Add `x.py setup tools` which enables `download-rustc` by default
Helps with https://github.com/rust-lang/rust/issues/81930. I know I said in that issue that I should fix that rebasing rebuilds bootstrap, but the compile time improvement is so good I think it's ok to leave that fix for later (I still plan to work on it). I think all the outright bugs have been fixed :)
This builds on https://github.com/rust-lang/rust/pull/83368 so I can set the option to `if-unchanged`.
r? ```@Mark-Simulacrum```
Use tikv-jemallocator in rustc/rustdoc in addition to jemalloc-sys when enabled.
In https://github.com/rust-lang/rust/pull/81782 it was mentioned that one reason rustc may benefit from minimalloc is it doesn't use the `sdallocx` api from jemalloc.
Currently, on unix, rust uses jemalloc by importing its symbols to use them with the default, System (libc) global allocator.
This PR switches its global alloc to `tikv-jemallocator`, which correctly uses sized deallocation (https://docs.rs/tikv-jemallocator/0.4.1/src/tikv_jemallocator/lib.rs.html#121-126). `tikv-jemallocator`, as far as I can tell, is a more up-to-date set of bindings to jemalloc than `jemallocator`
The perf results of this pr are in large part due to the version upgrade of jemalloc, but sized deallocation has a non-trivial improvement, particularly to rustdoc.
This pr also includes changes to bootstrap to correctly pass the jemalloc feature through to the rustdoc build
fix `missing_panics_doc` not detecting `assert_eq!` and `assert_ne!`
fixes#6997
changelog: `missing_panics_doc` detects `assert_eq!` and `assert_ne!`
---
searching for `assert_eq!` and `assert_ne!` in `FindPanicUnwrap`
New Lint: `branches_sharing_code`
This lint checks if all `if`-blocks contain some statements that are the same and can be moved out of the blocks to prevent code duplication. Here is an example:
```rust
let _ = if ... {
println!("Start"); // <-- Lint for code duplication
let _a = 99;
println!("End"); // <-- Lint for code duplication
false
} else {
println!("Start");
let _b = 17;
println!("End");
false
};
```
This could be written as:
```rust
println!("Start");
let _ = if ... {
let _a = 99;
false
} else {
let _b = 17;
false
};
println!("End");
```
---
This lint will get masked by the `IF_SAME_THEN_ELSE` lint. I think it makes more sense to only emit one lint per if block. This means that the folloing example:
```rust
if ... {
let _a = 17;
} else {
let _a = 17;
}
```
Will only trigger the `IF_SAME_THEN_ELSE` lint and not the `SHARED_CODE_IN_IF_BLOCKS` lint.
---
closes: #5234
changelog: Added a new lint: `branches_sharing_code`
And hello to the one that is writing the changelog for this release :D
Prevent very long compilation runtimes in LateBoundRegionNameCollector
Fixes https://github.com/rust-lang/rust/issues/83150
On recursive types such as in the example given in https://github.com/rust-lang/rust/issues/83150, the current implementation of `LateBoundRegionNameCollector` has very long compilation runtimes. To prevent those we store the types visited in the `middle::ty::Visitor` implementation of `LateBoundRegionNameCollector` in a `SsoHashSet`.
Remove author requirement for `cargo_common_metadata`
This PR follows https://github.com/rust-lang/cargo/pull/9282, I'm not fully informed about all of this, it would be great if somebody knowledgeable about this topic agrees.
changelog: Changed `cargo_common_metadata` to stop linting on the optional author field.
Rollup of 5 pull requests
Successful merges:
- #83368 (Add `download-rustc = "if-unchanged"`)
- #83705 (Give a better error when --theme is not a CSS file)
- #83853 (Disallow the use of high byte registes as operands on x86_64)
- #83877 (Remove unnecessary exceptions to the platform-specific code check)
- #83878 (Fix racing file access in tidy)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
Fix racing file access in tidy
That should fix the failure in https://github.com/rust-lang/rust/pull/83776#issuecomment-813311289
The file is only created for a brief moment during the bins checks in the source directories while other checks may also be visiting the same directory. By skipping it we avoid file not found errors.
Remove unnecessary exceptions to the platform-specific code check
Some of these were just wrong, like src/librustc. Some looked outdated,
like std::f64. Not sure what was going on with the others - maybe this
check isn't as smart as it needs to be? But in the meantime it seems
silly to ignore the check if it will pass anyway.
Give a better error when --theme is not a CSS file
Before:
```
error: invalid argument: "bacon.toml"
```
After:
```
error: invalid argument: "bacon.toml"
|
= help: arguments to --theme must be CSS files
```
cc https://github.com/rust-lang/rust/pull/83478
Add `download-rustc = "if-unchanged"`
This allows keeping the setting to a fixed value without having to
toggle it when you want to work on the compiler instead of on tools.
This sets `BOOTSTRAP_DOWNLOAD_RUSTC` in bootstrap.py so rustbuild doesn't have to try and replicate its logic.
Helps with https://github.com/rust-lang/rust/issues/81930.
r? `@Mark-Simulacrum` cc `@camelid`
Previously, the types looked like this:
- None means this is not an associated item (but may be a variant field)
- Some(Err) means this is known to be an error. I think the only way that can happen is if it resolved and but you had your own anchor.
- Some(Ok(_, None)) was impossible.
Now, this returns a nested Option and does the error handling and
fiddling with the side channel in the caller. As a side-effect, it also
removes duplicate error handling.
This has one small change in behavior, which is that
`resolve_primitive_associated_item` now goes through `variant_field` if
it fails to resolve something. This is not ideal, but since it will be
quickly rejected anyway, I think the performance hit is worth the
cleanup.
This also fixes a bug where struct fields would forget to set the side
channel, adds a test for the bug, and ignores `private_intra_doc_links`
in rustc_resolve (since it's always documented with
--document-private-items).
* Added expression check for shared_code_in_if_blocks
* Finishing touches for the shared_code_in_if_blocks lint
* Applying PR suggestions
* Update lints yay
* Moved test into subfolder
Rollup of 8 pull requests
Successful merges:
- #81922 (Let `#[allow(unstable_name_collisions)]` work for things other than function)
- #82483 (Use FromStr trait for number option parsing)
- #82739 (Use the beta compiler for building bootstrap tools when `download-rustc` is set)
- #83650 (Update Source Serif to release 4.004)
- #83826 (List trait impls before deref methods in doc's sidebar)
- #83831 (Add `#[inline]` to IpAddr methods)
- #83863 (Render destructured struct function param names as underscore)
- #83865 (Don't report disambiguator error if link would have been ignored)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
Don't report disambiguator error if link would have been ignored
Fixes#83859.
This prevents us from warning on links such as `<hello@example.com>`.
Note that we still warn on links such as `<hello@localhost>` because
they have no dots in them. However, the links will still work, even
though a warning is reported.
r? ````@jyn514````
List trait impls before deref methods in doc's sidebar
This PR is acting directly on a suggestion made by ```````@jyn514``````` in #83133. I've tested the changes locally, and can confirm that it does in fact properly achieve what he thought it would. This PR also in turn closes#83133.
