IRQ latency (was Re: RTEMS-MVME5500 BSP v1.3 available)

[Kate Feng]

Kate Feng wrote:

> My impression is that Till Strumann improved the clockIsr()
> routine of c/src/lib/libcpu/powerpc/mpc6xx/clock/c_clock.c
> to optimize the interrupt  overhead  for all the PowerPCs.
> He filed the  PR #773.
>
> However, in the  following thread :
> http://www.rtems.com/ml/rtems-users/2005/march/msg00134.html
>
> Till Straumann wrote:
>
> >BTW: I should also have disabled interrupts around the update
> >to avoid the race condition [higher priority IRQ between mfdec
> >and mtdec].
>

Peter Dufault  wrote :

> Me? No, I *never* disable interrupts! (well, almost never).

Yes, it's better to limit it to the minimum.

> But seriously, no I didn't disable interrupts. I'm a firm believer in
>lock-free systems, which to my thinking means trying to push the
>locks down to the hardware level.

I  am not sure  if you will get an  accurate timestamp in you test as
you mention  that you use  system clock, especially  under the heavy
load.   Anyway,  Till should make a  conclusion about  its pedigree.
He is the one who made the enhancement  in RTEMS4.7 PPC
shared port.


Thanks,
Kate