interpret: make typed copies lossy wrt provenance and padding
A "typed copy" in Rust can be a lossy process: when copying at type `usize` (or any other non-pointer type), if the original memory had any provenance, that provenance is lost. When copying at pointer type, if the original memory had partial provenance (i.e., not the same provenance for all bytes), that provenance is lost. When copying any type with padding, the contents of padding are lost.
This PR equips our validity-checking pass with the ability to reset provenance and padding according to those rules. Can be reviewed commit-by-commit. The first three commits are just preparation without any functional change.
Fixes https://github.com/rust-lang/miri/issues/845
Fixes https://github.com/rust-lang/miri/issues/2182
rustdoc: use strategic boxing to shrink `clean::Item`
* `inline_stmt_id` is never a cross-crate DefId, so save space by not storing it.
* Instead of two inner boxes for `Item`, use one.
Rollup of 9 pull requests
Successful merges:
- #129929 (`rustc_mir_transform` cleanups, round 2)
- #130022 (Dataflow/borrowck lifetime cleanups)
- #130064 (fix ICE in CMSE type validation)
- #130067 (Remove redundant check in `symlink_hard_link` test)
- #130131 (Print a helpful message if any tests were skipped for being up-to-date)
- #130137 (Fix ICE caused by missing span in a region error)
- #130153 (use verbose flag as a default value for `rust.verbose-tests`)
- #130154 (Stabilize `char::MIN`)
- #130158 (Update books)
r? `@ghost`
`@rustbot` modify labels: rollup
Update books
## rust-lang/edition-guide
1 commits in eeba2cb9c37ab74118a4fb5e5233f7397e4a91f8..b3ca7ade0f87d7e3fb538776defc5b2cc4188172
2024-09-03 22:59:56 UTC to 2024-09-03 22:59:56 UTC
- Rename `static_mut_refs` file to plural
## rust-embedded/book
1 commits in ff5d61d56f11e1986bfa9652c6aff7731576c37d..dbae36bf3f8410aa4313b3bad42e374735d48a9d
2024-09-04 20:49:11 UTC to 2024-09-04 20:49:11 UTC
- Fix small error in c bindings (rust-embedded/book#377)
## rust-lang/reference
6 commits in 0668397076da350c404dadcf07b6cbc433ad3743..687faf9958c52116d003b41dfd29cc1cf44f5311
2024-09-03 20:21:23 UTC to 2024-08-29 01:11:29 UTC
- `asm!`: clarify that `nomem` / `readonly` can access private memory (rust-lang/reference#1598)
- Switch warning blocks to use admonitions (rust-lang/reference#1595)
- dont use stdcall on x86_64 where it is not a valid ABI (rust-lang/reference#1600)
- Add capturing and precise capturing rules (rust-lang/reference#1577)
- Add a prefix to rule HTML IDs (rust-lang/reference#1593)
- Add a description of rule identifiers (rust-lang/reference#1594)
## rust-lang/rust-by-example
2 commits in 859786c5bc99301bbc22fc631a5c2b341860da08..c79ec345f08a1e94494cdc8c999709a90203fd88
2024-09-08 13:24:48 UTC to 2024-09-08 13:24:08 UTC
- Fix typo in conversion/from_into (rust-lang/rust-by-example#1882)
- Corrected x values for example rectangle (rust-lang/rust-by-example#1884)
## rust-lang/rustc-dev-guide
9 commits in fa928a6d19e1666d8d811dfe3fd35cdad3b4e459..0ed9229f5b6f7824b333beabd7e3d5ba4b9bd971
2024-09-09 10:58:45 UTC to 2024-08-31 13:02:08 UTC
- fix some comments (rust-lang/rustc-dev-guide#2059)
- Fix a link and create an implied internal link (rust-lang/rustc-dev-guide#2051)
- link to additional LLVM update PRs (rust-lang/rustc-dev-guide#2055)
- fix/improve rdg contributing howto (rust-lang/rustc-dev-guide#2056)
- fixed the typo (rust-lang/rustc-dev-guide#2058)
- fix typo maker -> marker (thanks `@lholten)` (rust-lang/rustc-dev-guide#2057)
- improve query system documentation (rust-lang/rustc-dev-guide#2041)
- add section on overlap checks (rust-lang/rustc-dev-guide#2042)
- clarify the role of rustc_const_unstable (rust-lang/rustc-dev-guide#2045)
Print a helpful message if any tests were skipped for being up-to-date
When running tests without the `--force-rerun` flag, compiletest will automatically skip any tests that (in its judgement) don't need to be run again since the last time they were run.
This is normally very useful, but can occasionally be confusing, especially in edge-cases where up-to-date checking is not completely accurate (or the test is flaky).
This PR makes bootstrap count the number of tests that were ignored for being up-to-date (via a hard-coded check on the ignore reason), and prints a helpful message when that number is nonzero.
---
Sample output:
```text
test result: ok. 4 passed; 0 failed; 17578 ignored; 0 measured; 0 filtered out; finished in 463.79ms
help: ignored 17295 up-to-date tests; use `--force-rerun` to prevent this
Build completed successfully in 0:00:07
```
Remove redundant check in `symlink_hard_link` test
We support macOS 10.12 and above, so it now always uses `linkat`, and so the check is redundant.
This was missed in #126351.
``@rustbot`` label O-macos
Distribute rustc_codegen_cranelift for Windows
With support for raw-dylib recently added to cg_clif, and inline assembly support working on Windows for quite a while now, all blockers for distributing cg_clif on Windows that I mentioned in https://github.com/rust-lang/rust/pull/81746#issuecomment-1774099637 are fixed now.
const: make ptr.is_null() stop execution on ambiguity
This seems better than saying `false` -- saying `false` is in fact actively unsound if `NonNull` then uses this to permit putting this pointer inside of it, but at runtime it turns out to be null.
Part of https://github.com/rust-lang/rust/issues/74939
Cc ```@rust-lang/wg-const-eval```
remove 'const' from 'Option::iter'
This is kind of pointless to be a `const fn` since you can't do anything with the iterator. It is also the only `const fn iter*` in the entire standard library. It probably got constified when `~const` traits got added everywhere, and then was forgotten to be de-constified when that was undone.
The rest of the const_option feature seems like it can reasonably be stabilized, but this one IMO should not be stabilized, and it's not worth creating a new tracking issue.
Cc https://github.com/rust-lang/rust/issues/67441
run-make-support: Add llvm-pdbutil
Add llvm-pdbutil to run-make-support, so we can write better unit tests for PDB specific features.
r? ````@jieyouxu````
````@rustbot```` label: +O-windows
Use sysroot crates maximally in `rustc_codegen_gcc`.
This shrinks `compiler/rustc_codegen_gcc/Cargo.lock` quite a bit. The only remaining dependencies in `compiler/rustc_codegen_gcc/Cargo.lock` are `gccjit`, `lang_tester`, and `boml`, all of which aren't used in any other compiler crates.
The commit also reorders and adds comments to the `extern crate` items so they match those in miri.
r? ```@Mark-Simulacrum```
rustdoc: normalise type/field names
Updates #100961
- `Import` -> `Use`, to better reflect the terminology of Rust & its syntax
- `TypeBinding` -> `AssocItemConstraint`, to sync up with `clean`
- `FnDecl` -> `FunctionSignature`, because that's what it is
- `Header` -> `FunctionHeader`, because `Header` is a very word that's very heavily loaded with different meanings
- `ItemEnum::AssocType`: `default` -> `type`, because those items appear in `impl` blocks as well, where they're _not_ the "default"
- `ItemEnum::AssocConst`: `default` -> `value`, see the previous point
- `ForeignType` -> `ExternType`, because "foreign" is not the right word there
- boolean fields' names made to consistently be a phrase that can be a yes/no answer, e.g. `async` -> `is_async`
The docs of `ItemEnum::AssocType::type_` & of `ItemEnum::AssocConst::value` are also updated to be up to date with the clarification of the name of the fields
detect incompatible CI LLVM options more precisely
Previously, the logic here was simply checking whether the option was set in `config.toml`. This approach was not manageable in our CI runners as we set so many options in config.toml. In reality, those values are not incompatible since they are usually the same value used to generate the CI llvm. Now, the new logic compares the configuration values with the values used to generate the CI llvm, so we get more precise results and make the process more manageable.
Fixes#129153
It's a very thin wrapper that pairs `MoveDataBuilder` with a `Location`,
and it has four lifetime arguments. This commit removes it by just
adding a `Location` to `MoveDataBuilder`.
There are four related dataflow structs: `MaybeInitializedPlaces`,
`MaybeUninitializedPlaces`, and `EverInitializedPlaces`,
`DefinitelyInitializedPlaces`. They all have a `&Body` and a
`&MoveData<'tcx>` field. The first three use different lifetimes for the
two fields, but the last one uses the same lifetime for both.
This commit changes the first three to use the same lifetime, removing
the need for one of the lifetimes. Other structs that also lose a
lifetime as a result of this are `LivenessContext`, `LivenessResults`,
`InitializationData`.
It then does similar things in various other structs.
