Add reset function to cortexm::support #3404
Conversation
| scb::reset(); | ||
| } | ||
| loop { | ||
| // This is required to avoid the empty loop clippy | ||
| // warning #[warn(clippy::empty_loop)] | ||
| nop(); | ||
| } | ||
| } |
There was a problem hiding this comment.
I think that scb::reset should also have the signature fn () -> ! and the nop loop should be moved into there.
|
I think in theory, boards would wrap the unsafe cortexm::reset function in a safe version, and pass that to pconsole. That would effectively be using the board's authority to call unsafe to determine that it is ok for that capsule to have what is an unsafe function. But, I think that adds unnecessary overhead. What is here looks fine. |
|
I agree that |
|
@lschuermann I am not sure what you want me to do 😄 , my git skills are not that good. |
Pull Request Overview
This pull request adds a reset function to the
cortexmcrate incortexm::support::reset. The function can be directly used by the process console as a reset function. Before this, boards had to define a functionfn reset() -> !that used thecortexm::scb::resetfunction.As discussed in #3398 (review), the
resetfunction is marked as being safe as capsules do jot depend on thecortexmcrate. I am not sure about this, maybe we should guard this using a permission so that even if a capsules does depend ion the arch crate, it cannot use this function.Testing Strategy
N/A
TODO or Help Wanted
N/A
Documentation Updated
/docs, or no updates are required.Formatting
make prepush.