On Fri, 28 Sep 2001, Stephen D. Cohen wrote:
> Mr. Culianu,
>
> Did you see my post earlier on this subject? It covered these
> issues nicely.
>
> > Wait.. so in the kernel the rtf_put() call can fail if the fifo is too
> > full?
>
> No, Mr. Kavaler is wrong.
>
> > I thought it clobbered the existing data, but, not actually
> > failed.. hmm.
>
> Despite what I wrote earlier, Mr. Kavaler is correct on this issue.
> The call will fail with -ENOSPC if insufficient space exists in the FIFO.
> This is consistent with the man page for rtf_put.
>
Yes, I should have read the docs. Now I am actually reading the source..
so I am being better today. :)
> > And you are also telling me that a read()
> > system call in
> > userland will NOT be interrupted for the duration of the system call?
> > That is good news on one hand, but on the other I suspect such things
> > affect real-time scheduler jitter :)
>
> It is not obvious to me from the source that this is guaranteed.
> Only a small portion of the read is protected with spinlocks and they do
> just a small fraction of the total work. I would not count on this behavior
> unless you test it thoroughly yourself.
Hmm I see.. I truly wonder what happens.. because race conditions on
fifos are never pretty.
Well at any rate you have been tremendously helpful (as this mailing list
always is) and I appreciate your help!
Regards,
-Calin
>
> Regards,
>
> Steve Cohen
>
> --------------------------
> Stephen D. Cohen
> Xybion Sensor Positioning Systems
> 11528 53rd Street North
> Clearwater, FL 33760
> Voice: (727) 299-0150
> Fax: (727) 299-0804
> [EMAIL PROTECTED]
> www.xybion.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/
