bors [Sat, 14 Jan 2023 03:04:40 +0000 (03:04 +0000)]
Auto merge of #106520 - ehuss:update-mdbook, r=Mark-Simulacrum
Update mdbook
This updates mdbook from 0.4.21 to 0.4.25. The list of changes is [here](https://github.com/rust-lang/mdBook/blob/master/CHANGELOG.md#mdbook-0425). The only user-visible changes are some changes around the theme picker, and change to the copy-to-clipboard ignoring hidden lines.
Internally there were some dependency updates and small fixes.
This also updates `clap` from 4.0.15 to 4.0.32 whose changelog is [here](https://github.com/clap-rs/clap/blob/master/CHANGELOG.md#4032---2022-12-22). This impacts tools like cargo. I don't see anything particularly noteworthy there, though there are some small user-visible changes.
Unfortunately this required adding a hack for building `rustix` with a bootstrap tool. The comment explains why. I am unable to think of some other workaround (or even a cleaner way to set the rustflag). Ideas are welcome if you can think of alternatives. I'm struggling to even think of a long-term solution, other than asking projects not to do auto-nightly feature detection.
One medium-term solution is to avoid the clap dependency for the mdbook library (which is how rustix gets pulled in). That is one of my goals for the 0.5 release of mdbook, but that probably won't happen until later this year. It would also require dropping clap from `rustbook` and using some other means to parse arguments (there's only two options, so it can probably be done manually).
Matthias Krüger [Fri, 13 Jan 2023 18:16:45 +0000 (19:16 +0100)]
Rollup merge of #106797 - FawazTirmizi:dev/issues/104284, r=bjorn3
riscv: Fix ELF header flags
The previous version added both `EF_RISCV_FLOAT_ABI_DOUBLE` and `EF_RISCV_RVC` if the "D" extension was enabled on riscv64 targets. riscv32 targets were not accounted for. This patch changes this so that:
- Only add `EF_RISCV_RVC` if the "C" extension is enabled
- Add `EF_RISCV_FLOAT_ABI_SINGLE` if the "F" extension is enabled and the "D" extension is not
- Add these ELF flags for riscv32 as well
Matthias Krüger [Fri, 13 Jan 2023 18:16:44 +0000 (19:16 +0100)]
Rollup merge of #106793 - Mark-Simulacrum:normalize-test, r=compiler-errors
Normalize test output more thoroughly
This prevents differences in local environments, which may (for example) end up with a longer backtrace with more digits in the backtrace prefix, as happened to me. While we're at it, clean more of the output up, including the exact location of the error in the compiler.
cc https://github.com/rust-lang/rust/pull/106521 which introduced this test
Matthias Krüger [Fri, 13 Jan 2023 18:16:43 +0000 (19:16 +0100)]
Rollup merge of #106678 - Veykril:proc-macro-panic-abort, r=eholk
Warn when using panic-strategy abort for proc-macro crates
See https://github.com/rust-lang/rust/issues/82320, this simply warns for now as that seems like the best step that can be immediately taken (opposed to straight up rejecting or ignoring)
Matthias Krüger [Fri, 13 Jan 2023 18:16:42 +0000 (19:16 +0100)]
Rollup merge of #106489 - jschwe:fix_linker_detection, r=petrochenkov
Fix linker detection for linker (drivers) with a version postfix (e.g. clang-12 instead of clang)
Linker (drivers) such as clang / gcc or lld often have a version postfix matching the regex "-\d+$".
Previously, linker detection did not account for the possible version postfix and the fallback value was used, which can cause linker errors due to wrong arguments.
Also remove the check for `-clang`, since there are no architecture specific variants of clang (to my knowledge).
Matthias Krüger [Fri, 13 Jan 2023 18:16:41 +0000 (19:16 +0100)]
Rollup merge of #104645 - yukiomoto:log-backtrace-option, r=oli-obk
Add log-backtrace option to show backtraces along with logging
according to #90698, I added a compiler option, `-Zlog-backtrace=filter`, where `filter` is a module name, to show backtraces for logging without rebuilding.
Linker drivers such as gcc, clang or lld often have a version postfix,
e.g clang-12. The previous logic would not account for this and would
fall back to guessing the linker flavor to be the default linker flavor
for the target, which causes linker errors when this is not the case.
By accounting for the possible version postfix and also considering
g++ and clang++, we considerably reduce the amount of times the
fallback guess has to be used.
To simplify matching check for a version postfix and match against the
linker stem without any version postfix.
In contrast to gcc, clang supports all architectures in one binary.
This means there are no variants like `aarch64-linux-gnu-clang` and
there is no need to check for `-clang` variants.
Yuki Okushi [Fri, 13 Jan 2023 07:54:22 +0000 (16:54 +0900)]
Rollup merge of #106608 - compiler-errors:missing-generics-verbose, r=estebank
Render missing generics suggestion verbosely
It's a bit easier to read like this, especially ones that are appending new generics onto an existing list, like ": `, T`" which render somewhat poorly inline.
Also don't suggest `dyn` as a type parameter to add, even if technically that's valid in edition 2015.
Fawaz [Mon, 9 Jan 2023 08:17:58 +0000 (00:17 -0800)]
riscv: Fix ELF header flags
The previous version added both `EF_RISCV_FLOAT_ABI_DOUBLE` and
`EF_RISCV_RVC` if the "D" extension was enabled on riscv64 targets.
riscv32 targets were not accounted for. This patch changes this
so that:
- Only add `EF_RISCV_RVC` if the "C" extension is enabled
- Add `EF_RISCV_FLOAT_ABI_SINGLE` if the "F" extension is enabled
and the "D" extension is not
- Add these ELF flags for riscv32 as well
Mark Rousskov [Fri, 13 Jan 2023 02:28:20 +0000 (21:28 -0500)]
Normalize test output more thoroughly
This prevents differences in local environments, which may (for example)
end up with a longer backtrace with more digits in the backtrace prefix,
as happened to me. While we're at it, clean more of the output up,
including the exact location of the error in the compiler.
bors [Fri, 13 Jan 2023 02:13:00 +0000 (02:13 +0000)]
Auto merge of #101138 - Rejyr:diagnostic-migration-rustc-lint-pt2, r=davidtwco
Migrate `rustc_lint` lint diagnostics
Part 2 of [Migrate `rustc_lint` errors to `SessionDiagnostic`](https://github.com/rust-lang/rust/pull/100776)
r? `@davidtwco`
# TODO
- [x] Refactor some lints manually implementing `DecorateLint` to use `Option<Subdiagnostic>`.
- [x] Add `#[rustc_lint_diagnostics]` to lint functions in `context.rs`.
- [x] Migrate `hidden_unicode_codepoints.rs`.
- [x] Migrate `UnsafeCode` in `builtin.rs`.
- [x] Migrate the rest of `builtin.rs`.
bors [Thu, 12 Jan 2023 22:58:14 +0000 (22:58 +0000)]
Auto merge of #106786 - JohnTitor:rollup-8f4vk8m, r=JohnTitor
Rollup of 8 pull requests
Successful merges:
- #105795 (Stabilize `abi_efiapi` feature)
- #106446 ([LSDA] Take ttype_index into account when taking unwind action)
- #106675 (Mark ZST as FFI-safe if all its fields are PhantomData)
- #106740 (Adding a hint on iterator type errors)
- #106741 (Fix reexport of `doc(hidden)` item)
- #106759 (Revert "Make nested RPITIT inherit the parent opaque's generics.")
- #106772 (Re-add mw to review rotation)
- #106778 (Exclude formatting commit from blame)
Yuki Okushi [Thu, 12 Jan 2023 20:47:21 +0000 (05:47 +0900)]
Rollup merge of #106446 - bzEq:fix-unwind-lsda, r=Amanieu
[LSDA] Take ttype_index into account when taking unwind action
If `cs_action != 0`, we should check the `ttype_index` field in action record. If `ttype_index == 0`, a clean up action is taken; otherwise catch action is taken.
This can fix unwind failure on AIX which uses LLVM's libunwind by default. IIUC, rust's LSDA is borrowed from c++ and I'm assuming itanium-cxx-abi https://itanium-cxx-abi.github.io/cxx-abi/exceptions.pdf should be followed, so the fix follows what libcxxabi does. See https://github.com/llvm/llvm-project/blob/ec48682ce9f61d056361c5095f21e930b8365661/libcxxabi/src/cxa_personality.cpp#L152 for use of `ttype_index`.
bors [Thu, 12 Jan 2023 17:36:00 +0000 (17:36 +0000)]
Auto merge of #10149 - jyn514:duplicate-sysroot, r=flip1995
Don't pass `--sysroot` twice if SYSROOT is set
This is useful for rust-lang/rust to allow setting a sysroot that's *only* for build scripts, different from the regular sysroot passed in RUSTFLAGS (since cargo doesn't apply RUSTFLAGS to build scripts or proc-macros).
That said, the exact motivation is not particularly important: this fixes a regression from
https://github.com/rust-lang/rust-clippy/pull/9881/commits/5907e9155ed7f1312d108aa2110853472da3b029#r1060215684.
Note that only RUSTFLAGS is tested in the new integration test; passing --sysroot through `clippy-driver` never worked as far as I can tell, and no one is using it, so I didn't fix it here.
Helps with https://github.com/rust-lang/rust/pull/106394.
---
changelog: other: `SYSROOT` and `--sysroot` can now be set at the same time
[#10149](https://github.com/rust-lang/rust-clippy/pull/10149)
<!-- changelog_checked -->
Philipp Krones [Thu, 12 Jan 2023 17:30:51 +0000 (18:30 +0100)]
Add cargo-clippy sysroot test
Whne SYSROOT is defined, clippy-driver will insert a --sysroot argument
when calling rustc. However, when a sysroot argument is already defined,
e.g. through RUSTFLAGS=--sysroot=... the `cargo clippy` call would
error. This tests that the sysroot argument is only passed once and that
SYSROOT is ignored in this case.
Joshua Nelson [Tue, 3 Jan 2023 18:04:28 +0000 (18:04 +0000)]
Don't pass `--sysroot` twice if SYSROOT is set
This is useful for rust-lang/rust to allow setting a sysroot that's
*only* for build scripts, different from the regular sysroot passed in
RUSTFLAGS (since cargo doesn't apply RUSTFLAGS to build scripts or
proc-macros).
That said, the exact motivation is not particularly important: this
fixes a regression from
https://github.com/rust-lang/rust-clippy/pull/9881/commits/5907e9155ed7f1312d108aa2110853472da3b029#r1060215684.
Note that only RUSTFLAGS is tested in the new integration test; passing
--sysroot through `clippy-driver` never worked as far as I can tell, and
no one is using it, so I didn't fix it here.
nils [Thu, 12 Jan 2023 14:44:51 +0000 (15:44 +0100)]
Rollup merge of #106322 - compiler-errors:CollectAllMismatches-infer-vars, r=oli-obk
Handle inference variables in `CollectAllMismatches` correctly
1. Fix #106240
2. Treat int/float type variables correctly (see `src/test/ui/iterators/invalid-iterator-chain-with-int-infer.rs`), so we can point out things like "`Iterator::Item` changed to `{integer}` here"
bors [Thu, 12 Jan 2023 13:59:30 +0000 (13:59 +0000)]
Auto merge of #105603 - oli-obk:non_repeatable_queries, r=petrochenkov
Harden the pre-tyctxt query system against accidental recomputation
While the current compiler has no issues where we `take` and then compute the query again, in https://github.com/rust-lang/rust/pull/105462 I accidentally introduced such a case.
I also took the opportunity to remove `peek_mut`, which is only ever used for `global_tcx` to then invoke `enter`. I added an `enter` method directly on the query.
bors [Thu, 12 Jan 2023 12:10:56 +0000 (12:10 +0000)]
Auto merge of #10179 - llogiq:trim-suspicious-to-owned-paths, r=xFrednet
trim paths in `suspicious_to_owned`
This continues my path trimming spree. I'm not going to add yet another changelog entry, we should have one "trim paths in some applicable lints" entry instead.
bors [Thu, 12 Jan 2023 10:46:19 +0000 (10:46 +0000)]
Auto merge of #106760 - compiler-errors:rollup-0bogyco, r=compiler-errors
Rollup of 8 pull requests
Successful merges:
- #103236 (doc: rewrite doc for signed int::{carrying_add,borrowing_sub})
- #103800 (Stabilize `::{core,std}::pin::pin!`)
- #106097 (Migrate mir_build diagnostics 2 of 3)
- #106170 (Move autoderef to `rustc_hir_analysis`)
- #106323 (Stabilize f16c_target_feature)
- #106360 (Tweak E0277 `&`-removal suggestions)
- #106524 (Label `struct/enum constructor` instead of `fn item`, mention that it should be called on type mismatch)
- #106739 (Remove `<dyn AstConv<'tcx>>::fun(c, ...)` calls in favour of `c.astconv().fun(...)`)
bors [Thu, 12 Jan 2023 07:40:32 +0000 (07:40 +0000)]
Auto merge of #106757 - matthiaskrgr:rollup-9j8830g, r=matthiaskrgr
Rollup of 10 pull requests
Successful merges:
- #106167 (Fix invalid syntax and incomplete suggestion in impl Trait parameter type suggestions for E0311)
- #106309 (Prefer non-`[type error]` candidates during selection)
- #106532 (Allow codegen to unsize `dyn*` to `dyn`)
- #106596 (Hide more of long types in E0271)
- #106638 (std tests: use __OsLocalKeyInner from realstd)
- #106676 (Test that we cannot use trait impl methods arguments as defining uses)
- #106702 (Conserve cause of `ImplDerivedObligation` in E0599)
- #106732 (rustc_llvm: replace llvm::makeArrayRef with ArrayRef constructors.)
- #106733 (Revert "warn newer available version of the x tool")
- #106748 (Clean up `OnUnimplementedFormatString::verify`)
Michael Goulet [Thu, 12 Jan 2023 06:25:48 +0000 (22:25 -0800)]
Rollup merge of #106170 - compiler-errors:autoderef-to-analysis, r=lcnr
Move autoderef to `rustc_hir_analysis`
Not sure if this is a change we actually want, but autoderef really is only (functionally) used by `rustc_hir_analysis` and `rustc_hir_typeck`, so it probably should live there.
Instead, implement a separate autoderef helper in `TypeErrCtxt` for the one use-case that goes against the ordering of the crate graph..