This commit removes the `dep_graph` field from the `Session` type according to
issue #44390. Most of the fallout here was relatively straightforward and the
`prepare_session_directory` function was rejiggered a bit to reuse the results
in the later-called `load_dep_graph` function.
Closes#44390
Fix include! in doc tests
By making the path relative to the current file.
Fixes#43153
[breaking-change] - if you use `include!` inside a doc test, you'll need to change the path to be relative to the current file rather than relative to the working directory.
This attribute has two effects:
1. Items with this attribute and their children will have the "This is
supported on **** only" message attached in the documentation.
2. The items' doc tests will be skipped if the configuration does not
match.
This prevents compilation failure we want to document a platform-specific
module. Every function is replaced by `loop {}` using the same construct
as `--unpretty everybody_loops`.
Note also a workaround to #43636 is included: `const fn` will retain their
bodies, since the standard library has quite a number of them.
Prior to this PR, when we aborted because a "critical pass" failed, we
displayed the number of errors from that critical pass. While that's the
number of errors that caused compilation to abort in *that place*,
that's not what people really want to know. Instead, always report the
total number of errors, and don't bother to track the number of errors
from the last pass that failed.
This changes the compiler driver API to handle errors more smoothly,
and therefore is a compiler-api-[breaking-change].
Fixes#42793.
This change allows the user to add an `#[allow_fail]` attribute to
tests that will cause the test to compile & run, but if the test fails
it will not cause the entire test run to fail. The test output will
show the failure, but in yellow instead of red, and also indicate that
it was an allowed failure.
Consequently, session creation can no longer initialize LLVM.
The few places that use the compiler without going through
rustc_driver/CompilerCalls thus need to be careful to manually
initialize LLVM (via rustc_trans!) immediately after session
creation.
This means librustc is not rebuilt when LLVM changes.
Move the code for loading metadata from rlibs and dylibs from
rustc_metadata into rustc_trans, and introduce a trait to avoid
introducing a direct dependency on rustc_trans.
This means rustc_metadata is no longer rebuilt when LLVM changes.
rustdoc: fix doctests with non-feature crate attrs
Fixes#38129.
The book says that any top-level crate attributes at the beginning of a doctest are moved outside the generated `fn main`, but it was only checking for `#![feature`, not `#![`.
These attributes previously caused warnings but were then ignored, so in theory this could change the behavior of doctests in the wild.
Threads spawned by the test framework have their output captured by default, so
for `rustdoc --test` threads this propagates that capturing to the spawned
thread that we now have.
Closes#39327
run rustdoc tests in the same sort of thread rustc runs in
Not sure yet if this is the problem in #38973 but seems like an improvement regardless.
r? @brson
Add the actually_rustdoc option to the session that
is passed to phase_2_configure_and_expand function
inside the rustdoc test module.
This prevents the warning code from triggering when
parsing proc_macro_derive attributes, just like when
`--test` is normally invoked.
[10/n] Split constants and functions' arguments into disjoint bodies.
_This is part of a series ([prev](https://github.com/rust-lang/rust/pull/38053) | [next]()) of patches designed to rework rustc into an out-of-order on-demand pipeline model for both better feature support (e.g. [MIR-based](https://github.com/solson/miri) early constant evaluation) and incremental execution of compiler passes (e.g. type-checking), with beneficial consequences to IDE support as well.
If any motivation is unclear, please ask for additional PR description clarifications or code comments._
<hr>
Finishes the signature-body split started in #37918, namely:
* `trait` items are separated just like `impl` items were, for uniformity, closing #37712
* `static`s, `const`s (including associated ones), `enum` discriminants and array lengths get bodies
* even the count in "repeat expressions", i.e. `n` in `[x; n]`, which fixes#24414
* arguments' patterns are moved to the bodies, with the types staying in `FnDecl`
* `&self` now desugars to `self: &Self` instead of `self: &_` (similarly for other `self` forms)
* `astconv`'s and metadata's (for rustdoc) informative uses are explicitly ignored for the purposes of the dep graph. this could be fixed in the future by hashing the exact information being extracted about the arguments as opposed to generating a dependency on *the whole body*
