25 +00:00