Re: new IRQ scalability changes in 2.3.48

From: Larry Sendlosky (larry@scrugs.lkg.dec.com)
Date: Fri Mar 10 2000 - 14:18:35 EST


> >> The current default for dp264 should be ok at lest for 2-way and maybe ok
> >> for 4-way too (max number of CPU supported by dp264).
> >
> >Eh. It still seems sloppy.
>
> 2-way is fine here. 4-way I don't know yet :)
>
  I have to agree with Richard here. From what I've been told, it's
  fairly "expensive" for an Alpha to "jump" for an IRQ interrupt and
  not get it - i.e. wasted time. If the code exists to support it, the
  "default setup" at init time could be to assign IRQs rotating
  through the list of CPUs. This way no IRQ would be sent to more than
  one CPU (if I'm understanding correctly how all this works)
  therefore incurring no wasted cycles.

  larry

______________________________________________________________________
 Larry Sendlosky ATS
 larry@scrugs.lkg.dec.com (978) 506-6640 Compaq
----------------------------------------------------------------------

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Wed Mar 15 2000 - 21:00:18 EST