Miri: basic dyn* support
As usual I am very unsure about the dynamic dispatch stuff, but it passes even the `Pin<&mut dyn* Trait>` test so that is something.
TBH I think it was a mistake to make `dyn Trait` and `dyn* Trait` part of the same `TyKind` variant. Almost everywhere in Miri this lead to the wrong default behavior, resulting in strange ICEs instead of nice "unimplemented" messages. The two types describe pretty different runtime data layout after all.
Strangely I did not need to do the equivalent of [this diff](https://github.com/rust-lang/rust/pull/106532#discussion_r1087095963) in Miri. Maybe that is because the unsizing logic matches on `ty::Dynamic(.., ty::Dyn)` already? In `unsized_info` I don't think the `target_dyn_kind` can be `DynStar`, since then it wouldn't be unsized!
r? `@oli-obk` Cc `@eholk` (dyn-star) https://github.com/rust-lang/rust/issues/102425
Improve building compiler artifacts output
Fixes#108051
``@Manishearth,`` ``@jyn514`` mentioned you might be interested in these changes to the outputs.
Including unstable lints in the lint group produces unintuitive behavior
on stable (see #106289). Meanwhile, if we only included unstable lints
on nightly and not on stable, we could end up with confusing bugs that
were hard to compare across versions of Rust that lacked code changes.
I think that only including stable lints in `rustdoc::all`, no matter
the release channel, is the most intuitive option. Users can then
control unstable lints individually, which is reasonable since they have
to enable the feature gates individually anyway.
Make codegen choose whether to emit overflow checks
ConstProp and DataflowConstProp currently have a specific code path not to propagate constants when they overflow. This is meant to have the correct behaviour when inlining from a crate with overflow checks (like `core`) into a crate compiled without.
This PR shifts the behaviour change to the `Assert(Overflow*)` MIR terminators: if the crate is compiled without overflow checks, just skip emitting the assertions. This is already what happens with `OverflowNeg`.
This allows ConstProp and DataflowConstProp to transform `CheckedBinaryOp(Add, u8::MAX, 1)` into `const (0, true)`, and let codegen ignore the `true`.
The interpreter is modified to conform to this behaviour.
Fixes#35310
Change src/etc/vscode_settings.json to always treat ./library as the sysroot source
See
https://rust-lang.zulipchat.com/#narrow/stream/185405-t-compiler.2Frust-analyzer/topic/False.20error.20report.20for.20.60rust-analyzer.28private-field.29.60 for further discussion; previously this had various bugs.
I tested go-to-definition on:
- `use std::io::Write` in `src/bootstrap/setup.rs`
- `use std::cell::RefCell` in `src/librustdoc/core.rs`
- `use rustc_span::symbol::sym` in `src/librustdoc/core.rs`
- `use std::fmt` in `compiler/rustc_span/src/symbol.rs`
- `Global` in `library/alloc/src/alloc/tests.rs`
The following things still don't work:
- `Global.deallocate` in alloc/tests.rs. This function is under `cfg(not(test))`, so it can't be enabled without disabling RA in `tests.rs` altogether. I think this might be fixable by moving `library/alloc/src/alloc/tests.rs` to `library/alloc/tests/alloc/lib.rs`, so it's in a different crate, but I'd like to avoid blocking this improvement on that change.
cc `@thomcc` `@BoxyUwU` `@spastorino` - you've had issues with RA in the past, does this fix them? Are there any other use cases I should test? You can try these changes out by running `cp src/etc/vscode_settings.json .vscode/settings.json`, or running `x setup` and picking a random profile (it won't overwrite config.toml if it already exists). See https://github.com/rust-lang/rust/pull/108135 for plans to make updating the config easier.
r? `@Veykril`
Correctly handle links starting with whitespace
Part of https://github.com/rust-lang/rust/issues/107995.
I just got this issue, wrote a fix and then saw the issue. So here's the PR. ^^'
r? `@petrochenkov`
Fix json reexports of different items with same name
Fixes #107677.
I renamed `from_item_id*` functions into `id_from_item` instead because it makes more sense now. I also simplified the logic around it a bit so that the `ids` function will now directly pass `&clean::Item` to `id_from_item` and the ID will be consistently generated (it caused an issue when I updated the ID for imports).
So now, the big change of this PR: I changed how imports' ID is generated: it now includes the target item's ID at the end of the ID. It's to prevent two reexported items with the same name (but different types).
r? `@aDotInTheVoid`
rustdoc: simplify DOM for `.item-table`
This switches from using `<div>` to the more semantic `<ul>`, and using class names that rhyme with the classes the search results table uses.
Update cargo
10 commits in 39c13e67a5962466cc7253d41bc1099bbcb224c3..17b3d0de0897e1c6b8ca347bd39f850bb0a5b9f6 2023-02-12 02:01:08 +0000 to 2023-02-17 19:45:09 +0000
- fix: unsupported protocol error on old macos version (rust-lang/cargo#11733)
- Error on invalid alphanumeric token for crates.io (rust-lang/cargo#11600)
- Add clippy lints (rust-lang/cargo#11722)
- chore: Make dependencies alphabetical order (rust-lang/cargo#11719)
- chore: bump mdbook to 0.4.27 (rust-lang/cargo#11716)
- Amend `mdman` tests. (rust-lang/cargo#11715)
- Run CI for macOS on nightly (rust-lang/cargo#11712)
- doc: doc comments and intra-doc links for `core::compiler` (rust-lang/cargo#11711)
- Ensure em dashes are recognizable in markup (rust-lang/cargo#11646)
- Set CARGO_BIN_NAME environment variable also for binary examples (rust-lang/cargo#11705)
r? `@ghost`
Add sanitizer support for modern iOS platforms
asan and tsan generally support iOS, but that previously wasn't configured in rust. This only adds support for the simulator architectures, and arm64 device architecture, not the older 32 bit architectures.
Add `kernel-address` sanitizer support for freestanding targets
This PR adds support for KASan (kernel address sanitizer) instrumentation in freestanding targets. I included the minimal set of `x86_64-unknown-none`, `riscv64{imac, gc}-unknown-none-elf`, and `aarch64-unknown-none` but there's likely other targets it can be added to. (`linux_kernel_base.rs`?) KASan uses the address sanitizer attributes but has the `CompileKernel` parameter set to `true` in the pass creation.
See
https://rust-lang.zulipchat.com/#narrow/stream/185405-t-compiler.2Frust-analyzer/topic/False.20error.20report.20for.20.60rust-analyzer.28private-field.29.60
for further discussion; previously this had various bugs.
I tested go-to-definition on:
- `use std::io::Write` in `src/bootstrap/setup.rs`
- `use std::cell::RefCell` in `src/librustdoc/core.rs`
- `use rustc_span::symbol::sym` in `src/librustdoc/core.rs`
- `use std::fmt` in `compiler/rustc_span/src/symbol.rs`
- `Global` in `library/alloc/src/alloc/tests.rs`
The following things still don't work:
- `Global.deallocate` in alloc/tests.rs. This function is under
`cfg(not(test))`, so it can't be enabled without disabling RA in
`tests.rs` altogether. I think this might be fixable by moving
`library/alloc/src/alloc/tests.rs` to
`library/alloc/tests/alloc/lib.rs`, so it's in a different crate, but
I'd like to avoid blocking this improvement on that change.
There are several `mk_foo`/`intern_foo` pairs, where the former takes an
iterator and the latter takes a slice. (This naming convention is bad,
but that's a fix for another PR.)
This commit changes several `mk_foo` occurrences into `intern_foo`,
avoiding the need for some `.iter()`/`.into_iter()` calls. Affected
cases:
- mk_type_list
- mk_tup
- mk_substs
- mk_const_list
