Messages in this thread | | | Date | Sat, 4 Dec 1999 17:23:36 -0800 (PST) | From | Robert Dinse <> | Subject | SS-10 -> Ultra Question... |
| |
Well, those of you who have seen my postings before are probably familiar with my Sparc-32 SMP stability problems.
To summarize, 4/670MP with quad Ross Hypersparc CPU's, 384MB of RAM, and a 10-base-T/SCSI combo card runs totally stable under 2.2.13 as long as I don't try to turn up two ethernet interfaces.
SS-10 on the other hand won't run for more than a few hours before spin_lock hell. This is moving the exact same CPU modules, video card, and SCSI controller that are completely stable in the 4/670MP.
Today I tried two things; I got an SS-20 chassis and tried that. Unfortunately, same spin_lock() hell, it seemed to happen even faster actually, longest uptime was 59 minutes.
When I've bothered David about this before, answer I got was "Get an Ultra".. Ok, maybe that's truncated a bit but that's what it amounted to. So I got an Ultra-1. Probably not a screamer but if I can find a stable platform that doesn't raise the room temperature ten degrees I'll be elated.
I was under the impression that the same kernel that would boot on an SS-10 would also boot on the Ultras. So I hooked up the disks that were connected to the 4/670MP / SS-10 / SS-20, to the Ultra-1 and attempted to boot.
The cursor spins but then before even getting the SILO prompt I get:
Fast Data Access MMU Miss
... and it returns to the PROM prompt.
I tried booting a RedHat 6.0 CD and it boots fine. But I don't want to wipe out everything I have and start over. Since it doesn't even get to the SILO prompt I assume this is SILO related?
Can anybody tell me what I need to do to get this machine to boot and run or do I have to re-install Linux and recompile all of my software?
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |