Handle Miri sysroot entirely outside the Miri driver #3411
Conversation
|
The issue with this is that it breaks autocfg for cross-target interpretation. We set RUSTC to be the Miri driver, and autocfg ignores RUSTC_WRAPPER, so it directly invokes the driver and there is no way for us to inject the right sysroot. So this is probably blocked on cuviper/autocfg#26. (dtolnay's crates seem to already support RUSTC_WRAPPER.) |
|
☔ The latest upstream changes (presumably #3414) made this pull request unmergeable. Please resolve the merge conflicts. |
|
autocfg got patched. :) So this should pass CI now. It's still a bit fragile though due to rust-lang/cargo#10885. Also if we land this now cross-builds will break for anyone using autocfg older than 1.2. (But host builds should work, autocfg probes will just use the wrong sysroot.) |
|
Ah the eyre build probe also still fails to apply RUSTC_WRAPPER. So landing this would break everything that uses eyre... IOW, blocked on eyre-rs/eyre#84. |
cargo-miri, miri-script, tests/ui: misc simplifications and comments Parts of #3411 that can be landed now.
This comment was marked as resolved.
This comment was marked as resolved.
RalfJung
force-pushed
the
sysroot
branch
2 times, most recently
from
a49c23c
to
b0f6fea
Compare
|
Actually looks like eyre still works with this. I guess by using neither RUSTC_WRAPPER nor The one thing that does not work any more with this PR is setting So I think this is currently my preferred approach to getting rid of the |
RalfJung
force-pushed
the
sysroot
branch
3 times, most recently
from
0333893
to
268ad82
Compare
cargo-miri, miri-script, tests/ui: misc simplifications and comments Parts of rust-lang/miri#3411 that can be landed now.
|
Also this should only break with old autocfg when using |
Based on the downloads graph on crates.io, 1.2 seems like just over a third of downloads. I'm concerned we have a surprising amount of --target usage due to our much better shim support on x86_64-unknown-linux-gnu, so I would prefer to wait a week from now. |
|
☔ The latest upstream changes (presumably #3453) made this pull request unmergeable. Please resolve the merge conflicts. |
…ts passed to the driver
We have waited 2 weeks since then. Download stats for autocfg haven't really changed. Let's ship this; we can always revert if there are too many problems. |
Handle Miri sysroot entirely outside the Miri driver (Extracted from #3409) This entirely moves the responsibility of setting miri-sysroot to whatever *invokes* the Miri driver. cargo-miri knows whether it is inside rustdoc or not and can adjust accordingly. I previously avoided doing that because there are a bunch of places that are invoking the driver (cargo-miri, the ui test suite, `./miri run`, `./x.py run miri`) and they all need to be adjusted now. But it is also somewhat less fragile as we usually have more information there -- and we can just decide that `./miri run file.rs --sysroot path` is not supported. The advantage of this is that the driver is reasonably clean and doesn't need magic environment variables like MIRI_SYSROOT, and we don't have to fight rustc_driver to use a different default sysroot. Everything is done in cargo-miri (and the other much simpler driver wrappers) where it can hopefully be debugged much better.
|
The build has long finished, but somehow bors didn't notice...? |
|
☀️ Test successful - checks-actions |
(Extracted from #3409)
This entirely moves the responsibility of setting miri-sysroot to whatever invokes the Miri driver. cargo-miri knows whether it is inside rustdoc or not and can adjust accordingly. I previously avoided doing that because there are a bunch of places that are invoking the driver (cargo-miri, the ui test suite,
./miri run,./x.py run miri) and they all need to be adjusted now. But it is also somewhat less fragile as we usually have more information there -- and we can just decide that./miri run file.rs --sysroot pathis not supported. The advantage of this is that the driver is reasonably clean and doesn't need magic environment variables like MIRI_SYSROOT, and we don't have to fight rustc_driver to use a different default sysroot. Everything is done in cargo-miri (and the other much simpler driver wrappers) where it can hopefully be debugged much better.