You should review Java RTS for a similar design and it’s pros and cons. 

It is definitely doable - but you end up writing C code as Java - or in this 
case Go. 

> On Nov 29, 2020, at 1:45 AM, Fino <[email protected]> wrote:
> 
> 
> Although Go is a GC language, 
> is it any chance to use Go in Preempt_RT Linux (Xenomai, or other RTOS), for 
> real-time, deterministic, (industry control) software? 
> RTOS can offer a <50us schedule latency, it's the delay from hardware timer's 
> interrupt triggered to real-time thread being re-scheduled.
> which means if GC occupies the CPU for more than 50us, then the re-schedule 
> latency cannot be guaranteed < 50us, especially for a single core CPU.
> if the CPU have more than 2 cores, maybe the real-time thread can stay on one 
> core, and GC work on another core, I am not sure.
> 
> the idea is attractive because the dev speed of writing C is slower than a GC 
> language. and C have too much history burden,  header files for examples.  
> 
> would like to read your thinking, thanks! 
> 
> BR fino
> 
> 
> -- 
> You received this message because you are subscribed to the Google Groups 
> "golang-nuts" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to [email protected].
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/golang-nuts/c2cb3eb6-b753-42af-be88-f70beb4ff0c9n%40googlegroups.com.

-- 
You received this message because you are subscribed to the Google Groups 
"golang-nuts" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/golang-nuts/1FFF694B-EFBD-4B51-94C7-81B3FAB48977%40ix.netcom.com.

Reply via email to