alignment of `byval` on x86 in the process.
Commit 88e4d2c291 from five years ago removed
support for alignment on indirectly-passed arguments because of problems with
the `i686-pc-windows-msvc` target. Unfortunately, the `memcpy` optimizations I
recently added to LLVM 16 depend on this to forward `memcpy`s. This commit
attempts to fix the problems with `byval` parameters on that target and now
correctly adds the `align` attribute.
The problem is summarized in [this comment] by @eddyb. Briefly, 32-bit x86 has
special alignment rules for `byval` parameters: for the most part, their
alignment is forced to 4. This is not well-documented anywhere but in the Clang
source. I looked at the logic in Clang `TargetInfo.cpp` and tried to replicate
it here. The relevant methods in that file are
`X86_32ABIInfo::getIndirectResult()` and
`X86_32ABIInfo::getTypeStackAlignInBytes()`. The `align` parameter attribute
for `byval` parameters in LLVM must match the platform ABI, or miscompilations
will occur. Note that this doesn't use the approach suggested by eddyb, because
I felt it was overkill to store the alignment in `on_stack` when special
handling is really only needed for 32-bit x86.
As a side effect, this should fix#80127, because it will make the `align`
parameter attribute for `byval` parameters match the platform ABI on LLVM
x86-64.
[this comment]: https://github.com/rust-lang/rust/pull/80822#issuecomment-829985417
Replace RPITIT current impl with new strategy that lowers as a GAT
This PR replaces the current implementation of RPITITs with the new implementation that we had under -Zlower-impl-trait-in-trait-to-assoc-ty flag that lowers the RPIT as a GAT on the trait and on the impls that implement that trait.
Opening this PR as a draft because this goes after #112682, ~#112981~ and ~#112983~.
As soon as those are merged, I can rebase and we should run perf, crater and test a lot.
r? `@compiler-errors`
Add `Read`, `Write` and `Seek` impls for `Arc<File>` where appropriate
If `&T` implements these traits, `Arc<T>` has no reason not to do so
either. This is useful for operating system handles like `File` or
`TcpStream` which don't need a mutable reference to implement these
traits.
CC #53835.
CC #94744.
Rollup of 2 pull requests
Successful merges:
- #113331 (Add filter with following segment while lookup typo for path)
- #113524 (Remove the library/term exception in tidy's pal checker code)
r? `@ghost`
`@rustbot` modify labels: rollup
Add filter with following segment while lookup typo for path
From the discussion: https://github.com/rust-lang/rust/pull/112917#discussion_r1239150173
Seems we can not get the assoc items for `Struct`, `Enum` in the resolving phase.
A obvious filter is avoid suggesting the same name with the following segment path.
Use `following_seg` can extend the function `smart_resolve_partial_mod_path_errors` for more scenarios, such as `std::sync_error::atomic::AtomicBool` in test case.
r? `@estebank`
move pal cfgs in f32 and f64 to sys
I'd like to push forward on `sys` being a separate crate. To start with, most of these PAL exception cases are very simple little bits of code like this, so I thought I would try tidying them up.
Re-enable some coverage tests on Linux
These tests were originally disabled (on all platforms) in #110393, because those changes had made them start failing on Linux for unclear reasons.
I tried to re-enable them unconditionally in #111179, since they worked locally on my Mac, but I found that they were still failing on Linux, so I gave up at that time.
Later while working on #112300 I was able to re-enable them on Windows and Mac, since those changes made it possible to add specific `ignore-` directives to individual tests. I noticed at the time that the tests actually seemed to be working again on Linux, but by that point I didn't want to risk more CI failures, so I left them disabled on Linux with an intention to re-enable them later.
Now I'm going back to re-enable them on Linux too, since they seem to work fine.
---
Because `run-coverage` tests are sensitive to line numbers, and `x test tidy` doesn't like leading blank lines, I've replaced the old comment/ignore with an informative comment that occupies the same number of lines.
Copy stage0 `rustc` binaries to `stage0-sysroot`
This is basically a revival of https://github.com/rust-lang/rust/pull/101711 and https://github.com/rust-lang/rust/pull/107956, with an added check that the full sysroot will only be created if the original rustc comes from `stage0/bin`.
What is/should be tested:
- [x] `rustup toolchain link stage0` (new libstd is used correctly)
- [x] `python3 x.py fmt dist --stage 0`
- [x] Custom rustc/cargo in `config.toml` (in this case this logic is ignored)
- [x] Perfbot (try perf run has succeeded)
- [x] Real use case (https://github.com/rust-lang/backtrace-rs/pull/542)
(Hopefully) fixes: https://github.com/rust-lang/rust/issues/101691
This is not the "end all, be all" solution to this problem, but as long as it resolves the basic use-case, and doesn't break perfbot, I say ship it. This code will probably be nuked anyway Soon™ because of the stage redesign.
Add Tests for native wasm exceptions
### Motivation
In PR #111322, I added support for native WASM exceptions. I was asked by ``@davidtwco`` to add some tests for it in a follow up PR, which seems like a very good idea.
This PR adds three tests for this feature:
* codegen: ensure the correct LLVM instructions are used
* assembly: ensure the correct WASM instructions are used
* run-make: ensure the exception handling works; the WASM code is run using a small nodejs script which demonstrates the exception handling
### Complications
There are a few changes beside adding the tests, which were necessary
* Tests for the wasm32-unknown-unknown target are (as far as I know) only run on `test-various`. Its docker image uses nodejs-15, which is very old. Experimental support for wasm-exceptions was added in nodejs16. In nodejs 18.12 (LTS), they are stable.
- --> increase nodejs to 18.12 in `test-various`
* codegen/assembly tests are not performed for the wasm32-unknown-unknown target yet
- --> add those to `test-various` as well
Due to the last point, some tests are run which have not run before (assembly+codegen tests for wasm32-unknown-unknown). I added `// ignore wasm32-bare` for those which failed
### Local testing
I run all tests locally using both `test-various` and `wasm32`. As far as I know, none of the other systems run any test for wasm32 targets.
Always name the return place.
MIR opts more and more consider `_0` as just another local, so there is no point in keeping the special case in debug-info logic.
Update debuginfo test runner to provide more useful output
This change makes debuginfo tests more user friendly. Changes:
- Print all lines that fail to match the patterns instead of just the first
- Provide better error messages that also say what did match
- Strip leading whitespace from directives so they are not skipped if indented
- Improve documentation and improve nesting on some related items
As an example, given the following intentional fail (and a few not shown):
```rust
// from tests/debuginfo/rc_arc.rs
// cdb-command:dx rc,d
// cdb-check:rc,d : 111 [Type: alloc::rc::Rc<i32>]
// cdb-check: [Reference count] : 11 [Type: core::cell FAIL::Cell<usize>]
// cdb-check: [Weak reference count] : 2 [Type: core::cell FAIL::Cell<usize>]
```
The current output (tested in #113313) will show:
```
2023-07-04T08:10:00.1939267Z ---- [debuginfo-cdb] tests\debuginfo\rc_arc.rs stdout ----
2023-07-04T08:10:00.1942182Z
2023-07-04T08:10:00.1957463Z error: line not found in debugger output: [Reference count] : 11 [Type: core:: cell FAIL::Cell<usize>]
2023-07-04T08:10:00.1958272Z status: exit code: 0
```
With this chane, you are able to see all failures in that check group, as well as what parts were successful. The output is now:
```
2023-07-04T09:45:57.2514224Z error: check directive(s) from `C:\a\rust\rust\tests\debuginfo\rc_arc.rs` not found in debugger output. errors:
2023-07-04T09:45:57.2514631Z (rc_arc.rs:31) ` [Reference count] : 11 [Type: core::cell FAIL::Cell<usize>]`
2023-07-04T09:45:57.2514908Z (rc_arc.rs:32) ` [Weak reference count] : 2 [Type: core::cell FAIL::Cell<usize>]`
2023-07-04T09:45:57.2515181Z (rc_arc.rs:41) ` [Reference count] : 21 [Type: core::sync::atomic FAIL::AtomicUsize]`
2023-07-04T09:45:57.2515452Z (rc_arc.rs:50) `dyn_rc,d [Type: alloc::rc::Rc<dyn$<core::fmt FAIL::Debug> >]`
2023-07-04T09:45:57.2515695Z the following subset of check directive(s) was found successfully::
2023-07-04T09:45:57.2516080Z (rc_arc.rs:30) `rc,d : 111 [Type: alloc::rc::Rc<i32>]`
2023-07-04T09:45:57.2516312Z (rc_arc.rs:35) `weak_rc,d : 111 [Type: alloc::rc::Weak<i32>]`
2023-07-04T09:45:57.2516555Z (rc_arc.rs:36) ` [Reference count] : 11 [Type: core::cell::Cell<usize>]`
2023-07-04T09:45:57.2516881Z (rc_arc.rs:37) ` [Weak reference count] : 2 [Type: core::cell::Cell<usize>]`
...
```
Which makes it easier to see what did and didn't succeed without manual comparison against the source test file.
Port PGO/LTO/BOLT optimized build pipeline to Rust
This PR ports the `stage-build.py` PGO/LTO/BOLT optimization script from Python to Rust, to make it easier to use dependencies, and make it a bit more robust. The PR switches both the Linux and Windows dist runners to the Rust script and removes the old Python script.
Funnily enough, the Rust port has less lines of code than the Python script :) I think that clearly shows that the Python script really lacked dependencies.
Don't call `query_normalize` when reporting similar impls
Firstly, It's sketchy to be using `query_normalize` at all during HIR typeck -- it's asking for an ICE 😅. Secondly, we're normalizing an impl trait ref that potentially has parameter types in `ty::ParamEnv::empty()`, which is kinda sketchy as well.
The only UI test change from removing this normalization is that we don't evaluate anonymous constants in impls, which end up giving us really ugly suggestions:
```
error[E0277]: the trait bound `[X; 35]: Default` is not satisfied
--> /home/gh-compiler-errors/test.rs:4:5
|
4 | <[X; 35] as Default>::default();
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ the trait `Default` is not implemented for `[X; 35]`
|
= help: the following other types implement trait `Default`:
&[T]
&mut [T]
[T; 32]
[T; core::::array::{impl#30}::{constant#0}]
[T; core::::array::{impl#31}::{constant#0}]
[T; core::::array::{impl#32}::{constant#0}]
[T; core::::array::{impl#33}::{constant#0}]
[T; core::::array::{impl#34}::{constant#0}]
and 27 others
```
So just fold the impls with a `BottomUpFolder` that calls `ty::Const::eval`. This doesn't work totally correctly with generic-const-exprs, but it's fine for stable code, and this is error reporting after all.
Rollup of 7 pull requests
Successful merges:
- #112931 (Enable zlib in LLVM on aarch64-apple-darwin)
- #113158 (tests: unset `RUSTC_LOG_COLOR` in a test)
- #113173 (CI: include workflow name in concurrency group)
- #113335 (Reveal opaques in new solver)
- #113390 (CGU formation tweaks)
- #113399 (Structurally normalize again for byte string lit pat checking)
- #113412 (Add basic types to SMIR)
r? `@ghost`
`@rustbot` modify labels: rollup
Structurally normalize again for byte string lit pat checking
We need to structurally normalize the pointee of a match scrutinee when trying to match byte string patterns -- we used[^1] to call `structurally_resolve_type`, which errors for type vars[^2], but lcnr added `try_structurally_resolve_type`[^3] in the mean time, which is the right thing to use here since it's totally opportunistic.
Fixes rust-lang/trait-system-refactor-initiative#38
[^1]: #112428
[^2]: #112993
[^3]: #113086
Reveal opaques in new solver
We were testing against the wrong reveal mode 😨
Also a couple of misc commits that I don't want to really put in separate prs
r? ``@lcnr``
CI: include workflow name in concurrency group
Currently, this won't change anything, because we only have one relevant workflow (`CI`), but for future proofing we should probably include the workflow name in the concurrency group.
Found by ``@klensy`` [here](https://github.com/rust-lang/rust/pull/113059#discussion_r1247213606).
