You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've been experimenting on and off with doing this more broadly, and it has helped me find various problems that I've already submitted fixes for. I'm hoping it will also help give us confidence around issues like #685; we can't know exactly what a real UEFI implementation does in all cases, so we can never be 100% sure that Rust's abstract machine isn't being violated in some way, but I think with a sensible stub implementation we can get closer.
For now I don't have code that's even close to being PR-able yet, but filing this as a tracking issue.
The text was updated successfully, but these errors were encountered:
Right now our Miri coverage is pretty limited since most of
uefi
requires interacting with the UEFI environment. There's a solution though, which is to create stubs for that environment. We do a small version of this in the file tests: https://github.com/rust-osdev/uefi-rs/blob/main/uefi/src/proto/media/file/mod.rs#L447I've been experimenting on and off with doing this more broadly, and it has helped me find various problems that I've already submitted fixes for. I'm hoping it will also help give us confidence around issues like #685; we can't know exactly what a real UEFI implementation does in all cases, so we can never be 100% sure that Rust's abstract machine isn't being violated in some way, but I think with a sensible stub implementation we can get closer.
For now I don't have code that's even close to being PR-able yet, but filing this as a tracking issue.
The text was updated successfully, but these errors were encountered: