Re: scheduling problem test13-pre7

From: Mike Galbraith (mikeg@wen-online.de)
Date: Sun Dec 31 2000 - 06:44:22 EST


On Sun, 31 Dec 2000, Matti Aarnio wrote:

> On Sun, Dec 31, 2000 at 10:42:26AM +0100, Mike Galbraith wrote:
> > Hi,
> >
> > While running iozone, I notice severe stalls of vmstat output
> > despite vmstat running SCHED_RR and mlockall().
>
> Lets eliminate the obvious:
>
> - Are you running with IDE disk ?

Yes.

> - Does hdparm /dev/hda(whatever) report:
>
> /dev/hda:
> unmaskirq = 0 (off)
> using_dma = 0 (off)

No.

/dev/hda:
 multcount = 0 (off)
 I/O support = 1 (32-bit)
 unmaskirq = 1 (on)
 using_dma = 1 (on)
 keepsettings = 1 (on)
 nowerr = 0 (off)
 readonly = 0 (off)
 readahead = 8 (on)
 geometry = 2482/255/63, sectors = 39876480, start = 0

>
> The IKD uses local interrupts, so this isn't necessarily true...

I just did a (mondo) trace covering 8 seconds of kernel time, and
vmstat ran twice. (Those two times were before I noticed the stall
and started counting down toward 'poke the freeze-frame button')

        -Mike

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



This archive was generated by hypermail 2b29 : Sun Dec 31 2000 - 21:00:15 EST