Messages in this thread |  | | Date | Mon, 4 Nov 1996 22:33:55 -0500 (EST) | From | Tyson D Sawyer <> | Subject | Realtime Linux |
| |
I am presently evalutating the specs of RT-Linux as available from http://luz.cs.nmt.edu/~rtlinux/. It looks like a simple and clean implementation that should work about as well as PC hardware can work for simple Hard-RT applications.
One of my concerns that that interest in RT-Linux may be dropped and it will discontinue being updated. I am quite willing to help support the project but don't have the resources to keep it up to date on my own. I wrote to Michael Barabanov <baraban@nmt.edu> with a few questions and haven't recieved any reply. To be fair is has only been a couple of days other than the weekend since I wrote. Michael Barabanov is the developer/maintainer of the web site but it is unclear who developed the patch.
It would be helpful to hear from those who are using RT-Linux, those that are interested in RT-Linux and from Linus with any input he might have on RT-Linux sometime being included in the standard kernel sources. Under what conditions might a version (or other implementation) of RT-Linux be included?
Note that RT-Linux goes far beyond POSIX-RT in that it goes under the kernel and tests have shown the processes can be scheduled and interrupts can be services with aproximately 15uS latency/resolution. I suspect that RT-Linux is (at least presently) incompatible with SMP. This is one aspect I would hope might change sometime.
I need to have some idea of the level of interest/support for RT-Linux before choosing it for a few projects.
Cheers! Ty
|  |