Subject: Re: NVRAM/OpenBoot2.6 Power-On Self Test FAILED
To: beaker <beaker@freeshell.org>
From: Johan A. van Zanten <johan@brahma.giantfoo.org>
List: port-sparc
Date: 10/14/2002 07:11:23
---In message <Pine.NEB.4.44.0210140714100.13624-100000@otaku.freeshell.org>

>I've done this already - the Host ID and Ethernet Address are set
>correctly and the system DOES boot normaly, so whatever the problem is, it
>doesn't seem to have any real impact on anything once I'm up and running.
>I suppose I could just keep typing "b" each time I do a cold boot, but I'd
>prefer to find out whats going on and "fix it.

 I had a problem with some Mouser M48T02 200ns chips i bought to replace
failed chips in SPARC 2s. The SPARC 2s had ROM rev 2.4 and 2.9 and. The
new Mouser chip seem to take (and remember) the new hostid and Ethernet
address, but there was some sort of problem in the TOD area that prevented
the machine from auto-booting:

WARNING: Unable to determine keyboard type
    FAILURE: TOD (f20007fa) Bit Failure, Exp = 000000ff, Obs = 0000007fscreen not found.
Can't open input device.
CPUengine/SS2, No Keyboard
ROM Rev. 2.4, 64 MB memory installed, Serial #12648430.
Ethernet address 8:0:20:13:de:ad, Host ID: 55c0ffee.


Power-On SelfTest FAILED ... Replace NVRAM


 I tried 4 different (new) MK48T02 chips in two different SPARC-2s (one's
a clone), with the same results in each situation.

 However, the SPARC 1+ had no trobule with the same chip.

If i remember correctly, googling, i came across a post by someone
suggesting that Mouser had made some changes to the chip for y2000
compliance, and this caused the Mouser chip to misbehave with some SPARCs.

 From MemoryX (memoryx.com), i purchased a Dallas Semiconductor's chip to
use instead of the Mouser (sorry, i don't have the model # handy), and the
SPARC 2 was happier.

 --johan