The structure of the old translator as well as MIR assumed that drop glue cannot possibly panic and
translated the drops accordingly. However, in presence of `Drop::drop` this assumption can be
trivially shown to be untrue. As such, the Rust code like the following would never print number 2:
```rust
struct Droppable(u32);
impl Drop for Droppable {
fn drop(&mut self) {
if self.0 == 1 { panic!("Droppable(1)") } else { println!("{}", self.0) }
}
}
fn main() {
let x = Droppable(2);
let y = Droppable(1);
}
```
While the behaviour is allowed according to the language rules (we allow drops to not run), that’s
a very counter-intuitive behaviour. We fix this in MIR by allowing `Drop` to have a target to take
on divergence and connect the drops in such a way so the leftover drops are executed when some drop
unwinds.
Note, that this commit still does not implement the translator part of changes necessary for the
grand scheme of things to fully work, so the actual observed behaviour does not change yet. Coming
soon™.
We used to have CallKind only because there was a requirement to have all successors in a
contiguous memory block. Now that the requirement is gone, remove the CallKind and instead just
have the necessary information inline.
bors [Thu, 4 Feb 2016 13:41:42 +0000 (13:41 +0000)]
Auto merge of #31161 - sfackler:slice-to-socket-addrs, r=alexcrichton
This is useful when you have an API that takes a `T: ToSocketAddrs` and needs to turn that into an owned value which will be passed to another API taking `T: ToSocketAddrs` at a later time, for example: https://github.com/sfackler/rust-hyper-socks/blob/master/src/lib.rs#L15
Changed the description of the `make_mut` copy-on-write behaviour in arc.rs
The sentence "doesn't have one strong reference and no weak references." is a
hard to understand double negative, which can be much more easily explained.
bors [Thu, 4 Feb 2016 06:07:26 +0000 (06:07 +0000)]
Auto merge of #31326 - sdleffler:master, r=nikomatsakis
After the truly incredible and embarrassing mess I managed to make in my last pull request, this should be a bit less messy.
Fixes #31267 - with this change, the code mentioned in the issue compiles.
Found and fixed another issue as well - constants of zero-size types, when used in ExprRepeats inside associated constants, were causing the compiler to crash at the same place as #31267. An example of this:
```
struct Bar;
const BAZ: Bar = Bar;
struct Foo([Bar; 1]);
struct Biz;
impl Biz {
const BAZ: Foo = Foo([BAZ; 1]);
}
fn main() {
let foo = Biz::BAZ;
println!("{:?}", foo);
}
```
However, I'm fairly certain that my fix for this is not as elegant as it could be. The problem seems to occur only with an associated constant of a tuple struct containing a fixed size array which is initialized using a repeat expression, and when the element to be repeated provided to the repeat expression is another constant which is of a zero-sized type. The fix works by looking for constants and associated constants which are zero-width and consequently contain no data, but for which rustc is still attempting to emit an LLVM value; it simply stops rustc from attempting to emit anything. By my logic, this should work fine since the only values that are emitted in this case (according to the comments) are for closures with side effects, and constants will never have side effects, so it's fine to simply get rid of them. It fixes the error and things compile fine with it, but I have a sneaking suspicion that it could be done in a far better manner.
bors [Thu, 4 Feb 2016 02:46:44 +0000 (02:46 +0000)]
Auto merge of #30962 - Amanieu:non_volatile_atomic, r=alexcrichton
Rust currently emits atomic loads and stores with the LLVM `volatile` qualifier. This is unnecessary and prevents LLVM from performing optimization on these atomic operations.
Avoid quadratic growth of functions due to cleanups
If a new cleanup is added to a cleanup scope, the cached exits for that
scope are cleared, so all previous cleanups have to be translated
again. In the worst case this means that we get N distinct landing pads
where the last one has N cleanups, then N-1 and so on.
As new cleanups are to be executed before older ones, we can instead
cache the number of already translated cleanups in addition to the
block that contains them, and then only translate new ones, if any and
then jump to the cached ones, getting away with linear growth instead.
For the crate in #31381 this reduces the compile time for an optimized
build from >20 minutes (I cancelled the build at that point) to about 11
seconds. Testing a few crates that come with rustc show compile time
improvements somewhere between 1 and 8%. The "big" winner being
rustc_platform_intrinsics which features code similar to that in #31381.
bors [Wed, 3 Feb 2016 22:40:32 +0000 (22:40 +0000)]
Auto merge of #31078 - nbaksalyar:illumos, r=alexcrichton
This pull request adds support for [Illumos](http://illumos.org/)-based operating systems: SmartOS, OpenIndiana, and others. For now it's x86-64 only, as I'm not sure if 32-bit installations are widespread. This PR is based on #28589 by @potatosalad, and also closes #21000, #25845, and #25846.
Required changes in libc are already merged: https://github.com/rust-lang-nursery/libc/pull/138
Here's a snapshot required to build a stage0 compiler:
https://s3-eu-west-1.amazonaws.com/nbaksalyar/rustc-sunos-snapshot.tar.gz
It passes all checks from `make check`.
There are some changes I'm not quite sure about, e.g. macro usage in `src/libstd/num/f64.rs` and `DirEntry` structure in `src/libstd/sys/unix/fs.rs`, so any comments on how to rewrite it better would be greatly appreciated.
Also, LLVM configure script might need to be patched to build it successfully, or a pre-built libLLVM should be used. Some details can be found here: https://llvm.org/bugs/show_bug.cgi?id=25409
bors [Wed, 3 Feb 2016 19:26:05 +0000 (19:26 +0000)]
Auto merge of #30834 - reem:rwlock-read-guard-map, r=alexcrichton
This is very useful when the RwLock is synchronizing access to a data
structure and you would like to return or store guards which contain
references to data inside the data structure instead of the data structure
itself.
bors [Wed, 3 Feb 2016 06:38:01 +0000 (06:38 +0000)]
Auto merge of #31263 - dhuseby:fixing_bsd_builds, r=alexcrichton
Something went haywire with github last night and the old PR https://github.com/rust-lang/rust/pull/31230 got closed somehow. This new PR is to replace the old one. This incorporates all of the feedback from the other PR.
@alexcrichton I incorporated the suggestion from @semarie and the result is cleaner and clearer. I think this is ready to go.
bors [Wed, 3 Feb 2016 03:06:52 +0000 (03:06 +0000)]
Auto merge of #31319 - alexcrichton:msvc-backtraces, r=michaelwoerister
This mirrors the behavior of `clang-cl.exe` by adding a `CodeView` global
variable when emitting debug information. This should in turn help stack traces
that are generated when code is compiled with debuginfo enabled.
Rollup merge of #31329 - quodlibetor:no-const-doc-in-stable, r=alexcrichton
Fixes #31098
AFAICT this is the only place where rustdoc explicitly checks if we are on stable before emitting content, so I can't tell if this is the sane way to handle this, or if anything else should be done to make sure that nobody forgets to remove this check when `const` is stabilized.
Rollup merge of #27499 - barosl:macro-doc-raw-str-hashes, r=nikomatsakis
Any documentation comments that contain raw-string-looking sequences may pretty-print invalid code when expanding them, as the current logic always uses the `r"literal"` form, without appending any `#`s.
This commit calculates the minimum number of `#`s required to wrap a comment correctly and appends `#`s appropriately.
bors [Tue, 2 Feb 2016 17:11:48 +0000 (17:11 +0000)]
Auto merge of #31312 - alexcrichton:no-le-in-powerpc64le, r=alexcrichton
Currently the `mipsel-unknown-linux-gnu` target doesn't actually set the
`target_arch` value to `mipsel` but it rather uses `mips`. Alternatively the
`powerpc64le` target does indeed set the `target_arch` as `powerpc64le`,
causing a bit of inconsistency between theset two.
As these are just the same instance of one instruction set, let's use
`target_endian` to switch between them and only set the `target_arch` as one
value. This should cut down on the number of `#[cfg]` annotations necessary and
all around be a little more ergonomic.
bors [Tue, 2 Feb 2016 14:00:50 +0000 (14:00 +0000)]
Auto merge of #31254 - tmiasko:macro-pretty-print-fix, r=sfackler
Pretty printing of macro with braces but without terminated semicolon
removed more boxes from stack than it put there, resulting in panic.
This fixes the issue #30731.
Steve Klabnik [Tue, 2 Feb 2016 05:32:20 +0000 (00:32 -0500)]
Rollup merge of #31348 - alexcrichton:shuffle-tiers, r=steveklabnik
Some other shufflings as well:
* Three powerpc triples for Linux have been added recently
* An armv7 linux triple was added recently
* The 64-bit Solaris triple is now mentioned in tier 3
We are currently now also building nightlies for iOS, powerpc triples, and
armv7, but there hasn't been much vetting of the triples themselves so I've left
them in tier 3 for now.
Steve Klabnik [Tue, 2 Feb 2016 05:32:19 +0000 (00:32 -0500)]
Rollup merge of #31344 - steveklabnik:gh31334, r=alexcrichton
Fixes #31334
This is just a quicker fix for this issue; since I'm working on the next draft of the book, I don't want to put a huuuge amount of work into improving it here.
Steve Klabnik [Tue, 2 Feb 2016 05:32:18 +0000 (00:32 -0500)]
Rollup merge of #31339 - rthomas:doc, r=alexcrichton
When trying to run a specific test, I found the contributing docs a bit confusing and through a bit of googling found out that TESTNAME takes the fully qual'd name of the test.
I'm unsure if this can also take the source file, but I was unable to get that to work.
Steve Klabnik [Tue, 2 Feb 2016 05:32:18 +0000 (00:32 -0500)]
Rollup merge of #31327 - dirk:dirk/process-child-safety-docs, r=alexcrichton
`Drop` is not implemented for `Child`, so if it goes out of scope in Rust-land and gets deallocated the child process will continue to exist and execute. If users want a guarantee that the process has finished running and exited they must manually use `kill`, `wait`, or `wait_with_output`.
Steve Klabnik [Tue, 2 Feb 2016 05:32:18 +0000 (00:32 -0500)]
Rollup merge of #31270 - ruud-v-a:improve-e0507, r=steveklabnik
E0507 can occur when you try to move out of a member of a mutably borrowed struct, in which case `mem::replace` can help. Mentioning that here hopefully saves future users a trip to Google.
Steve Klabnik [Tue, 2 Feb 2016 05:32:17 +0000 (00:32 -0500)]
Rollup merge of #30971 - SDX2000:docfixes, r=steveklabnik
Updated documentation to clarify the difference between `and_then` and `map`. This also explains why we need `and_then` in addition to `map`. Please look at the diff for more information.
Alex Crichton [Sat, 30 Jan 2016 21:27:00 +0000 (13:27 -0800)]
Remove "powerpc64le" and "mipsel" target_arch
Currently the `mipsel-unknown-linux-gnu` target doesn't actually set the
`target_arch` value to `mipsel` but it rather uses `mips`. Alternatively the
`powerpc64le` target does indeed set the `target_arch` as `powerpc64le`,
causing a bit of inconsistency between theset two.
As these are just the same instance of one instruction set, let's use
`target_endian` to switch between them and only set the `target_arch` as one
value. This should cut down on the number of `#[cfg]` annotations necessary and
all around be a little more ergonomic.