docs/62402: easily circumventable Blade150 problem

Lodewijk Vöge lodewijk at cope.nl
Thu Feb 5 21:11:28 UTC 2004


>Number:         62402
>Category:       docs
>Synopsis:       easily circumventable Blade150 problem
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    freebsd-doc
>State:          open
>Quarter:        
>Keywords:       
>Date-Required:
>Class:          change-request
>Submitter-Id:   current-users
>Arrival-Date:   Thu Feb 05 13:10:16 PST 2004
>Closed-Date:
>Last-Modified:
>Originator:     Lodewijk Vöge
>Release:        5.2-RELEASE
>Organization:
>Environment:
FreeBSD Blade 5.2.1-RC FreeBSD 5.2.1-RC #3: Thu Feb  5 13:30:14 CET 2004     root at Blade:/usr/src/sys/sparc64/compile/BLADE  sparc64

>Description:
This is a request to append a note to the documentation for sparc64, and a way to get the solution into Google in any case.

We had installed 5.2-RELEASE on a Blade 150. it would work okay for a while, but would reliably die during a cvsup with a

IOMMU fault at virtual address 0xC3000000                                       
Uncorrectable DMA error AFAR 0x696000 AFSR 0x210000ff000000000

in the end the solution turned out to be to unplug the USB keyboard. I had found one message in the archives indicating the Blade150 needed to be installed without the keyboard because it wasn't supported at all, but this appeared not to apply: the machine installed fine and works for a while before panicing, always under network load. But, as it turned out, that was it.


>How-To-Repeat:

>Fix:

Add a note to the sparc64 documentation that you really shouldn't have the keyboard plugged in a Blade150 even if it appears to work fine at first sight.
>Release-Note:
>Audit-Trail:
>Unformatted:



More information about the freebsd-doc mailing list