Fix ci #15
Fix ci #15
Conversation
This drops support for calling .retire on a *mut HazPtrObjectWrapper
There was a problem hiding this comment.
Thanks for looking into this! I think you'll need to merge from main to get the no_std changes though.
Also, separately, I just noticed that the file name for the CI file is misspelled, which is probably why CI isn't taking effect. It should be azure-pipelines.yml (the e and the n are swapped). Could you fix that too?
src/domain.rs
Outdated
| use crate::{Deleter, HazPtrRecord, Reclaim}; | ||
| use alloc::boxed::Box; | ||
| use alloc::collections::BTreeSet; | ||
| use core::marker::PhantomData; | ||
| use core::sync::atomic::Ordering; | ||
|
|
||
| #[cfg(feature = "std")] | ||
| #[cfg(all(feature = "std", not(miri)))] |
There was a problem hiding this comment.
I'm not sure I follow why it's necessary to turn this off under miri?
There was a problem hiding this comment.
It goes with the other "time" related things so I believe it makes things cleaner when due_time is disabled (see comment below)
There was a problem hiding this comment.
Hmm, I still don't understand after reading your comment. From what I can tell, miri already mocks time just fine (rust-lang/miri#975), and supports Instant::now (rust-lang/miri#1243), and I don't see anything about time-related limitations in the miri readme? What error do you get if you make miri run with time supported?
There was a problem hiding this comment.
Actually, I'm going to merge this as-is just so we can iterate just on the miri point and have working CI in the meantime!
There was a problem hiding this comment.
Ah, I was getting the clock_gettime not available when isolation is enabled error, not knowing about -Zmiri-disable-isolation. This should be pretty easy to address in a PR.
|
Took another pass! |
This PR ports over most of Flutty's CI config, including:
This disables due_time checking when on miri, and places the
due_timevariable behind a #[cfg] guard.Also this adds support for the stable channel by getting rid of the
arbitrary_self_typesfeature. This means all instances where.retirecould be called on a*mut HazPtrObjectWrappermust be replaced withHazPtrObjectWrapper::retire.This is a little sad. Possibly we could add a
retirefunction in the crate's root that callsHazPtrObjectWrapper::retireto make user code a little cleaner, buthaphazard::retire(..., deleter)is still much more cumbersome than(...).retire(deleter).Clippy isn't happy at the moment, but all other tasks pass.