Der Herr Hofrat wrote:
>
> Hi !
>
> rtlinux-2.3/examples/measurement/rt_process.o + monitor
>
> platform i486/75
> rtlinux 2.3
> kernel 2.2.14
> egcs 2.91.66
>
> this is what it looks like at the beginning:
> min: 4526, max: 14147
> min: 4509, max: 13427
> min: 4509, max: 12656
> ....
> min: 4509, max: 10292
> min: 4526, max: 12903804 <- then the first jump occures:
> min: 12841919, max: 12916945
> min: 12842002, max: 12913023
> min: 12841952, max: 12872710
> ....
> min: 12819441, max: 12924940
> min: 12817077, max: 12922460
> min: 12817144, max: 25650581 <- the next jump :
> min: 24841919, max: 24890863
> min: 24842036, max: 24885650
> ...
> min: 24842002, max: 24918956
> min: 24842069, max: 24895054
> min: 24842203, max: 38351112 <- etc....
> min: 37842203, max: 37913710
> min: 37842036, max: 37894199
>
> if I restart the test it repeats begining at 8us jitter again...
> what is happening here ? I guess its some rolleover or so that is happening
> but where ?
>
> any hint ? did anybody else get this result with 2.3 ?
>
Hi Nick,
I have seen this if you let the mode default to periodic on 486's, the
emulated tsc will overflow (8254). Try this in one-shot mode and it
should work.
Regards, Stuart.
-- [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/
