koenig@tat.physik.uni-tuebingen.de said:
> I've tested this long time ago with Linux-1.0...1.3 with my 486DX2/
> 66. there I got ~11-12 usec interrupt latency (I just read the timer
> latch in the clock timer ISR, then you exactly know how many usec it
> took from timer countdown to get into the ISR...)
This roughly corresponds with my experiences with a 75MHz Pentium in
a Packard Hell. I actually measured the width if the PCI INTA# signal from
my board, and the pulse width was < 30uSec. This of course included some
PCI I/O to get the interrupt request info and clear the IRQ source. YMMV.
It is not clear how Linux, the Pentium and the chipset contribute to that
delay.
So, although I suspect 2.0 is a bit slower then 1.x, I'd say not by much.
Note that I also did the exact same test with Windows-NT 4.0 and got INTA#
widths of 60uSec. I performed exactly the same operations to clear the
interrupt in both cases, and used the exact same hardware.
-- Steve Williams steve@icarus.com steve@picturel.com"The woods are lovely, dark and deep. But I have promises to keep, And lines to code before I sleep, And lines to code before I sleep."