bors [Sun, 23 Oct 2022 11:18:11 +0000 (11:18 +0000)]
Auto merge of #2613 - RalfJung:scalar-only-field-retagging, r=RalfJung
add scalar-abi-only field retagging option
`@saethlin` has requested a Stacked Borrows field retagging mode that matches when we do in terms of emitting `noalias` for LLVM. So here you go! Types with `Scalar` and `ScalarPair` ABI are being recursively retagged, everything else is not.
Given that many types can get `ScalarPair` ABI, I don't think this actually helps to make many things sound that are unsound under full field retagging -- but it might still be useful for some experimentation.
bors [Fri, 21 Oct 2022 16:33:36 +0000 (16:33 +0000)]
Auto merge of #2611 - pfmooney:libffi, r=RalfJung
Update libffi-sys to 2.0.1
Prior version of libffi [could not be cross-compiled to illumos](https://github.com/tov/libffi-rs/pull/59) due to host-triple complications. This should fix rustup builds of miri for the illumos platform.
Patrick Mooney [Thu, 20 Oct 2022 16:48:13 +0000 (11:48 -0500)]
Update libffi-sys to 2.0.1
Prior version of libffi could not be cross-compiled to illumos due to
host-triple complications. This should fix rustup builds of miri for
the illumos platform.
bors [Fri, 21 Oct 2022 12:23:11 +0000 (12:23 +0000)]
Auto merge of #2609 - RalfJung:is_terminal, r=RalfJung
Use is_terminal to implement isatty
This means that Linux targets on Windows hosts should give a correct reply. We still don't have an implementation for Windows targets though, that requires some tracking of handles.
bors [Thu, 20 Oct 2022 21:08:28 +0000 (21:08 +0000)]
Auto merge of #103322 - matthiaskrgr:rollup-m9zgpft, r=matthiaskrgr
Rollup of 9 pull requests
Successful merges:
- #103221 (Fix `SelfVisitor::is_self_ty` ICE)
- #103230 (Clarify startup)
- #103281 (Adjust `transmute{,_copy}` to be clearer about which of `T` and `U` is input vs output)
- #103288 (Fixed docs typo in `library/std/src/time.rs`)
- #103296 (+/- shortcut now only expand/collapse, not both)
- #103297 (fix typo)
- #103313 (Don't label `src/test` tests as `A-testsuite`)
- #103315 (interpret: remove an incorrect assertion)
- #103319 (Improve "`~const` is not allowed here" message)
Matthias Krüger [Thu, 20 Oct 2022 20:42:41 +0000 (22:42 +0200)]
Rollup merge of #103315 - RalfJung:interpret-switchint-ice, r=bjorn3
interpret: remove an incorrect assertion
This fixes an ICE in Miri, [reported](https://rust-lang.zulipchat.com/#narrow/stream/269128-miri/topic/SwitchInt.20with.20no.20targets.3F) by `@saethlin.` The faulty assertion was introduced by https://github.com/rust-lang/rust/commit/432535da2b5144d467056efcfa6864d35ba2de0f, when a previously correct assertion checking that the `otherwise` target exists got replaced by this assertion checking that at least one more target beyond `otherwise` exists.
Sadly we don't have a small reproducer so I don't think we can easily add a testcase.
Matthias Krüger [Thu, 20 Oct 2022 20:42:39 +0000 (22:42 +0200)]
Rollup merge of #103288 - johnmatthiggins:master, r=thomcc
Fixed docs typo in `library/std/src/time.rs`
* Changed comment from `Previous rust versions panicked when self was earlier than the current time.` to `Previous rust versions panicked when the current time was earlier than self.`
* Resolves #103282.
bors [Thu, 20 Oct 2022 16:20:50 +0000 (16:20 +0000)]
Auto merge of #103092 - petrochenkov:weaklto, r=wesleywiser
linker: Fix weak lang item linking with combination windows-gnu + LLD + LTO
In https://github.com/rust-lang/rust/pull/100404 this logic was originally disabled for MSVC due to issues with LTO, but the same issues appear on windows-gnu with LLD because that LLD uses the same underlying logic as MSVC LLD, just with re-syntaxed command line options.
So this PR just disables it for LTO builds in general.
bors [Thu, 20 Oct 2022 11:09:41 +0000 (11:09 +0000)]
Auto merge of #2606 - cbiffle:nostdio, r=RalfJung
Magic functions for writing to stdout/stderr.
This enables I/O in no_std contexts (or, really, any Miri-specific OS-independent context). Combined with the `abort` intrinsic it should allow a reasonable test framework in no_std.
**Question for maintainers:** So, the `no_std` panic test needs work, for two reasons:
- First, its stdout includes Miri's whole message about the abort intrinsic having been used. I guess whatever panic handler you use in `std` contexts exits cleanly without triggering this message. Comparing the entire output with backtrace as golden seems fragile.
- Second, likely for the same reason, the test framework appears to expect the test to exit successfully, when in fact it exits with status 1 due to the abort. This means the test doesn't actually pass right now.
bors [Thu, 20 Oct 2022 10:40:09 +0000 (10:40 +0000)]
Auto merge of #103185 - chenyukang:yukang/fix-span-next-point, r=davidtwco
Fix the bug of next_point in source_map
There is a bug in `next_point`, the new span won't move to next position when be called in the first time.
For this reason, our current code is working like this:
1. When we really want to move to the next position, we called two times of `next_point`
2. Some code which use `next_point` actually done the same thing with `shrink_to_hi`
This fix make sure when `next_point` is called, span will move with the width at least 1, and also work correctly in the scenario of multiple bytes.
As mentioned in the issue, https://github.com/rust-lang/rust/pull/102924 introduced this regression. This PR partially reverts it and adds a regression test.
Matthias Krüger [Thu, 20 Oct 2022 05:58:55 +0000 (07:58 +0200)]
Rollup merge of #103197 - est31:stabilize_proc_macro_source_text, r=petrochenkov
Stabilize proc_macro::Span::source_text
Splits `proc_macro::Span::source_text` into a new feature gate and stabilizes it. The [FCP is complete](https://github.com/rust-lang/rust/issues/101991#issuecomment-1279393265).
bors [Thu, 20 Oct 2022 05:56:56 +0000 (05:56 +0000)]
Auto merge of #103220 - compiler-errors:deny-infers, r=lcnr
Deny hashing ty/re/ct inference variables
cc `@cjgillot` and https://github.com/rust-lang/rust/pull/102695#issuecomment-1275706528
r? `@lcnr`
best reviewed one commit at a time, mostly because the second commit that fixes `ClosureOutlivesRequirement` is mostly noise because of losing its `<'tcx>` lifetime parameter.
Cliff L. Biffle [Wed, 19 Oct 2022 20:16:29 +0000 (13:16 -0700)]
Magic functions for writing to stdout/stderr.
This enables I/O in no_std contexts (or, really, any Miri-specific
OS-independent context). Combined with the `abort` intrinsic it should
allow a reasonable test framework in no_std.
bors [Wed, 19 Oct 2022 21:17:09 +0000 (21:17 +0000)]
Auto merge of #103264 - matthiaskrgr:rollup-3ja4spo, r=matthiaskrgr
Rollup of 7 pull requests
Successful merges:
- #103211 (rustdoc: remove class name `location` from sidebar sibling nav)
- #103223 (Use already checked RHS ty for LHS deref suggestions)
- #103237 (Clean up codeblock-tooltip rustdoc-gui test)
- #103239 (Allow #[unstable] impls for fn() with unstable abi.)
- #103246 (Mark `rust-analyzer` as a host-only tool)
- #103257 (rustdoc: move `setting-line` color CSS to settings.css)
- #103258 (Make miri read_dir test a little more robust)
Matthias Krüger [Wed, 19 Oct 2022 19:38:41 +0000 (21:38 +0200)]
Rollup merge of #103246 - ferrocene:pa-rust-analyzer-hosts, r=Mark-Simulacrum
Mark `rust-analyzer` as a host-only tool
All tools meant to be shipped with host toolchains only should be marked as `ONLY_HOSTS = true`, but rust-analyzer was marked as `ONLY_HOSTS = false` incorrectly. This meant that bootstrap attempted to build rust-analyzer for cross-compilation-only targets, causing errors because libstd is not present on some of them.
It will still be possible to cross-compile rust-analyzer by passing a different `--host` flag to `./x`, like you can cross-compile other tools.
Matthias Krüger [Wed, 19 Oct 2022 19:38:40 +0000 (21:38 +0200)]
Rollup merge of #103223 - compiler-errors:deref-sugg-slow, r=wesleywiser
Use already checked RHS ty for LHS deref suggestions
There's no reason to do the `check_lhs_assignable` and RHS `check_expr_with_hint` in that order, so invert them and use the typeck results to avoid exponential blowup on error.
This change tweaks the CSS to apply most of its styles to `.sidebar h2`, cleaning up a few redundant rules from `.mobile-topbar .location` and restoring useful navigation aids in mobile mode.
`@RalfJung` I prefer this approach because the whole GC system is sloppy already in order to be efficient (doesn't run often, ignores small stacks) so a bit more imprecision for a simple implementation seems worth it to me. But I'm of course willing to be convinced otherwise.
bors [Wed, 19 Oct 2022 13:53:06 +0000 (13:53 +0000)]
Auto merge of #103214 - Nilstrieb:set-theory, r=petrochenkov
Use Set instead of Vec in transitive_relation
Helps with #103195. It doesn't fix the underlying quadraticness but it makes it _a lot_ faster to an extent where even doubling the amount of nested references still takes less than two seconds (50s on nightly).
I want to see whether this causes regressions (because the vec was usually quite small) or improvements (as lookup for bigger sets is now much faster) in real code.
bors [Wed, 19 Oct 2022 13:10:32 +0000 (13:10 +0000)]
Auto merge of #2604 - RalfJung:ret-scalar, r=RalfJung
use Scalar return types for Windows shims
I started doing this while working on https://github.com/rust-lang/miri/issues/2595; now I don't need the rest of this patch but this part still makes sense.
Pietro Albini [Wed, 19 Oct 2022 12:33:17 +0000 (14:33 +0200)]
mark rust-analyzer as a host-only tool
All tools meant to be shipped with host toolchains only should be marked
as `ONLY_HOSTS = true`, but rust-analyzer was marked as `ONLY_HOSTS =
false` incorrectly. This meant that bootstrap attempted to build
rust-analyzer for cross-compilation-only targets, causing errors because
libstd is not present on some of them.
It will still be possible to cross-compile rust-analyzer by passing a
different --host flag to ./x, like you can cross-compile other tools.
bors [Wed, 19 Oct 2022 11:12:10 +0000 (11:12 +0000)]
Auto merge of #103228 - Dylan-DPC:rollup-31yiauw, r=Dylan-DPC
Rollup of 6 pull requests
Successful merges:
- #102863 (Standardize "use parentheses to call" suggestions between typeck and trait selection)
- #103034 (Let expressions on RHS shouldn't be terminating scopes)
- #103127 (Make transpose const and inline)
- #103153 (Allow `Vec::leak` when using `no_global_oom_handling`)
- #103182 (Clean up query descriptions)
- #103216 (Consider patterns in fn params in an `Elided(Infer)` lifetime rib.)