| Age | Commit message (Collapse) | Author | Lines |
|---|
|
This should produce more efficient code, with fewer calls to
__tls_get_addr. The tradeoff is that libraries using it won't work with
dlopen, but that shouldn't be a problem for tools or for our own
internal libraries.
Co-authored-by: Mark Rousskov <mark.simulacrum@gmail.com>
|
|
Print a summary of which test suite failed
Especially on CI, where cross-compiling is common and single builder may end up
with multiple hosts and multiple targets, it can be annoying to scroll back to
the nearest start of test marker. This prints out a summary of the test suite
being run directly in compiletest.
For example, on a mir-opt failure, this would show something like this:
```
failures:
[mir-opt] mir-opt/while-storage.rs
test result: FAILED. 140 passed; 1 failed; 2 ignored; 0 measured; 0 filtered out
Some tests failed in compiletest suite=mir-opt mode=mir-opt host=x86_64-unknown-linux-gnu target=x86_64-unknown-linux-gnu
```
Fixes #78517
|
|
Allow using 1/2/3/4 for `x.py setup` options
This undocumented feature allows you to typo 'a' as '1'.
r? ```@Mark-Simulacrum```
cc ```@Lokathor```
|
|
Fix `x.py clippy`
I don't think this ever worked.
Fixes https://github.com/rust-lang/rust/issues/77309. `--fix` support is a work in progress, but works for a very small subset of `libtest`.
This works by using the host `cargo-clippy` driver; it does not use `stage0.txt` at all. To mitigate confusion from this, it gives an error if you don't have `rustc +nightly` as the default rustc in `$PATH`. Additionally, it means that bootstrap can't set `RUSTC`; this makes it no longer possible for clippy to detect the sysroot itself. Instead, bootstrap passes the sysroot to cargo.
r? `@ghost`
|
|
Here's the error for rustdoc:
```
Checking rustdoc artifacts (x86_64-unknown-linux-gnu -> x86_64-unknown-linux-gnu)
error: no library targets found in package `rustdoc-tool`
```
|
|
This fixes ongoing issues where x.py will detect the wrong host triple
between MSVC and GNU.
- Add line to changelog
|
|
|
|
Remove legacy promote-release support from build-manifestg
Now that we're not running the [legacy `promote-release`](https://github.com/rust-lang/rust-central-station/tree/master/promote-release) anymore, this PR removes support from it from `bootstrap` and `build-manifest`.
r? `@Mark-Simulacrum`
|
|
Fix panic in bootstrap for non-workspace path dependencies.
If you add a `path` dependency to a `Cargo.toml` that is located outside of the workspace, then the `in_tree_crates` function can panic because it finds a path dependency that is not defined (since it uses `cargo metadata --no-deps`). This fixes it by skipping over those entries, which are usually not things you select on the command-line.
Fixes #78617
|
|
Suggest library/std when running all stage 0 tests
r? ``@Mark-Simulacrum``
cc ``@ijackson``
For context, this came out of a discord conversation where ``@ijackson`` was running `test --stage 1` when they were only adding doc-tests to the standard library.
|
|
|
|
Especially on CI, where cross-compiling is common and single builder may end up
with multiple hosts and multiple targets, it can be annoying to scroll back to
the nearest start of test marker. This prints out a summary of the test suite
being run directly in compiletest.
|
|
|
|
This commit removes support for the legacy promote-release, as that's
not executed anymore on the nightly channel.
|
|
|
|
|
|
See https://github.com/rust-lang/rust/issues/78509 for discussion.
|
|
This undocumented feature allows you to typo 'a' as '1'.
|
|
add system-llvm-libunwind config option
allows using the system-wide llvm-libunwind as the unwinder
Workaround for #76020
|
|
Not 100% sure this will _always_ work, but it works currently.
|
|
|
|
|
|
Clippy does its own runtime detection of the sysroot, which was
incorrect in this case (it used the beta sysroot). This overrides the
sysroot to use `stage0-sysroot` instead.
- Get `x.py clippy` to work on nightly
- Give a nice error message if nightly clippy isn't installed
|
|
Add cg_clif as optional codegen backend
Rustc_codegen_cranelift is an alternative codegen backend for rustc based on Cranelift. It has the potential to improve compilation times in debug mode. In my experience the compile time improvements over debug mode LLVM for a clean build are about 20-30% in most cases.
This PR adds cg_clif as optional codegen backend. By default it is only enabled for `./x.py check`. It can be enabled for `./x.py build` too by adding `cranelift` to the `rust.codegen-backends` array in `config.toml`.
MCP: https://github.com/rust-lang/compiler-team/issues/270
r? `@Mark-Simulacrum`
|
|
|
|
|
|
Unignore test for #36710 on MUSL
This now works fine thanks to autodetected `-C link-self-contained`.
Closes #36710
|
|
|
|
Sync LLVM submodule if it has been initialized
Since having enabled the download-ci-llvm option,
and having rebased on top of #76864,
I've noticed that I had to update the llvm-project
submodule manually if it was checked out.
Orignally, the submodule update logic was
introduced to reduce the friction for contributors
to manage the submodules, or in other words, to prevent
getting PRs that have unwanted submodule rollbacks
because the contributors didn't run git submodule update.
This commit adds logic to ensure there is no inadvertent
LLVM submodule rollback in a PR if download-ci-llvm
(or llvm-config) is enabled. It will detect whether the
llvm-project submodule is initialized, and if so, update
it in any case. If it is not initialized, behaviour is
kept to not do any update/initialization.
An alternative to the chosen implementation would
be to not pass the --init command line arg to
`git submodule update` for the src/llvm-project
submodule. This would show a confusing error message
however on all builds with an uninitialized repo.
We could pass the --silent param, but we still want
it to print something if it is initialized and has
to update something.
So we just do a manual check for whether the
submodule is initialized.
|
|
Package more llvm-* tools in the rust-dev component, for run-make-fulldeps tests
Fixes https://github.com/rust-lang/rust/issues/78110
|
|
Haiku: explicitly set CMAKE_SYSTEM_NAME when cross-compiling
This resolves issues where the cross-build of LLVM fails because it tries to
link to the host's system libraries instead of the target's system libraries.
|
|
Signed-off-by: Marc-Antoine Perennou <Marc-Antoine@Perennou.com>
|
|
Since having enabled the download-ci-llvm option,
and having rebased on top of f05b47ccdfa63f8b4b9fb47a9aa92381801d3ff1,
I've noticed that I had to update the llvm-project
submodule manually if it was checked out.
Orignally, the submodule update logic was
introduced to reduce the friction for contributors
to manage the submodules, or in other words, to prevent
getting PRs that have unwanted submodule rollbacks
because the contributors didn't run git submodule update.
This commit adds logic to ensure there is no inadvertent
LLVM submodule rollback in a PR if download-ci-llvm
(or llvm-config) is enabled. It will detect whether the
llvm-project submodule is initialized, and if so, update
it in any case. If it is not initialized, behaviour is
kept to not do any update/initialization.
An alternative to the chosen implementation would
be to not pass the --init command line arg to
`git submodule update` for the src/llvm-project
submodule. This would show a confusing error message
however on all builds with an uninitialized repo.
We could pass the --silent param, but we still want
it to print something if it is initialized and has
to update something.
So we just do a manual check for whether the
submodule is initialized.
|
|
|
|
Fixes https://github.com/rust-lang/rust/issues/78110
|
|
[x.py setup] Allow setting up git hooks from other worktrees
Closes https://github.com/rust-lang/rust/issues/77684
r? @caass
|
|
This allows moving a lot of run-make-fulldeps tests to just run-make
tests, and allows running those on target-only platforms
|
|
This allows us to use the C++ compiler configured via
`CXX_target_triple` env vars
|
|
It now says "finished in xxx seconds".
Also slightly improved some wording in the README.
|
|
Modify executable checking to be more universal
This uses a dummy file to check if the filesystem being used supports the executable bit in general.
Supersedes #74753.
|
|
This resolves issues where the cross-build of LLVM fails because it tries to
link to the host's system libraries instead of the target's system libraries.
|
|
Detect configuration for LLVM during setup
This is a first draft to address #77579, setting `download-ci-llvm` to true on Linux, but I could also implement the `if-available` setting mentioned in the issue.
On other platforms I was thinking about using [the which crate](https://crates.io/crates/which), if adding a dependency on it is considered okay of course, to detect the presence of `llvm-config` in the path, and use it if found. Still a work in progress of course.
|
|
EOF is not an error; it just causes read_line to produce "".
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
We need a fresh input buffer each time, or we reuse the previous
data (since `read_line` appends).
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
We understand these profile names because we use .to_str().
Mention them in the question.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
Co-authored-by: Joshua Nelson <joshua@yottadb.com>
|
|
Put all()'s otuput in the order we want to print things in, and add a
comment about why they are in this order. Provide purpose() and
all_for_help(). Use these things everywhere.
Move all the abbrev character ("a", "b", etc.) processing into
interactive_path.
No functional change.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
bootstrap: only use compiler-builtins-c if they exist
The assignment of `features` above was added in rust-lang#60981, but
never used. Presumably the intent was to replace the string literal here
with it.
While I'm in the area, `compiler_builtins_c_feature` doesn't need to be
a `String`.
I'm not entirely sure of a great way to locally test this -- `./x.py test`
passed on my machine, but 🤷♂️.
r? @alexcrichton
|
|
bootstrap: set correct path for the build-manifest binary
This changes the path of the binary inside the tarball to be:
```
build-manifest-{release}-{target}/build-manifest/bin/build-manifest
```
...instead of:
```
build-manifest-{release}-{target}/build-manifest/bin/build-manifest/build-manifest
```
r? @Mark-Simulacrum
|
|
Include `llvm-dis`, `llc` and `opt` in `llvm-tools-preview` component
Fixes #55890
It's useful to have `llc` and `opt` available when debugging an LLVM
miscompilation,.
|
