Currently we don't emit lifetime end markers when translating the
unwinding code. I omitted that when I added the support for lifetime
intrinsics, because I initially made the mistake of just returning true
in clean_on_unwind(). That caused almost all calls to be translated as
invokes, leading to quite awful results.
To correctly emit the lifetime end markers, we must differentiate
between cleanup that requires unwinding and such cleanup that just wants
to emit code during unwinding.
The translation is based on an early version of tutorial.md, thus most
of entries have been marked as fuzzy and actually they are incorrect.
Now tutorial.md is planed to be replaced with guide.md, so I'd suggest
removing translation files for a while.
/cc @gifnksm
librustc: Stop desugaring `for` expressions and translate them directly.
This makes edge cases in which the `Iterator` trait was not in scope
and/or `Option` or its variants were not in scope work properly.
This breaks code that looks like:
struct MyStruct { ... }
impl MyStruct {
fn next(&mut self) -> Option<int> { ... }
}
for x in MyStruct { ... } { ... }
Change ad-hoc `next` methods like the above to implementations of the
`Iterator` trait. For example:
impl Iterator<int> for MyStruct {
fn next(&mut self) -> Option<int> { ... }
}
Closes#15392.
[breaking-change]
This makes edge cases in which the `Iterator` trait was not in scope
and/or `Option` or its variants were not in scope work properly.
This breaks code that looks like:
struct MyStruct { ... }
impl MyStruct {
fn next(&mut self) -> Option<int> { ... }
}
for x in MyStruct { ... } { ... }
Change ad-hoc `next` methods like the above to implementations of the
`Iterator` trait. For example:
impl Iterator<int> for MyStruct {
fn next(&mut self) -> Option<int> { ... }
}
Closes#15392.
[breaking-change]
The translation is based on an early version of tutorial.md, thus most
of entries have been marked as fuzzy and actually they are incorrect.
Now tutorial.md is planed to be replaced with guide.md, so I'd suggest
removing translation files for a while.
Signed-off-by: OGINO Masanori <masanori.ogino@gmail.com>
Apparently the default getFile implementation for a memory buffer in LLVM ends
up requiring a null terminator at the end of the file. This isn't true a good
bit of the time apparently on OSX. There have been a number of failed
nightly/snapshot builds recently with this strange assertion.
This modifies the calls to MemoryBuffer::getFile to explicitly not ask for a
null terminator.
Apparently the default getFile implementation for a memory buffer in LLVM ends
up requiring a null terminator at the end of the file. This isn't true a good
bit of the time apparently on OSX. There have been a number of failed
nightly/snapshot builds recently with this strange assertion.
This modifies the calls to MemoryBuffer::getFile to explicitly not ask for a
null terminator.
Closes#15807 (Deprecate some unsafe functions in `str::raw` and remove `OwnedStr` trait)
Closes#15859 (Implement `Show` for `CString` and fix warning compiling tests for libcollections)
Closes#15911 (Updated LLVM for iOS)
Closes#15925 (libsyntax: Remove `~self` and `mut ~self` from the language.)
Closes#15930 (Add examples for Checked[Add|Sub|Mul|Div])
Closes#15933 (rustdoc: make table of contents optional)
Closes#15937 (librustc: Make bare functions implement the `FnMut` trait.)
Closes#15938 (librustc: Check structure constructors against their types.)
Closes#15939 (rustdoc: Add a --crate-name option)
Closes#15942 (Document trie collections)
Closes#15943 (Document SmallIntMap)
This is done entirely in the libraries for functions up to 16 arguments.
A macro is used so that more arguments can be easily added if we need.
Note that I had to adjust the overloaded call algorithm to not try
calling the overloaded call operator if the callee is a built-in
function type, to prevent loops.
Closes#15448.
rustdoc currently determines whether to produce a table of
contents (along with numbered sections) from the input type: yes for
markdown input, no for Rust input. This commit adds a flag to disable
the table of contents for markdown input, which is useful for embedding
the output in a larger context.
This eliminates the last vestige of the `~` syntax.
Instead of `~self`, write `self: Box<TypeOfSelf>`; instead of `mut
~self`, write `mut self: Box<TypeOfSelf>`, replacing `TypeOfSelf` with
the self-type parameter as specified in the implementation.
Closes#13885.
[breaking-change]
This also removes checks in other methods of `CString`
Breaking changes:
* `CString::new` now fails if `buf` is null. To avoid this add a check
before creatng a new `CString` .
* The `is_null` and `is_not_null` methods are deprecated, because a
`CString` cannot be null.
* Other methods which used to fail if the `CString` was null do not fail anymore
[breaking-change]
This trait was only implemented by `String`. It provided the methods
`into_bytes` and `append`, both of which **are already implemented as normal
methods** of `String` (not as trait methods). This change improves the
consistency of strings.
This shouldn't break any code, except if somebody has implemented
`OwnedStr` for a user-defined type.
The alignment of the line continuation backslashes is rather inconsistent. These commits solve that by removing the extra whitespace and adding a space where there previously was none. An alternative solution would be to fix the alignment.
The allocas used in match expression currently don't get good lifetime
markers, in fact they only get lifetime start markers, because their
lifetimes don't match to cleanup scopes.
While the bindings themselves are bog standard and just need a matching
pair of start and end markers, they might need them twice, once for a
guard clause and once for the match body.
The __llmatch alloca OTOH needs a single lifetime start marker, but
when there's a guard clause, it needs two end markers, because its
lifetime ends either when the guard doesn't match or after the match
body.
With these intrinsics in place, LLVM can now, for example, optimize
code like this:
````rust
enum E {
A1(int),
A2(int),
A3(int),
A4(int),
}
pub fn variants(x: E) {
match x {
A1(m) => bar(&m),
A2(m) => bar(&m),
A3(m) => bar(&m),
A4(m) => bar(&m),
}
}
````
To a single call to bar, using only a single stack slot. It still fails
to eliminate some of checks.
````gas
.Ltmp5:
.cfi_def_cfa_offset 16
movb (%rdi), %al
testb %al, %al
je .LBB3_5
movzbl %al, %eax
cmpl $1, %eax
je .LBB3_5
cmpl $2, %eax
.LBB3_5:
movq 8(%rdi), %rax
movq %rax, (%rsp)
leaq (%rsp), %rdi
callq _ZN3bar20hcb7a0d8be8e17e37daaE@PLT
popq %rax
retq
````
Refs #15665
Semaphores are not currently designed to handle this case correctly, leading to
very strange behavior. Semaphores as written are intended to count *resources*
and it's not possible to have a negative number of resources.
This alters the behavior and documentation to note that the task will be failed
if the initial count is 0.
Closes#15758
At the moment, writing generic functions for integer types that involve shifting is rather verbose. For example, a function at shifts an integer left by 1 currently requires
use std::num::One;
fn f<T: Int>(x : T) -> T {
x << One::one()
}
If the shift amount is not 1, it's even worse:
use std::num::FromPrimitive;
fn f<T: Int + FromPrimitive>(x: T) -> T {
x << FromPrimitive::from_int(2).unwrap()
}
This patch allows the much simpler implementation
fn f<T: Int>(x: T) -> T {
x << 2
}
It accomplishes this by changing the built-in integer types (and the `Int` trait) to implement `Shl<uint, T>` instead of `Shl<T, T>` as it currently is defined. Note that the internal implementations of `shl` already cast the right-hand side to `uint`. `BigInt` also implements `Shl<uint, BigInt>`, so this increases consistency.
All of the above applies similarly to right shifts, i.e., `Shr<uint, T>`.
This fixes naming conventions for `push`/`pop` from either end of a structure by partially implementing @erickt's suggestion from https://github.com/rust-lang/rust/issues/10852#issuecomment-30823343, namely:
* push/pop from the 'back' are called `push` and `pop`.
* push/pop from the 'front' are called `push_front` and `pop_front`.
* `push`/`pop` are declared on the `MutableSeq` trait.
* Implement `MutableSeq` for `Vec`, `DList`, and `RingBuf`.
* Add `MutableSeq` to the prelude.
I did not make any further refactorings because there is some more extensive thought that needs to be put into the collections traits. This is an easy first step that should close https://github.com/rust-lang/rust/issues/10852.
I left the `push_back` and `pop_back` methods on `DList` and `RingBuf` deprecated. Because `MutableSeq` is in the prelude it shouldn't break many, but it is a breaking change.
