Repository navigation
Rollup of 8 pull requests - #165808
Rollup of 8 pull requests#165808
Conversation
Co-authored-by: alexey semenyuk <alexsemenyuk88@gmail.com>
Co-authored-by: Kevin Reid <kpreid@switchb.org> Co-authored-by: 许杰友 Jieyou Xu (Joe) <39484203+jieyouxu@users.noreply.github.com> Co-authored-by: Tim (Theemathas) Chirananthavat <theemathas@gmail.com>
which was renamed and started taking an LibraryOptionsContext argument in llvm/llvm-project#239919
Co-authored-by: Josh Stone <cuviper@gmail.com>
"appendex" -> "appendix". Signed-off-by: Hardik Kumar <hardikxk@gmail.com>
Co-authored-by: Josh Stone <cuviper@gmail.com>
See <rust-lang#165796>. This debuginfo test is flaky for ARM64 MSVC.
Co-authored-by: León Orell Valerian Liehr <me@fmease.dev>
Add relnotes for 1.100 r? cuviper cc @rust-lang/release
…,fmease Support underscore as associated const name in inherent impls This PR doesn't change the current behavior of the `dead_code` lint for unused associated consts with underscore-prefixed names and treats associated consts named `_` consistently. There might still be some places that should've been changed to use `named_items_in_definition_order()` instead of `in_definition_order()`, though I'm not quite sure. [The RFC](https://github.com/dtolnay/rfcs/blob/assocunderscore/text/0000-associated-const-underscore.md#reference-level-explanation) does not propose const underscore inside traits or trait impls, so this PR only implements support for const underscore in inherent impls. Unresolved questions: - Whether to lint `impl Private { pub const _: Private = Private; }`, where the visibility qualifier here has no effect. It seems like there wasn't any disagreement on linting this, though what's proposed in the RFC currently diverges from that: [RFC: Associated const underscore](https://github.com/dtolnay/rfcs/blob/assocunderscore/text/0000-associated-const-underscore.md#reference-level-explanation). Tracking: - rust-lang#158944
…ouwer,GuillaumeGomez Properly feature gate doc(rust_logo) fixes rust-lang#161984 It's a breaking change, technically. I didn't find it in the crater rollup, but will rerun just in case.
…nFlags, r=nikic Update call to llvm::codegen::InitTargetOptionsFromCodeGenFlags which was renamed and started taking a LibraryOptionsContext argument in llvm/llvm-project#239919
Bail out ill-formed aliases when assembling trait object bounds candidates Fixes rust-lang#155761 r? adwinwhite
…r=tshepang docs: Fix a typo in rustc-dev-guide [NFC] "appendex" -> "appendix".
…cdb-basic-stepping, r=Enselic Disable flaky `tests/debuginfo/basic-stepping.rs` on aarch64 msvc See <rust-lang#165796>. This debuginfo test is flaky for aarch64 msvc. ### History - Test added in rust-lang#153840 - Removed `ignore-aarch64` in rust-lang#153877 - Seems that we started getting failures soon after, e.g. rust-lang#153935 (comment) - rust-lang#153939 also wanted to add `ignore-aarch64-pc-windows-msvc` but was closed.
…ercote Add a regression test for fixed next-solver ICE Closes rust-lang#155035, which has been fixed already(cc rust-lang#155035 (comment))
|
@bors r+ p=5 force |
This comment has been minimized.
This comment has been minimized.
|
The job Click to see the possible cause of the failure (guessed by this bot) |
What is this?This is an experimental post-merge analysis report that shows differences in test outcomes between the merged PR and its parent PR.Comparing 9de7469 (parent) -> e88737b (this PR) Test differencesShow 66 test diffsStage 1
Stage 2
Additionally, 34 doctest diffs were found. These are ignored, as they are noisy. Job group index
Test dashboardRun cargo run --manifest-path src/ci/citool/Cargo.toml -- \
test-dashboard e88737b98536eae309f98e207b8af6a2cafa0350 --output-dir test-dashboardAnd then open Job duration changes
How to interpret the job duration changes?Job durations can vary a lot, based on the actual runner instance |
|
📌 Perf builds for each rolled up PR:
parent commit: 9de7469383 In the case of a perf regression, run the following command with the SHAs of each PR you suspect might be the cause: |
|
Finished benchmarking commit (e88737b): comparison URL. Overall result: no relevant changes - no action needed@rustbot label: -perf-regression Instruction countThis perf run didn't have relevant results for this metric. Max RSS (memory usage)Results (primary 1.3%, secondary -0.5%)A less reliable metric. May be of interest, but not used to determine the overall result above.
CyclesResults (secondary 2.8%)A less reliable metric. May be of interest, but not used to determine the overall result above.
Binary sizeThis perf run didn't have relevant results for this metric. Bootstrap: 483.908s -> 484.585s (0.14%) |
…esleywiser Adjust `unsized.rs` debuginfo test expected output for cdb `10.0.28000.2957` This was a cdb test that was failing locally when I updated Windows SDK to `10.0.28000.2957` (and thus cdb and WinDbg). Currently our Windows CI seems to use Windows SDKs around `10.0.26100.*`. See example run: - rust-lang#165808 - => https://github.com/rust-lang/rust/actions/runs/43010199577 - => https://github.com/rust-lang/rust/actions/runs/43010199577/job/130488011527 - => Runner image `windows-2025-vs2026 20261103.298.1` - => https://github.com/actions/runner-images/blob/win25-vs2026/20261103.298/images/windows/Windows2025-VS2026-Readme.md - => `10.0.26100.33674` Seems that for whatever reason eventual visualizations for `c` and `_box` changed between these cdb (WinDbg) versions. For ```rs let c: &Foo<dyn std::fmt::Debug> = &Foo { value: 7i32 }; ``` the `10.0.26100.*` series seem to output e.g. ``` //@ cdb-check:c [Type: ref$<unsized::Foo<dyn$<core::fmt::Debug> > >] //@ cdb-check: [+0x000] pointer : 0x[...] [Type: unsized::Foo<dyn$<core::fmt::Debug> > *] //@ cdb-check: [...] vtable : 0x[...] [Type: unsigned [...]int[...] (*)[4]] ``` but `10.0.28000.2957` output changed to a single line e.g. ``` c : 7 [Type: i32 (derived from ref$<unsized::Foo<dyn$<core::fmt::Debug> > >)] ``` So likely one of: - Default visualizers got updated - WinDbg internals changed - cdb <-> WinDbg engine interactions changed I cannot find cdb/WinDbg changelogs anywhere for these specific Windows SDK versions, so the best I can do is use `10.0.28000.2957` output as new expected output and gate this test on a minimum cdb version of `10.0.28000.2957`. This of course means that, before our GHA windows runner images update to a Windows SDK that's at least `10.0.28000.2957` (and thus cdb since they're distributed and versioned together), this test will not run in CI.
Rollup merge of #165815 - jieyouxu:jieyouxu/fix-unsized, r=wesleywiser Adjust `unsized.rs` debuginfo test expected output for cdb `10.0.28000.2957` This was a cdb test that was failing locally when I updated Windows SDK to `10.0.28000.2957` (and thus cdb and WinDbg). Currently our Windows CI seems to use Windows SDKs around `10.0.26100.*`. See example run: - #165808 - => https://github.com/rust-lang/rust/actions/runs/43010199577 - => https://github.com/rust-lang/rust/actions/runs/43010199577/job/130488011527 - => Runner image `windows-2025-vs2026 20261103.298.1` - => https://github.com/actions/runner-images/blob/win25-vs2026/20261103.298/images/windows/Windows2025-VS2026-Readme.md - => `10.0.26100.33674` Seems that for whatever reason eventual visualizations for `c` and `_box` changed between these cdb (WinDbg) versions. For ```rs let c: &Foo<dyn std::fmt::Debug> = &Foo { value: 7i32 }; ``` the `10.0.26100.*` series seem to output e.g. ``` //@ cdb-check:c [Type: ref$<unsized::Foo<dyn$<core::fmt::Debug> > >] //@ cdb-check: [+0x000] pointer : 0x[...] [Type: unsized::Foo<dyn$<core::fmt::Debug> > *] //@ cdb-check: [...] vtable : 0x[...] [Type: unsigned [...]int[...] (*)[4]] ``` but `10.0.28000.2957` output changed to a single line e.g. ``` c : 7 [Type: i32 (derived from ref$<unsized::Foo<dyn$<core::fmt::Debug> > >)] ``` So likely one of: - Default visualizers got updated - WinDbg internals changed - cdb <-> WinDbg engine interactions changed I cannot find cdb/WinDbg changelogs anywhere for these specific Windows SDK versions, so the best I can do is use `10.0.28000.2957` output as new expected output and gate this test on a minimum cdb version of `10.0.28000.2957`. This of course means that, before our GHA windows runner images update to a Windows SDK that's at least `10.0.28000.2957` (and thus cdb since they're distributed and versioned together), this test will not run in CI.
Successful merges:
tests/debuginfo/basic-stepping.rson aarch64 msvc #165797 (Disable flakytests/debuginfo/basic-stepping.rson aarch64 msvc)r? @ghost
Create a similar rollup