Closes #3973.

Closes #3973.

Even if they used to test different things in the past, they are
now identical to other files.

Closes #11496

/usr/bin/env appears to be more portable.

Reject `struct Foo();` to fix #15095.

fix #12291

Turns out field names of struct variants are not encoded in crate metadata.

Closes #15031.

In other words, Late-bound regions that occur non-free should be
skipped.

Fix #10846 (specifically the ICE, not the weakness in the current type inference).

See http://static.rust-lang.org/doc/master/rustc/front/feature_gate/index.html for the problem this fixes.

This is a couple micro-optimizations I've been sitting on for a while. This inlines a couple functions that are important to the `std::io::mem`. Ultimately, this results in about a 15% performance increase in some micro-benchmarks for my [libserialize](https://github.com/erickt/rust-serde) rewrite.

Fixes #15080.

Some cleanups I'm sick of rebasing.

r? @pcwalton (do you agree with new names?)

Unit-like structs are written as `struct Foo;`, but we erroneously
accepted `struct Foo();` and took it to mean the same thing. Now we
don't, so use the `struct Foo;` form!

[breaking-change]

The #![feature(...)] line had been rendering in the docs as a header
because of Markdown syntax.

There were still Total{Ord,Eq} in docs and src/etc

Even if they used to test different things in the past, they are
now identical to other files.

Closes #11496

There were still Total{Ord,Eq} in docs and src/etc

In other words, Late-bound regions that occur non-free should be
skipped.

Fix #10846.

We currently compiled bools to i8 values, because there was a bug in
LLVM that sometimes caused miscompilations when using i1 in, for
example, structs.

Using i8 means a lot of unnecessary zero-extend and truncate operations
though, since we have to convert the value from and to i1 when using for
example icmp or br instructions. Besides the unnecessary overhead caused
by this, it also sometimes made LLVM miss some optimizations.

First, we have to fix some bugs concerning the handling of
attributes in foreign function declarations and calls. These
are required because the i1 type needs the ZExt attribute when
used as a function parameter or return type.

Then we have to update LLVM to get a bugfix without which LLVM
sometimes generates broken code when using i1.

And then, finally, we can switch bools over to i1.

This allows llvm to optimize away much of the overhead from using
the MemReader/MemWriters. My benchmarks showed it to shave 15% off
of my in progress serialization/json encoding.

Generally speaking, inlining doesn't really help out with
constructing vectors, except for when we construct a zero-sized
vector. This patch allows llvm to optimize this case away in
a lot of cases, which shaves off 4-8ns. It's not much, but it
might help in some inner loop somewhere.

before:

running 12 tests
test bench_extend_0          ... bench:       123 ns/iter (+/- 6)
test bench_extend_5          ... bench:       323 ns/iter (+/- 11)
test bench_from_fn_0         ... bench:         7 ns/iter (+/- 0)
test bench_from_fn_5         ... bench:        49 ns/iter (+/- 6)
test bench_from_iter_0       ... bench:        11 ns/iter (+/- 0)
test bench_from_iter_5       ... bench:       176 ns/iter (+/- 11)
test bench_from_slice_0      ... bench:         8 ns/iter (+/- 1)
test bench_from_slice_5      ... bench:        73 ns/iter (+/- 5)
test bench_new               ... bench:         0 ns/iter (+/- 0)
test bench_with_capacity_0   ... bench:         6 ns/iter (+/- 1)
test bench_with_capacity_100 ... bench:        41 ns/iter (+/- 3)
test bench_with_capacity_5   ... bench:        40 ns/iter (+/- 2)

after:

test bench_extend_0          ... bench:       123 ns/iter (+/- 7)
test bench_extend_5          ... bench:       339 ns/iter (+/- 27)
test bench_from_fn_0         ... bench:         7 ns/iter (+/- 0)
test bench_from_fn_5         ... bench:        54 ns/iter (+/- 4)
test bench_from_iter_0       ... bench:        11 ns/iter (+/- 1)
test bench_from_iter_5       ... bench:       182 ns/iter (+/- 16)
test bench_from_slice_0      ... bench:         4 ns/iter (+/- 0)
test bench_from_slice_5      ... bench:        62 ns/iter (+/- 3)
test bench_new               ... bench:         0 ns/iter (+/- 0)
test bench_with_capacity_0   ... bench:         0 ns/iter (+/- 0)
test bench_with_capacity_100 ... bench:        41 ns/iter (+/- 1)
test bench_with_capacity_5   ... bench:        41 ns/iter (+/- 3)

Closes #15031.

Fixes #15080.

We currently compiled bools to i8 values, because there was a bug in
LLVM that sometimes caused miscompilations when using i1 in, for
example, structs.

Using i8 means a lot of unnecessary zero-extend and truncate operations
though, since we have to convert the value from and to i1 when using for
example icmp or br instructions. Besides the unnecessary overhead caused
by this, it also sometimes made LLVM miss some optimizations.

Fixes #8106.

To fix #8106, we need an LLVM version that contains r211082 aka 0dee6756
which fixes a bug that blocks that issue.

There have been some tiny API changes in LLVM, and cmpxchg changed its
return type. The i1 part of the new return type is only interesting when
using the new weak cmpxchg, which we don't do.

When calling a foreign function, some arguments and/or return value
attributes are required to conform to the foreign ABI. Currently those
attributes are only added to the declaration of foreign functions. With
direct calls, this is no problem, because LLVM can see that those
attributes apply to the call. But with an indirect call, LLVM cannot do
that and the attribute is missing.

To fix that, we have to add those attribute to the calls to foreign
functions as well.

This also allows to remove the special handling of the SRet attribute,
which is ABI-dependent and will be set via the `attr` field of the
return type's `ArgType`.

The ArgType type gives us a generic way to specify an attribute for a
type to ensure ABI conformance for foreign functions. But the code that
actually sets the argument attributes in the function declaration
only sets the attribute for the return type when the type is indirect.

Since LLVMAddAttribute() doesn't allow to set attributes on the return
type, we have to use LLVMAddFunctionAttribute() instead.

This didn't cause problems yet, because currently only some indirect
types require attributes to be set.

Using something like:

```rust
box "string"
```

yields
```shell
"`~\"string\"` has been removed; use `\"string\".to_string()` instead"
```

Should the error message maybe say `box "string"` instead?

This will break code that looks like `Box<Trait+>`. Change that code to
`Box<Trait>` instead.

Closes #14925.

[breaking-change]

r? @brson

fix #12291

This commit makes several changes to the stability index infrastructure:

* Stability levels are now inherited lexically, i.e., each item's
  stability level becomes the default for any nested items.

* The computed stability level for an item is stored as part of the
  metadata. When using an item from an external crate, this data is
  looked up and cached.

* The stability lint works from the computed stability level, rather
  than manual stability attribute annotations. However, the lint still
  checks only a limited set of item uses (e.g., it does not check every
  component of a path on import). This will be addressed in a later PR,
  as part of issue #8962.

* The stability lint only applies to items originating from external
  crates, since the stability index is intended as a promise to
  downstream crates.

* The "experimental" lint is now _allow_ by default. This is because
  almost all existing crates have been marked "experimental", pending
  library stabilization. With inheritance in place, this would generate
  a massive explosion of warnings for every Rust program.

  The lint should be changed back to deny-by-default after library
  stabilization is complete.

* The "deprecated" lint still warns by default.

The net result: we can begin tracking stability index for the standard
libraries as we stabilize, without impacting most clients.

Closes #13540.