Repository navigation
Rollup of 13 pull requests - #165582
Rollup of 13 pull requests#165582
Conversation
This phase contains a lot of checks, not only borrowck
Local runs don't emit metrics, trying to update the metrics.json file will always fail.
I became a bit lost when I encountered these links, for example at: <https://doc.rust-lang.org/nightly/nightly-rustc/rustc_errors/diagnostic/struct.Diag.html#method.span_label> It appears that these were caused by commit f6f8779 ("Reduce capabilities of `Diagnostic`."). Formerly, the docs for with_foo and foo would both link to the docs for foo on an "impl" struct, but that commit moved those methods over to the outer struct and changed the links to point to the original method there, for both the original and with_ variants. This *is* the appropriate target ... except you don't need to go see it when you're already seeing it. It's a bit of a big refactoring moving a bunch of code, so it is understandable that such a small detail wasn't really noticed with all of those method bodies flying around.
right now, an opt-dist run starts with: - clear its opt-artifacts folder - copy a rustc-perf tree there - build the collector So, on every run, the built rustc-perf is deleted and rebuilt, even though there is a "prebuilt rustc-perf" configuration. This commit does not rebuild rustc-perf if there's a prebuilt dir, and returning that as the rustc-perf path to run benchmarks in.
Local runs are usually more concerned about optimizing the compiler, so optimizing rustdoc is off by default for these, but it still is on by default otherwise, to keep optimizing it on CI as we do today.
Make optimizing cranelift and clippy off by default for local runs.
The default is still on for CI, because we optimize by default, but off for local runs.
At least for now, only `lto=fat` is supported
…, r=JohnTitor Specialize more builtin types for `<[MaybeUninit<_>]>::write_default()` r? libs
…bzol Use the `2.38` binutils version in CI Blocked on rust-lang/ci-mirrors#61. r? @Kobzol
…rs, r=notriddle rustdoc: Make implementors collapseable if there is more than one line of docs Fixes rust-lang#165383. Apply the same logic as for non-implementors impl blocks: if there is an empty impl block only containing more than one line of docs, it should be collapseable. r? @notriddle
|
Rollup of everything. @bors r+ rollup=never p=5 force |
This comment has been minimized.
This comment has been minimized.
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 a267dc9 (parent) -> e30d778 (this PR) Test differencesShow 478 test diffsStage 1
Stage 2
Additionally, 456 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 e30d778f3029b78bb9c2e0dbe11a3b7389b652c4 --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 |
|
Finished benchmarking commit (e30d778): comparison URL. Overall result: ✅ improvements - no action needed@rustbot label: -perf-regression Instruction countOur most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.
Max RSS (memory usage)Results (secondary 6.1%)A less reliable metric. May be of interest, but not used to determine the overall result above.
CyclesThis perf run didn't have relevant results for this metric. Binary sizeResults (primary -0.1%)A less reliable metric. May be of interest, but not used to determine the overall result above.
Bootstrap: 483.513s -> 484.651s (0.24%) |
|
📌 Perf builds for each rolled up PR:
parent commit: a267dc9278 In the case of a perf regression, run the following command with the SHAs of each PR you suspect might be the cause: |
Successful merges:
<[MaybeUninit<_>]>::write_default()#164914 (Specialize more builtin types for<[MaybeUninit<_>]>::write_default())2.38binutils version in CI #165078 (Use the2.38binutils version in CI)ExitCodeinstead of callingexit#165444 (Suggest usingExitCodeinstead of callingexit)std::sysmodules #165379 (Autolabelstd::sysmodules)str::split_at_[mut_]uncheckedandOsStr::split_at_unchecked#165546 (Exposestr::split_at_[mut_]uncheckedandOsStr::split_at_unchecked)-Zvirtual-function-eliminationerror message #165570 (fix-Zvirtual-function-eliminationerror message)r? @ghost
Create a similar rollup