Hi Hosam,
You can find more information about Linux and POSIX.4 (POSIX 1003.1b after
the big renumbering) in the following url:
ftp://ftp.informatik.uni-erlangen.de/local/cip/mskuhn/misc/linux-posix.1b
One thing that you have to take into account is the fact that POSIX.1b is
structured as a set of options that a vendor can support or not. Only some
functions are compulsory to say that a system is POSIX.1b compliant (If 'm
not wrong these options are the queue signals and not much more).
RTLinux, as far I know, is not a POSIX way but is a REAL way to get a RT
system working. Another POSIXish alternative is QNX (www.qnx.com) for
example, but it's commercial.
Finally, if you're interested in the POSIX way I strongly recommend the
following book "POSIX.4: Programming for the real world, Bill O.
Gallmeister, O'Reilly & Associates, 1995"
best regards,
Juanjo
At 03:23 23/05/99 PDT, you wrote:
>Dear ...,
>
>I am a student in Mech. Eng., Kansas State Univ., and I'm working on
>developing software for a GPS-based avionics package. Initially, I was told
>to develop all the software in Windows NT because "the people who give us
>our research money say so". One of the requirements for the software package
>was that it had to run in "hard real time". Well, I built the software
>package, with multiple threads and timers and critical sections and
>asynchronous I/O and memory page locking and all the rest of the thread
>timing/synchronization tools that I needed. And unfortunately, after I built
>the whole package in the Win32 API, I started testing, and the results were
>disastrous: the timing errors were on the same order of magnitude as the
>desired computational time period! I guess I learned the "hard way" that
>Windows NT can't be used for "hard real-time" applications (excuse the pun).
>I then found out about RT Linux, and the performance of RT Linux seems to be
>simply superior to anything I have achieved with NT. So I was asked by my
>advisors to port the software to RT Linux. I need to mention that both
>myself and my advisors are "mechanical engineering" people: we know
>absolutely nothing about operating systems. We "think" that both RT Linux
>and Windows NT support POSIX, and hence we "think" it makes sense to port
>the software to POSIX first, then move it over to RT Linux (our software
>also uses MFC for displays, so we figured out that by porting to POSIX,
>running it within NT, and THEN worrying about doing the rest of the porting,
>the task will be simpler). We are also aware that for real-time
>applications, we need POSIX.4. So I guess my questions are:
>
>1-Does RT Linux support POSIX.4? If yes, does NT also support it?
>2-What strategy would you recommend for porting the software?
>3-Our software needs to communicate with 10 GPS receivers simultaneously
>through COM connections. How hard is it to develop Linux drivers for
>special-purpose serial boards?
>
>Finally, thank you very much.
>
>Sincerely Yours,
>Hosam K. Fathy.
>
>
>_______________________________________________________________
>Get Free Email and Do More On The Web. Visit http://www.msn.com
>--- [rtl] ---
>To unsubscribe:
>echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
>echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
>----
>For more information on Real-Time Linux see:
>http://www.rtlinux.org/~rtlinux/
>
>
>
--- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
----
For more information on Real-Time Linux see:
http://www.rtlinux.org/~rtlinux/
