The existing implementation uses Python to launch a set of Rust-written
binaries. Unfortunately, this is currently broken; it seems that some
updates meant it no longer compiles.
There is also a problem that support for more float types (`f16`,
`f128`) would be difficult to add since this is very specialized to
`f32` and `f64`.
Because of these sortcomings, migrate to a version written in Rust. This
version should be significantly faster; test generators can execute in
parallel, and test cases are chunked and parallelized. This should also
resolve the preexisting "... the worker processes are leaked and stick
around forever" comment.
This change also introduces genericism over float types and properties,
meaning it will be much easier to extend support to newly added types.
`num::BigRational` is used in place of Python's fractions for
infinite-precision calculations.
Add Natvis visualiser and debuginfo tests for `f16`
To render `f16`s in debuggers on MSVC targets, this PR changes the compiler to output `f16`s as `struct f16 { bits: u16 }`, and includes a Natvis visualiser that manually converts the `f16`'s bits to a `float` which is can then be displayed by debuggers. `gdb`, `lldb` and `cdb` tests are also included for `f16` .
`f16`/`f128` MSVC debug info issue: #121837
Tracking issue: #116909
By default, `x fmt` formats/checks modified files. But it also lets you
choose one or more paths instead.
This adds significant complexity to `x fmt`. Explicit paths are
specified via `WalkBuilder::add` rather than `OverrideBuilder::add`. The
`ignore` library is not simple, and predicting the interactions between
the two mechanisms is difficult.
Here's a particularly interesting case.
- You can request a path P that is excluded by the `ignore` list in the
`rustfmt.toml`. E.g. `x fmt tests/ui/` or `x fmt tests/ui/bitwise.rs`.
- `x fmt` will add P to the walker (via `WalkBuilder::add`), traverse it
(paying no attention to the `ignore` list from the `rustfmt.toml`
file, due to the different mechanism), and call `rustfmt` on every
`.rs` file within it.
- `rustfmt` will do nothing to those `.rs` files, because it *also*
reads `rustfmt.toml` and sees that they match the `ignore` list!
It took me *ages* to debug and understand this behaviour. Not good!
`x fmt` even lets you name a path below the current directory. This was
intended to let you do things like `x fmt std` that mirror things like
`x test std`. This works by looking for `std` and finding `library/std`,
and then formatting that. Unfortuantely, this motivating case now gives
an error. When support was added in #107944, `library/std` was the only
directory named `std`. Since then, `tests/ui/std` was added, and so `x
fmt std` now gives an error.
In general, explicit paths don't seem particularly useful. The only two
cases `x fmt` really needs are:
- format/check the files I have modified (99% of uses)
- format/check all files
(While respecting the `ignore` list in `rustfmt.toml`, of course.)
So this commit moves to that model. `x fmt` will now give an error if
given an explicit path. `x fmt` now also supports a `--all` option. (And
running with `GITHUB_ACTIONS=true` also causes everything to be
formatted/checked, as before.) Much simpler!
Summary:
I landed a fix last year to enable `DW_TAG_variant_part` encoding in LLDBs (https://reviews.llvm.org/D149213). This PR is a corresponding fix in synthetic formatters to decode that information.
This is in no way perfect implementation but at least it improves the status quo. But most types of enums will be visible and debuggable in some way.
I've also updated most of the existing tests that touch enums and re-enabled test cases based on LLDB for enums.
Test Plan:
ran tests `./x test tests/debuginfo/`. Also tested manually in LLDB CLI and LLDB VSCode
Other Thoughs
A better approach would probably be adopting [formatters from codelldb](https://github.com/vadimcn/codelldb/blob/master/formatters/rust.py). There is some neat hack that hooks up summary provider via synthetic provider which can ultimately fix more display issues for Rust types and enums too. But getting it to work well might take more time that I have right now.
Support running library tests in Miri
This adds a new bootstrap subcommand `./x.py miri` which can test libraries in Miri. This is in preparation for eventually doing that as part of bors CI, but this PR only adds the infrastructure, and doesn't enable it yet.
`@rust-lang/bootstrap` should this be `x.py test --miri library/core` or `x.py miri library/core`? The flag has the advantage that we don't have to copy all the arguments from `Subcommand::Test`. It has the disadvantage that most test steps just ignore `--miri` and still run tests the regular way. For clippy you went the route of making it a separate subcommand. ~~I went with a flag now as that seemed easier, but I can change this.~~ I made it a new subcommand. Note however that the regular cargo invocation would be `cargo miri test ...`, so `x.py` is still going to be different in that the `test` is omitted. That said, we could also make it `./x.py miri-test` to make that difference smaller -- that's in fact more consistent with the internal name of the command when bootstrap invokes cargo.
`@rust-lang/libs` ~~unfortunately this PR does some unholy things to the `lib.rs` files of our library crates.~~
`@m-ou-se` found a way that entirely avoids library-level hacks, except for some new small `lib.miri.rs` files that hopefully you will never have to touch. There's a new hack in cargo-miri but there it is in good company...
Rename `pointer` field on `Pin`
A few days ago, I was helping another user create a self-referential type using `PhantomPinned`. However, I noticed an odd behavior when I tried to access one of the type's fields via `Pin`'s `Deref` impl:
```rust
use std::{marker::PhantomPinned, ptr};
struct Pinned {
data: i32,
pointer: *const i32,
_pin: PhantomPinned,
}
fn main() {
let mut b = Box::pin(Pinned {
data: 42,
pointer: ptr::null(),
_pin: PhantomPinned,
});
{
let pinned = unsafe { b.as_mut().get_unchecked_mut() };
pinned.pointer = &pinned.data;
}
println!("{}", unsafe { *b.pointer });
}
```
```rust
error[E0658]: use of unstable library feature 'unsafe_pin_internals'
--> <source>:19:30
|
19 | println!("{}", unsafe { *b.pointer });
| ^^^^^^^^^
error[E0277]: `Pinned` doesn't implement `std::fmt::Display`
--> <source>:19:20
|
19 | println!("{}", unsafe { *b.pointer });
| ^^^^^^^^^^^^^^^^^^^^^ `Pinned` cannot be formatted with the default formatter
|
= help: the trait `std::fmt::Display` is not implemented for `Pinned`
= note: in format strings you may be able to use `{:?}` (or {:#?} for pretty-print) instead
= note: this error originates in the macro `$crate::format_args_nl` which comes from the expansion of the macro `println` (in Nightly builds, run with -Z macro-backtrace for more info)
```
Since the user named their field `pointer`, it conflicts with the `pointer` field on `Pin`, which is public but unstable since Rust 1.60.0 with #93176. On versions from 1.33.0 to 1.59.0, where the field on `Pin` is private, this program compiles and prints `42` as expected.
To avoid this confusing behavior, this PR renames `pointer` to `__pointer`, so that it's less likely to conflict with a `pointer` field on the underlying type, as accessed through the `Deref` impl. This is technically a breaking change for anyone who names their field `__pointer` on the inner type; if this is undesirable, it could be renamed to something more longwinded. It's also a nightly breaking change for any external users of `unsafe_pin_internals`.
The internal, unstable field of `Pin` can conflict with fields from the
inner type accessed via the `Deref` impl. Rename it from `pointer` to
`__pointer`, to make it less likely to conflict with anything else.
bump bootstrap dependencies
This PR removes hard-coded patch versions, updates bootstrap's dependency stack to recent versions (some of the versions were released 3-4 years ago), and removes a few dependencies from bootstrap.
Removed dependencies:
add more niches to rawvec
Previously RawVec only had a single niche in its `NonNull` pointer. With this change it now has `isize::MAX` niches since half the value-space of the capacity field is never needed, we can't have a capacity larger than isize::MAX.
As bootstrap locks its entire build directory, parallel bootstrapping
for anything becomes impossible. This change enables developers to bypass
the locking mechanism when it is unnecessary for their specific use case.
Signed-off-by: onur-ozkan <work@onurozkan.dev>
Run tidy license checker on more workspaces
The license checker didn't run on several workspaces before this PR. The same applied to the "external package sources" check. There were also two missing lockfiles which I have added now.
add bootstrap flag `--skip-stage0-validation`
This change introduces the `--skip-stage0-validation` flag, which permits the use of any desired version of the stage0 compiler without verifying its version.
Additionally, stage0 compiler validation check is reverted(#115103) to its default enabled state.
Helps to #115065
r? Mark-Simulacrum
Rewrite gdb pretty-printer registration
Currently, the Rust pretty-printers are registered in gdb using the uninformative name "lookup":
(gdb) info pretty-printer
global pretty-printers:
[...]
objfile /home/tromey/[...]
lookup
It's nicer for users if the top-level registration is given a clear name. Additionally, gdb lets users individually enable and disable specific printers, provided they are registered correctly.
This patch implements both these ideas. Now the output looks like:
(gdb) info pretty-printer
global pretty-printers:
[...]
objfile /home/tromey/[...]
rust
StdArc
StdBTreeMap
StdBTreeSet
StdCell
StdHashMap
StdHashSet
StdNonZeroNumber
StdOsString
StdRc
StdRef
StdRefCell
StdRefMut
StdSlice
StdStr
StdString
StdVec
StdVecDeque
This change introduces the --skip-stage0-validation flag,
which permits the use of any desired version of the stage0
compiler without verifying its version.
Signed-off-by: onur-ozkan <work@onurozkan.dev>
GDB 14 has a "gdb.ValuePrinter" tag class that was introduced to let
GDB evolve the pretty-printer API. Users of this tag are required to
hide any local attributes or methods. This patch makes this change to
the Rust pretty-printers. At present this is just a cleanup,
providing the basis for any future changes.
Currently, the Rust pretty-printers are registered in gdb using the
uninformative name "lookup":
(gdb) info pretty-printer
global pretty-printers:
[...]
objfile /home/tromey/[...]
lookup
It's nicer for users if the top-level registration is given a clear
name. Additionally, gdb lets users individually enable and disable
specific printers, provided they are registered correctly.
This patch implements both these ideas. Now the output looks like:
(gdb) info pretty-printer
global pretty-printers:
[...]
objfile /home/tromey/[...]
rust
StdArc
StdBTreeMap
StdBTreeSet
StdCell
StdHashMap
StdHashSet
StdNonZeroNumber
StdOsString
StdRc
StdRef
StdRefCell
StdRefMut
StdSlice
StdStr
StdString
StdVec
StdVecDeque
