[rescue] Sparc 20 Diag question

[Mike Spooner]

It's worth checking that the lowest-numbered SIMM slot is populated. SS10 and SS20 are fussy about that. You can leave any other slot(s) empty, but not the lowest-addressed one. Also, the traditional "Re-seating of the SIMMs" maneuver may be wise.




-- Mike





On Wed, Dec 11, 2019 at 4:04 PM +0000, "Mouse" <mouse at Rodents-Montreal.ORG> wrote:










> Good news was when I pulled off the cover, there was a 4M VSIMM in
> the machine :-)

Score!

> Bad news is that there's a failure on diags that I can't find info
> on.  It begins diag as it comes up, and then I get to:

>  MXCC Non-Cache Block Zero Test
> CPU_00000000 >> error: target memory
> asi 20, addr 00010004, exp 00000000, obs 00e90000

According to the NetBSD include files, ASI decimal 20 is "flush i&d
cache line (user)" and ASI hex 20 is "sun ref mmu bypass, ie. direct
phys access".  I suspect the latter is more likely what's going on.

0x00010004 is just over 64k, so it's probably in the lowest-addressed
stick of memory.  I don't recall which that is, and at the moment I
don't have a 20 I can open up to check.  I daresay others here will
have that info in their heads.

> I'm bringing this up here since I can't seem to find any reference to
> these tests and what they actually mean or point to.  Then I may be
> able to narrow down my troubleshooting.

Yeah, it's annoying.  I have no reference for them either; I'm just
guessing based on what you quoted.  I suspect that those diagnostics
were never documented for customers; if so, all the info that's
available today was either leaked or reverse-engineered.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse at rodents-montreal.org
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue