Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
in fact, RT-linux is ... Linux !
In the linux kernel, multitasking is achieved by sharing the processor time between processes. this is done by a kernel module, the scheduler, that switch very quickly between them. Due to the way it is done, it's not possible for a process (or an application) to generate an event at a very precise and repeatable time (there are latency and jitter)
In RT-linux, scheduling is a two stage process. The applications are either " Real Time processes" or "normal processes". At the heart of RT Linux, a real time scheduler ensures that the "RT processes" are highly prioritized and time accurate. The whole linux kernel, as we already know it with all the services it offers, becomes a kind of "low priority" RT process. All the other usual processes (the "normal" ones) are then scheduled by the linux kernel scheduler as usualy. Being "served last", theses processes can't slow down the time critical RT processes.
The RT applications use a special programming API to request special real tme scheduling and other services from the RT kernel, like accurate timers and fast and predictable interrupts servicing delays.
(RT-Linux gurus will be horrified by this highly simplistic and somehow inaccurate presentation, but you get the idea...).
AFAIK; there are currently 2 implementations of RT linux, (RTAI and RT-Linux, if I remember well, Google will help you). RT-Linux is quite popular among professionals in time critical military, communication and other demanding applications.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
