Re: Measured overhead of timer interrupts

Alan Cox (alan@lxorguk.ukuu.org.uk)
Mon, 19 Jul 1999 17:01:57 +0100 (BST)

Messages sorted by: [ date ][ thread ][ subject ][ author ]
Next message: Matthew Kirkwood: "[patch] capability paranoia"
Previous message: yodaiken@chelm.cs.nmt.edu: "Re: Power PC."

> Summary: Single 300MHz Pentium II on 440LX chipset.
> Average timer interrupt overhead at 100Hz: 0.14225%
> Average cycles per timer interrupt: 41240.6
> Average time per timer interrupt: 137.47
>
> Extrapolating to 1024Hz, expect 1.4566% time to go servicing the timer
> interrupts.

Your extrapolation is broken. Try running a 1024Hz kernel. Timer work
does not accrue at 10.24 times the rate as the clock increases

-
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/

Next message: Matthew Kirkwood: "[patch] capability paranoia"
Previous message: yodaiken@chelm.cs.nmt.edu: "Re: Power PC."