This is a last-ditch attempt to help our pain with dealing with #38878 on the
bots. A new environment variable is added to the compiler,
`RUSTC_RETRY_LINKER_ON_SEGFAULT`, which will instruct the compiler to
automatically retry the final linker invocation if it looks like the linker
segfaulted (up to 2 extra times).
Unfortunately there have been no successful attempts to debug #38878. The only
information seems to be that the linker (e.g. `ld` on OSX) is segfaulting
somewhere in some thread pool implementation. This appears to be spurious as
failed PRs will later merge.
The hope is that this helps the queue keep moving without clogging and delaying
PRs due to #38878.
This change introduces a Dockerfile and script which builds a complete
Fuchsia toolchain which can be used to build Rust distribution for
Fuchsia. We only support cross-compiling at the moment, hence only
setting the target.
This commit alters the compiler to exit quickly if the only output being emitted
is `dep-info`, which doesn't need a lot of other information to generate.
Closes#40328
A task function is now given as a `fn` pointer to ensure that it carries
no state. Each fn can take two arguments, because that worked out to be
convenient -- these two arguments must be of some type that is
`DepGraphSafe`, a new trait that is intended to prevent "leaking"
information into the task that was derived from tracked state.
This intentionally leaves `DepGraph::in_task()`, the more common form,
alone. Eventually all uses of `DepGraph::in_task()` should be ported
to `with_task()`, but I wanted to start with a smaller subset.
Originally I wanted to use closures bound by an auto trait, but that
approach has some limitations:
- the trait cannot have a `read()` method; since the current method
is unused, that may not be a problem.
- more importantly, we would want the auto trait to be "undefined" for all types
*by default* -- that is, this use case doesn't really fit the typical
auto trait scenario. For example, imagine that there is a `u32` loaded
out of a `hir::Node` -- we don't really want to be passing that
`u32` into the task!
In recent months there have been a few different people investigating how to make a plugin that
registers a MIR-pass – one that isn’t intended to be eventually merged into rustc proper.
The interface to register MIR passes was added primarily for miri (& later was
found to make prototyping of rustc-proper MIR passes a tiny bit faster). Since miri does not use
this interface anymore it seems like a good time to remove this "feature".
For prototyping purposes a similar interface can be added by developers themselves in their custom
rustc build.
When declaring nested unsafe blocks (`unsafe {unsafe {}}`) that trigger
the "unnecessary `unsafe` block" error, point out the enclosing `unsafe
block` or `unsafe fn` that makes it unnecessary.
rustbuild: Use copies instead of hard links
The original motivation for hard links was to speed up the various stages of
rustbuild, but in the end this is causing problems on Windows (#39504).
This commit tweaks the build system to use copies instead of hard links
unconditionally to ensure that the files accessed by Windows are always
disjoint.
Locally this added .3s to a noop build, so it shouldn't be too much of a
regression hopefully!
Closes#39504
travis: Split the linux-tested-targets builder
Travis only gives us 30GB disk space and we don't currently have an option to
increase that. Each musl target generates "hello world" binaries of about 3.5MB
in size, and we're testing two targets in the same image. We have around 3k
run-pass tests and 2 musl targets which works out to around 20GB. That's
dangerously close to the limit and is causing PRs to bounce.
This PR splits up the builder in two, one for x86_64 musl and the other for
i686. Hopefully that'll keep us under the disk limit.
Closes#40359
Travis only gives us 30GB disk space and we don't currently have an option to
increase that. Each musl target generates "hello world" binaries of about 3.5MB
in size, and we're testing two targets in the same image. We have around 3k
run-pass tests and 2 musl targets which works out to around 20GB. That's
dangerously close to the limit and is causing PRs to bounce.
This PR splits up the builder in two, one for x86_64 musl and the other for
i686. Hopefully that'll keep us under the disk limit.
Closes#40359
This commit is a random stab in the dark to fix the spurious failures on #39518.
The leading theory of the spurious failures on Windows is that the compiler is
loading a path in the `deps` folder, passing it to `link.exe`, and then this is
racing with Cargo itself updating those paths.
This race, however, has a few unique properties:
* It's isolated to just libstd. Most crates are never passed to the linker and
simultaneously being worked on by Cargo. Cargo's typical execution of the
dependency graph never hits this problem.
* The crates are already all located in the sysroot in addition to the `deps`
folder. This means that the compiler actually has two candidates of crates to
load, and it's just arbitrarily rejecting one.
Together this means that we shouldn't need to fix this problem "in the large"
and we can instead just fix it in this isolated situation (hopefully). To solve
this the compiler's been updated to prefer crates from the sysroot to leave
Cargo's structure to itself.
We'll see if this actually allows the PR to land...
The original motivation for hard links was to speed up the various stages of
rustbuild, but in the end this is causing problems on Windows (#39504).
This commit tweaks the build system to use copies instead of hard links
unconditionally to ensure that the files accessed by Windows are always
disjoint.
Locally this added .3s to a noop build, so it shouldn't be too much of a
regression hopefully!
