IRQ latency and context switching on mvme5500 (was Re: RTEMS mvme5500 bsp)

[Kate Feng]

Peter Dufault wrote:

>
> On Jul 27, 2005, at 10:26 PM, Kate Feng wrote:
>
> >
> > Based on the test  result, it shows that  RTEMS-mvme5500 is
> > more deterministic  and  steadier than vxWorks-mvme5500
> > for the highest priority task.  For  both  the idle  and loaded
> > system,
> > RTEMS-mvme5500 "GUARANTEED"  twice or three times
> > faster response time in a steadier state.     The "worst case" is a
> > critical factor  in consideration of the real-time  system.
> >
> >
> >
>
> Can you also include all the identifying information about the
> vxWorks version, and whether you've modified anything from the
> distribution?

Steve changed  the cache setup  of vxWokrs from
its default (enabled though) to be the same as  that of
RTEMS-mvme5500.   Now,  I get a more comparable result.
This might  not  imply the  conclusion for RTEMS-mvme5500
yet, except I will not be able to  enhance  it further for a while.
Please note the  results  of the loaded system vary everyday
depending on the  network condition.  --Kate


Friday, July 29:
______________________________________________________
MVME5500    Interrupt   Latency    |   Context Switching
___________________________________________________________
                     max      (average)         |     max    (average)
____________________________________________________________
Idle System:

RTEMS        5.04       (3.45)            |    6.80     (0.96)

vxWorks      6.10       (1.58)            |    9.65     (0.91)

____________________________________________________________
Loaded System:

RTEMS        8.17       (3.74)            |   17.48     (1.69)

vxWorks     13.90       (1.68)           |   20.80     (1.90)
____________________________________________________________

** All units are in usec.