| Age | Commit message (Collapse) | Author | Lines |
|---|
|
Pick up the latest version of iai-callgrind, which includes some output
improvements.
Changelog: https://github.com/iai-callgrind/iai-callgrind/releases
|
|
This makes it more obvious what we intend to check rather than looking
for `--target`.
|
|
|
|
The submodule was causing issues in rust-lang/rust, so eliminiate it
here. `build-musl` is also removed from `libm-test`'s default features
so the crate doesn't need to be built by default.
|
|
iai-callgrind now correctly exits with error if regressions were found
[1], so we no longer need to check for regressions manually. Remove this
check and instead exit based on the exit status of the benchmark run.
[1] https://github.com/iai-callgrind/iai-callgrind/issues/337
|
|
We don't actually need this for now, but eventually it would be nice to
run icount benchmarks on multiple targets. Start tagging artifact names
with the architecture, and allow passing `--tag` to `ci-util.py` in
order to retrieve the correct one.
|
|
Currently we only build this, but it is possible to run the binary.
Change the CI script to do so here.
|
|
This should be less error-prone and adaptable than the `nm` version, and
have better cross-platform support without needing LLVM `nm` installed.
|
|
|
|
Move this to a script and ensure only `libm-test` gets built to avoid
default feature issues with `compiler-builtins`.
|
|
|
|
|
|
Many contributions to compiler-builtins don't have any need to touch
libm, and could get by with the few minutes of CI for compiler-builtins
rather than the ~30 minutes for libm. We already have some scripts that
handle changed file detection, so expand its use to skip libm CI if it
doesn't need to run.
|
|
This makes it possible to use nextest within the container.
|
|
Update `run.sh` to start testing `libm`. Currently this is somewhat
inefficient because `builtins-test` gets run more than once on some
targets; this can be cleaned up later.
|
|
|
|
These are still not yet covered in CI since we always name explicit
packages there, but all crates are now part of the workspace.
|
|
This crate doesn't need to be a default member since it requires the
opposite settings from everything else. Exclude it from the workspace
and run it only when explicitly requested.
This also makes `cargo t --no-default-features` work without additional
qualifiers. `--no-default-features` still needs to be passed to ensure
`#![compiler_builtins]` does not get set.
compiler-builtins needs doctests disabled in order for everything to
work correctly, since this causes an error running rustdoc that is
unrelated to features (our `compiler_builtins` is getting into the crate
graph before that from the sysroot, but `#![compiler_builtins]` is not
set).
We can also remove `test = false` and `doctest = false` in
`builtins-test` since these no longer cause issues. This is unlikely to
be used but it is better to not quietly skip if anything ever gets added
by accident.
|
|
Prepare for having the repositories combined by ensuring EMULATED,
RUST_BACKTRACE, and CI are set or forwarded as applicable. Also
re-indent the file to four spaces and do some reorganization.
|
|
Apply a handful of changes to reduce the diff between the two:
* Cancel running jobs on new pushes
* Enable log color and backtraces
* Add timeouts
* Specify CI runner versions
* Add an armv7 job
* Replace the name NO_STD with BUILD_ONLY
* Update the extension to the canonical .yaml
* Set AR_ and CC_ environments in docker
* Install requirements to build MPFR
|
|
The repo will soon have `libm` as a top-level crate, so make it clear
that this is only the test crate for `compiler-builtins`.
|
|
* copy_misaligned_words: avoid out-of-bounds accesses
* add test to make Miri able to detect OOB in memmove
* run Miri on CI
|
|
The current setup has the `Cargo.toml` for `compiler-builtins` at the
repository root, which means all support crates and other files are
located within the package root. This works for now but is not the
cleanest setup since files that should or shouldn't be included in the
package need to be configured in `Cargo.toml`. If we eventually merge
`libm` development into this repository, it would be nice to make this
separation more straightforward.
Begin cleaning things up by moving the crate source to a new
`compiler-builtins` directory and adding a virtual manifest. For now the
`libm` submodule is also moved, but in the future it can likely move
back to the top level (ideally `compiler-builtins/src` would contain a
symlink to `libm/src/math`, but unfortunately it seems like Cargo does
not like something about the submodule + symlink combination).
|
|
Rather than needing to copy the version and URL from the CI workflow,
put this into a script that can be directly run locally.
|
|
Currently there is an interesting situation with the way features get
enabled; `testcrate` enables `mangled-names`, but the `intrinsics.rs`
example requires this feature be disabled (otherwise the test fails with
missing symbols, as expected). This is also the reason that `testcrate`
is not a default workspace member, meaning `cargo test` doesn't actually
run `testcrate`'s tests; making it a default member would mean that
`compiler-builtins/mangled-names` gets enabled when
`examples/intrinsics.rs` gets built, due to the way features get
unified.
Simplify the situation by making moving the example to its own crate as
`builtins-test-intrinsics`. This also means `testcrate` can become a
default member so it is included in `cargo check` or `cargo test` when
run at the workspace root.
`testcrate` and `builtins-test-intrinsics` still can't be built at the
same time since there isn't a straightforward way to have Cargo build
`compiler-builtins` twice with different features. This is a side effect
of us using non-additive features, but there isn't really a better
option since enabling both mangled and unmangled names would render
`builtins-test-intrinsics` useless.
|
|
|
|
There is a proposal to promote `riscv64gc-unknown-linux-gnu` to tier 1
[1]. We do not currently test RISC-V in CI; add it here.
[1]: https://rust-lang.zulipchat.com/#narrow/stream/131828-t-compiler/topic/Imminent.20RFC.20PR.3A.20riscv64gc-unknown-linux-gnu.20to.20Tier-1
|
|
|
|
These were preventing building via Docker locally.
|
|
|
|
|
|
This adds comparisons among the compiler-builtins function, system
functions if available, and optionally handwritten assembly.
These also help us identify inconsistencies between this crate and
system functions, which may otherwise go unnoticed if intrinsics get
lowered to inline operations rather than library calls.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
There are currently a lot of warnings printed in CI, mostly dead code.
Update CI to deny warnings.
|
|
|
|
|
|
Allow using the `rust-lang/rust:nightly` docker image to run tests in
cases where the host rust and cargo cannot be used, such as non-linux
hosts.
|
|
|
|
|
|
|
|
|
|
* Use a no-asm feature instead of an asm feature
This works better as core/alloc/std have trouble supporting default
featues in this crate.
Signed-off-by: Joe Richey <joerichey@google.com>
* Have no-asm disable arm assembly intrinsics
Signed-off-by: Joe Richey <joerichey@google.com>
|
