Re: ulogd unaligned trap
On Sat, Jan 21, 2006 at 07:25:47PM -0500, Jim Crilly wrote:
> On 01/21/06 09:45:15PM +0000, Carlos Rodrigues wrote:
> > Hi!
> >
> > I have a couple of Alpha machines (XP1000 and ES40) which spit this
> > message regularly to /var/log/messages and the console:
> >
> > Jan 21 21:41:05 ulisses kernel: ulogd(26459): unaligned trap at
> > 00000200001e2ce8: 000002000020a0e5 28 4
> > Jan 21 21:41:05 ulisses kernel: ulogd(26459): unaligned trap at
> > 00000200001e2d64: 000002000020a0e9 28 4
> > Jan 21 21:41:05 ulisses kernel: ulogd(26459): unaligned trap at
> > 00000200001e2da8: 000002000020a0e1 28 3
> > Jan 21 21:41:05 ulisses kernel: ulogd(26459): unaligned trap at
> > 00000200001e2df4: 000002000020a0e1 28 2
> >
> > I think (not sure) this started happening since the latest kernel
> > security update.
> >
> > Both machines are running sarge, with the stock (debian) 2.4.27 kernel
> > (smp, in the ES40).
> >
> > I don't think this is affecting the operation of ulogd, but I'm not
> > sure. Anyways, I reported this to the BTS.
> >
> > Any ideas?
> >
>
> Ulogd will keep running just fine. What's happening is it's attempting
> to access memory that's not aligned properly, the hardware complains
> and the kernel fixes the access so the hardware is happy. It cause
> a bit of a slowdown since the kernel has to fix up the operation,
> but it doesn't cause any other problems.
It can be more than a little slowdown, if it happens a lot.
Check for the frequency via:
cat /proc/cpuinfo | grep unalign
and see how often/how much it changes.
BTW, the addresses spit out in the message look like they are from
a module, which may make tracking it down a bit more difficult...
Good luck.
--Jay++
---------------------------------------------------------------
Jay A Estabrook HPTC - XC I & B
Hewlett-Packard Company - ZKO1-3/D-B.8 (603) 884-0301
110 Spit Brook Road, Nashua NH 03062 Jay.Estabrook@hp.com
---------------------------------------------------------------
Reply to: