Re: Bug#243120: boot failure on sparc32 (Ultra1)
In article <[🔎] Pine.SOL.4.33.0410260933300.4964-100000@mayhem.atnf.CSIRO.AU>
Vince.McIntyre@atnf.csiro.au writes:
>
>I finally made some more tests, with netinst 20041024
>(http://cdimage.debian.org/pub/cdimage-testing/daily/sparc/20041024/sarge-sparc-netinst.iso)
>SILO 1.4.8
> Fast Data Access MMU miss
Please try the netinst iso image in http://people.debian.org/~blarson/
and let me know if that works or not.
>A note on the 32/64 question. The sparc chip in these boxes is indeed
>64-bit. However apparently there were some errors in the chip design
>that forced Sun to make Solaris run in 32-bit mode.
>Maybe linux does not have this problem?
The problem is with undefined instructions causing the processor to
hang. It affects any OS with 64-bit user mode and allows any user to
crash your system.
>Or maybe there is a PROM setting that needs tweaking, to force 64-bit?
No. 64-bit solaris should run on them with a minor tweek to it that
disables the disabling code. :-) (Search the archives of the
appropriate usenet newsgroups, I don't remember the exact tweek.)
--
Blars Blarson blarson@blars.org
http://www.blars.org/blars.html
With Microsoft, failure is not an option. It is a standard feature.
Reply to: