Sparc slowdown - problem identified...

[Thomas Moestl]

On Sat, 2003/08/16 at 14:27:27 -0700, David O'Brien wrote:
> On Fri, Aug 15, 2003 at 04:34:04PM +0200, Thomas Moestl wrote:
> > > The notes for OFW_NEWPCI say:
> > > 
> > > # New OpenFirmware PCI framework. This fixes a number of interrupt-
> > > # routing problems and changes the device enumeration to be hopefully
> > > # closer to Solaris. Be aware that, because of the latter, enabling or
> > > # disabling this option may require reconfiguration, and can even
> > > # cause the machine to not boot without manual intervention before the
> > > # fstab is adjusted.
> > > 
> > > What sort of changes are likely to occur that would affect fstab? The
> > > box is remote, so I can fix most things via a serial console as long as
> > > it'll boot :-)
> 
> Please, please, please, please consider making OFW_NEWPCI the default.

I'll probably do that soon, but of course the bug which leaves the
caches disabled should also be fixed.

> After turning it on, 'make buildworld':
> 
>         3h37m12.24s real  2h56m32.37s user   34m20.62s sys
> 
> still a little slower than before, but not the 4x degradation I
> experienced.
> 
> I'm curious why you didn't see this on your Blade 100 didn't experience
> this as mine did.  Could it be that you have no devices in your PCI slots
> and I have two SCSI cards?

No, but I'm using OFW_NEWPCI routinely on my boxen and did not measure
build times without this option.

	- Thomas

-- 
Thomas Moestl <t.moestl at tu-bs.de>	http://www.tu-bs.de/~y0015675/
              <tmm at FreeBSD.org>		http://people.FreeBSD.org/~tmm/
PGP fingerprint: 1C97 A604 2BD0 E492 51D0  9C0F 1FE6 4F1D 419C 776C