First draft at concurrency chapter #49
Conversation
adamgreig
added
T-resources
S-waiting-on-review
|
This is high quality content. Hats off! I am wondering about one point: The mutex-inside-critical-section example confuses me a bit. In a single-core scenario, it doesn't make much sense, because a critical section is already enough to keep the only contender for the shared data away? In a multi-core example, we could still get weird behavior if one core currently holds the lock when the other core in an ISR tries to lock it. Maybe I am missing something here? |
|
Thanks! In the single-core case, the mutex is really only ensuring you are in fact inside a critical section - it doesn't have its own state or concept of being locked or any other overhead. Really all it's doing is only letting you borrow its contents if you can give it a In a multi-core scenario the |
|
I was just asking because the introduction speaks about multi-core. Would it help to set the scope for each of the examples (single or multi or both) so that the reader can differentiate? |
|
Good idea, I'll add a note that the |
|
@andre-richter I've added some notes about multiple cores. Good observation: it's lead to needing to also note that |
|
Shall we bors this @adamgreig ? |
|
I've just had a typo pointed out so give me a moment to fix that... |
adamgreig
dismissed stale reviews from therealprof and andre-richter
via
9d3594b
|
Okay, fixed, if you're happy then go ahead and bors. Thanks! |
49: First draft at concurrency chapter r=andre-richter a=adamgreig I've mostly focused on building up to safely sharing peripherals, without very much focus on RTFM or higher level things like an RTOS, since I don't think there's very much we can usefully say about them at the moment. Very happy to be told I'm wrong on that front though! Closes #7. r? @rust-embedded/resources Co-authored-by: Adam Greig <adam@adamgreig.com>
Build succeeded |
49: First draft at concurrency chapter r=andre-richter a=adamgreig I've mostly focused on building up to safely sharing peripherals, without very much focus on RTFM or higher level things like an RTOS, since I don't think there's very much we can usefully say about them at the moment. Very happy to be told I'm wrong on that front though! Closes rust-embedded#7. r? @rust-embedded/resources Co-authored-by: Adam Greig <adam@adamgreig.com>
I've mostly focused on building up to safely sharing peripherals, without very much focus on RTFM or higher level things like an RTOS, since I don't think there's very much we can usefully say about them at the moment. Very happy to be told I'm wrong on that front though!
Closes #7.
r? @rust-embedded/resources