Hi Norm,

The fifos found in RTAI include support for named fifos, which open and
return the next free fifo, it the name is not already open.  This module
works with either rtai or rtl and provides all the usual API calls.

Regards, Stuart

"Dresner, Norman A." wrote:
> 
> How about a simple kernel module that tracks which FIFO's are in use and
> reserves and returns the next one.  A simple time-out on a reservation would
> protect against at least some faults?
> 
> Norman Dresner
> Fellow Systems Engineer
> Radar Systems Engineering Department
> Electronic Systems and Sensors Segment
> Northrop Grumman Corporation
> Baltimore-Washington International Airport
> 7323 Aviation Boulevard
> Baltimore Maryland 21240
> 
> Voice:  (410) 993 - 2096        Mornings; all-day voice-mail
>         (410) 969 - 8068        Afternoons with answering machine
> FAX:    (410) 993 - 8084        On-site
>         (410) 969 - 8068        Afternoons; call first to arrange
> E-Mail:Mornings:        [EMAIL PROTECTED]
>         Afternoons:     [EMAIL PROTECTED]
> 
> > -----Original Message-----
> > From: Eric Peterson [SMTP:[EMAIL PROTECTED]]
> > Sent: Thursday, December 14, 2000 6:58 PM
> > To:   Stephen D. Cohen
> > Cc:   '[EMAIL PROTECTED]'
> > Subject:      Re: [rtl] Using RTFifos with multiple writers?
> >
> > On Thu, Dec 14, 2000 at 11:37:29AM -0500, Stephen D. Cohen wrote:
> > > Hi,
> > >
> > >     David Olofson offered an excellent treatise on why multiple open of
> > a
> > >     FIFO is a bad idea.  I would like to offer a fairly simple solution.
> > >
> > >     Simply create a block of FIFOs on the RTL side of the house.  Attach
> > >     them all to the same handler (it gets the FIFO number when it is
> > >     called, but since you were planning to multi-open, it probably
> > doesn't
> > >     matter).
> >
> > >     Then simply have your user-mode programs try opening FIFOs until
> > they
> > >     succeed or run out of FIFOs to try.
> >
> > Steve,
> >
> >     Thanks for the suggestion - I thought a bit about doing this, but it
> >     doesn't scale too well (n**2).  Assuming I have a unique FIFO for each
> >     RT-task, and one for every Linux process, the number of FIFOs starts
> >     getting large.
> >
> >     David O's suggestion of a message queue seems to be a good solution -
> > now
> >     just to find one...
> >
> > Regards,
> > Eric
> >
> > --
> > God made everything out of nothing, but the nothingness shows through.
> >               -- Paul Valery
> >
> > Eric Peterson WB6PYK (805)370-3046 [EMAIL PROTECTED] HOME:
> > [EMAIL PROTECTED]
> > -- [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/


-- [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/

Reply via email to