Open davidbarsky opened 2 months ago
Name | Link |
---|---|
Latest commit | 0102ffb852d6a90ca509dc991b1a3c0cc3cf183a |
Latest deploy log | https://app.netlify.com/sites/salsa-rs/deploys/67118638fd98680008254f64 |
Comparing davidbarsky/push-lqummkkxvkzn
(0102ffb) with master
(710691d)
✅ 8
untouched benchmarks
The Miri errors appear to be coming from crossbeam-epoch
, but I'm not sure how bad of an error that is.
The Miri errors appear to be coming from
crossbeam-epoch
, but I'm not sure how bad of an error that is.
Judging from the issue tracker that crate seems to have a couple stacked borrows violations
Does miri have a way to add 'known failures'...
@davidbarsky I'd be ok moving the miri tests to "warn only" for the time being and then trying to investigate the use of crossbeam-epoch more closely.
@davidbarsky I'd be ok moving the miri tests to "warn only" for the time being and then trying to investigate the use of crossbeam-epoch more closely.
Lemme see about doing that. I think I still need to port over a test or two over as well.
Added a cancellation test.
@davidbarsky what's the status of this PR? We now have a use case where the hacky hand rolled concurrency no longer works :laughing:
Initial pass; very much not ready. Putting it up for initial feedback; I'm still figuring out how to wrap
&Db
reasonably.