Prevents LateContext::maybe_typeck_results() from returning data in a
nested item without a body. Consequently, LateContext::qpath_res is less
likely to ICE when called in a nested item. Would have prevented
rust-lang/rust-clippy#4545, presumably.

Actually fulfills the documented guarantees.

Add expected error

Add comment

Tweak comment wording

Fix after rebase to updated master

Fix after rebase to updated master

Distinguish mutation in normal and move closures

Tweak error message

Fix error message for nested closures

Refactor code showing mutated upvar in closure

Remove debug assert

B

The angle brackets were not rendered, so code like this:

    some_func: for<'a> fn(val: &'a i32) -> i32

would be rendered as:

    some_func: fn'a(val: &'a i32) -> i32

However, rendering with angle brackets is still invalid syntax:

    some_func: fn<'a>(val: &'a i32) -> i32

so now it renders correctly as:

    some_func: for<'a> fn(val: &'a i32) -> i32

-----

However, note that this code:

    some_trait: dyn for<'a> Trait<'a>

will still render as:

    some_trait: dyn Trait<'a>

which is not invalid syntax, but is still unclear. Unfortunately I think
it's hard to fix that case because there isn't enough information in the
`rustdoc::clean::Type` that this code operates on. Perhaps that case can
be fixed in a later PR.

cc rust-lang#79813

This PR adds an allow-by-default future-compatibility lint
`SEMICOLON_IN_EXPRESSIONS_FROM_MACROS`. It fires when a trailing semicolon in a
macro body is ignored due to the macro being used in expression
position:

```rust
macro_rules! foo {
    () => {
        true; // WARN
    }
}

fn main() {
    let val = match true {
        true => false,
        _ => foo!()
    };
}
```

The lint takes its level from the macro call site, and
can be allowed for a particular macro by adding
`#[allow(semicolon_in_expressions_from_macros)]`.

The lint is set to warn for all internal rustc crates (when being built
by a stage1 compiler). After the next beta bump, we can enable
the lint for the bootstrap compiler as well.

This commit adds a new stable codegen option to rustc,
`-Csplit-debuginfo`. The old `-Zrun-dsymutil` flag is deleted and now
subsumed by this stable flag. Additionally `-Zsplit-dwarf` is also
subsumed by this flag but still requires `-Zunstable-options` to
actually activate. The `-Csplit-debuginfo` flag takes one of
three values:

* `off` - This indicates that split-debuginfo from the final artifact is
  not desired. This is not supported on Windows and is the default on
  Unix platforms except macOS. On macOS this means that `dsymutil` is
  not executed.

* `packed` - This means that debuginfo is desired in one location
  separate from the main executable. This is the default on Windows
  (`*.pdb`) and macOS (`*.dSYM`). On other Unix platforms this subsumes
  `-Zsplit-dwarf=single` and produces a `*.dwp` file.

* `unpacked` - This means that debuginfo will be roughly equivalent to
  object files, meaning that it's throughout the build directory
  rather than in one location (often the fastest for local development).
  This is not the default on any platform and is not supported on Windows.

Each target can indicate its own default preference for how debuginfo is
handled. Almost all platforms default to `off` except for Windows and
macOS which default to `packed` for historical reasons.

Some equivalencies for previous unstable flags with the new flags are:

* `-Zrun-dsymutil=yes` -> `-Csplit-debuginfo=packed`
* `-Zrun-dsymutil=no` -> `-Csplit-debuginfo=unpacked`
* `-Zsplit-dwarf=single` -> `-Csplit-debuginfo=packed`
* `-Zsplit-dwarf=split` -> `-Csplit-debuginfo=unpacked`

Note that `-Csplit-debuginfo` still requires `-Zunstable-options` for
non-macOS platforms since split-dwarf support was *just* implemented in
rustc.

There's some more rationale listed on rust-lang#79361, but the main gist of the
motivation for this commit is that `dsymutil` can take quite a long time
to execute in debug builds and provides little benefit. This means that
incremental compile times appear that much worse on macOS because the
compiler is constantly running `dsymutil` over every single binary it
produces during `cargo build` (even build scripts!). Ideally rustc would
switch to not running `dsymutil` by default, but that's a problem left
to get tackled another day.

Closes rust-lang#79361

rustc: Stabilize `-Zrun-dsymutil` as `-Csplit-debuginfo`

This commit adds a new stable codegen option to rustc,
`-Csplit-debuginfo`. The old `-Zrun-dsymutil` flag is deleted and now
subsumed by this stable flag. Additionally `-Zsplit-dwarf` is also
subsumed by this flag but still requires `-Zunstable-options` to
actually activate. The `-Csplit-debuginfo` flag takes one of
three values:

* `off` - This indicates that split-debuginfo from the final artifact is
  not desired. This is not supported on Windows and is the default on
  Unix platforms except macOS. On macOS this means that `dsymutil` is
  not executed.

* `packed` - This means that debuginfo is desired in one location
  separate from the main executable. This is the default on Windows
  (`*.pdb`) and macOS (`*.dSYM`). On other Unix platforms this subsumes
  `-Zsplit-dwarf=single` and produces a `*.dwp` file.

* `unpacked` - This means that debuginfo will be roughly equivalent to
  object files, meaning that it's throughout the build directory
  rather than in one location (often the fastest for local development).
  This is not the default on any platform and is not supported on Windows.

Each target can indicate its own default preference for how debuginfo is
handled. Almost all platforms default to `off` except for Windows and
macOS which default to `packed` for historical reasons.

Some equivalencies for previous unstable flags with the new flags are:

* `-Zrun-dsymutil=yes` -> `-Csplit-debuginfo=packed`
* `-Zrun-dsymutil=no` -> `-Csplit-debuginfo=unpacked`
* `-Zsplit-dwarf=single` -> `-Csplit-debuginfo=packed`
* `-Zsplit-dwarf=split` -> `-Csplit-debuginfo=unpacked`

Note that `-Csplit-debuginfo` still requires `-Zunstable-options` for
non-macOS platforms since split-dwarf support was *just* implemented in
rustc.

There's some more rationale listed on rust-lang#79361, but the main gist of the
motivation for this commit is that `dsymutil` can take quite a long time
to execute in debug builds and provides little benefit. This means that
incremental compile times appear that much worse on macOS because the
compiler is constantly running `dsymutil` over every single binary it
produces during `cargo build` (even build scripts!). Ideally rustc would
switch to not running `dsymutil` by default, but that's a problem left
to get tackled another day.

Closes rust-lang#79361

…micolon, r=petrochenkov

Add `SEMICOLON_IN_EXPRESSIONS_FROM_MACROS` lint

cc rust-lang#79813

This PR adds an allow-by-default future-compatibility lint
`SEMICOLON_IN_EXPRESSIONS_FROM_MACROS`. It fires when a trailing semicolon in a
macro body is ignored due to the macro being used in expression
position:

```rust
macro_rules! foo {
    () => {
        true; // WARN
    }
}

fn main() {
    let val = match true {
        true => false,
        _ => foo!()
    };
}
```

The lint takes its level from the macro call site, and
can be allowed for a particular macro by adding
`#[allow(macro_trailing_semicolon)]`.

The lint is set to warn for all internal rustc crates (when being built
by a stage1 compiler). After the next beta bump, we can enable
the lint for the bootstrap compiler as well.

…llaumeGomez,jyn514

rustdoc: Render HRTB correctly for bare functions

The angle brackets were not rendered, so code like this:

    some_func: for<'a> fn(val: &'a i32) -> i32

would be rendered as:

    some_func: fn'a(val: &'a i32) -> i32

However, rendering with angle brackets is still invalid syntax:

    some_func: fn<'a>(val: &'a i32) -> i32

so now it renders correctly as:

    some_func: for<'a> fn(val: &'a i32) -> i32

-----

However, note that this code:

    some_trait: dyn for<'a> Trait<'a>

will still render as:

    some_trait: dyn Trait<'a>

which is not invalid syntax, but is still unclear. Unfortunately I think
it's hard to fix that case because there isn't enough information in the
`rustdoc::clean::Type` that this code operates on. Perhaps that case can
be fixed in a later PR.

r? ``@jyn514``

Use -target when linking binaries for Mac Catalyst

When running `rustc` with `-target x86_64-apple-ios-macabi`, the linker
eventually gets run with `-arch x86_64`, because the linker back end splits the
LLVM target triple and uses the first token as the target architecture. However,
this does not work for the Mac Catalyst ABI, which is a separate target from
Darwin.

Specifying the full target triple with `-target` allows Mac Catalyst binaries to
link and run.

closes rust-lang#80202

…1011

Point to span of upvar making closure FnMut

For rust-lang#80313.

Improve safety of `LateContext::qpath_res`

This is my first rustc code change, inspired by hacking on clippy!

The first change is to clear cached `TypeckResults` from `LateContext` when visiting a nested item. I took a hint from [here](https://github.com/rust-lang/rust/blob/5e91c4ecc09312d8b63d250a432b0f3ef83f1df7/compiler/rustc_privacy/src/lib.rs#L1300).

Clippy has a `qpath_res` util function to avoid a possible ICE in `LateContext::qpath_res`. But the docs of `LateContext::qpath_res` promise no ICE. So this updates the `LateContext` method to keep its promises, and removes the util function.

Related: rust-lang/rust-clippy#4545

CC ````````````@eddyb```````````` since you've done related work
CC ````````````@flip1995```````````` FYI

…illaumeGomez

Split rustdoc JSON types into separately versioned crate

For now just an in-tree change.

In the future, this may be exposed as a standalone crate with standard semver.

Fuse inner iterator in FlattenCompat and improve related tests

Fixes rust-lang#81248

clean up some const error reporting around promoteds

These are some error reporting simplifications enabled by rust-lang#80579.

Further simplifications are possible but could be blocked on making `const_err` a hard error.

r? ``````@oli-obk``````

…r=Nemo157

Fix rustdoc text selection for page titles

Fixes text selection of page titles by placing the `h1.fqn span.in-band` element before `h1.fqn span.out-of-band`.

Before vs after:
https://user-images.githubusercontent.com/11315492/105768203-55708700-5f5c-11eb-924b-4e7527ffe147.mp4

Retry of PR rust-lang#81397 due to merge conflicts confusing me. Recreated the same changes as in that PR, but now when I run `./x.py doc library/std`, no changes I make to `src/librustdoc/html/render/mod.rs` to the file are reflected in the built doc files, even if I delete the `build` folder. I'm guessing there's some cache I'm missing?

r? `@Nemo157`