Implementing the timer queue feature and even multi-core support doesn't seem like it would be too hard but IDK what would be the best way to do IO RTFM wants *signals* on IO events but I don't think any of the Linux IO API uses signals? (haven't read epoll et al man pages yet)
-
-
Prikaži ovu nit
-
P.S. I think this is the most systems program-y stuff I have done while using an OS (raw system calls) It's also the most segfaults I have gotten from unsafe Rust code, ever ABI contracts, assembly and non-standard calling conventions are all very easy to get wrong
Prikaži ovu nit
Kraj razgovora
Novi razgovor -
-
-
It's amazingly appealing even for faster dev cycle of rtfm app logic that will go into embedded
-
Definitively could be used to teach the RTFM task model w/o any embedded hardware. Cortex-M emulation on QEMU is OK-ish but the timer queue doesn't work there because the DWT CYCCNT (cycle counter) was never implemented.
- Još 1 odgovor
Novi razgovor -
-
-
Oooo! It could be used for plenty of cool things, porting SafeFlight to Raspberry Pi + Navio2 comes to mind. I’ve been working on a library for managing threads on PREEMPT_RT kernels, hard real-time RTFM on Linux would be awesome
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Might be useful for testing/debugging embedded stuff.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
ROS ( Robot Operating System ) could use a strong alternative, designed for real-time from ground up
-
There aren't that many frameworks to choose from in the first place, Finroc is one, YARP another, OpenRTM maybe another
Kraj razgovora
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.