--Signature_Mon__2_Jan_2006_19_05_24_-0500_5yIq.Z/fw6qWzgW4
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

Hello,

> >>>> NMI: system interrupts: 40000000<VME=3D0,SBUS=3D0,ME>
> >>> Wasn't that the MicroSPARC's way to complain about RAM parity
> >>> errors?
> >> Was it?  I don't know.  The M bit is not set (SINTR_M); that seems
> >> to be the memory error indicator, based on the code - or is that
> >> only for machines with real ECC, not just parity?
> > My SPARCbook ( which has a MS-II ) spewed errors like that when
> > something touched the right bits in a faulty SIMM ( and never did it
> > again after I replaced them ).  I'm just not 100% sure it's
> > /exactly/ the same number ( only 98.5% or so ;) )
>=20
> Is there anything akin to memtest86+ for SPARCs?  I did a little
> poking around and I saw some indications that some of the Linux people
> have developed something like that for Linux; anyone know of anything
> similar either standalone or NetBSD?

Don't think so. On my sparcbook the firmware memory test was kind of
unreliable, sometimes it found something on startup, sometimes not. But
after such an NMI it /always/ found the same bit error at the same
address.

So, I'd recommend setenv selftest-#megs and test-memory after such an
NMI.

have fun
Michael

--Signature_Mon__2_Jan_2006_19_05_24_-0500_5yIq.Z/fw6qWzgW4
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (NetBSD)

iQEVAwUBQ7m/xMpnzkX8Yg2nAQIrpgf+II0anVKH9nF9piYf7oTOoBW7hjzcYzPJ
LYcjPbt9iixt+Dpv+IUkVg+CQ8gw8ZqDz54nZMzzw43pS9/diA0t+P66LCZOQm0T
T0OQ2oBfK6gBAmZyvu8lx5toczdl4f5tuiIpOPbeAw93bkUDuN5b+UYGpefmiB5B
EQ24tHYrd458zQ9scSmY6Tf2h/rRK28d8GWgEfvCOXlWPF5FNmOePAnZvY+nvWbL
zRDjc0RPPx/jjbM9nStZC2F54Ccn3ka4Tnm8SP/E8E5XFn0I4u2of2+Z9+ZrP5ea
733lh892y/Kk67a/XbPCEE7sqi1+9p/ls2fUXU9LRkYlupt2CTgiJw==
=qKv5
-----END PGP SIGNATURE-----

--Signature_Mon__2_Jan_2006_19_05_24_-0500_5yIq.Z/fw6qWzgW4--