Bug#301478: [sparc][rc3][netinst] installation fails on u1/140

2005-03-27 Thread Vincent McIntyre

A bit more digging on the gentoo lists reveals this:

http://forums.gentoo.org/viewtopic-p-1766279.html#1766279
   
   The problem is basically related to the recoding of isofs starting at
   silo 1.3.0, it seems some kit (newer specially) doesn't like the new
   iso first. This affects up to the latest silo 1.4.8 (latest at the time
   i'm writing this). Using silo 1.4.8 doesn't cause problems for regular
   (i.e. hard disk) boot AFAIK, at least if you don't use an initrd and/or
   iso filesystem.
   

posted 2004.11.15.
However I tried this release and still have the same problems, see
my previous message.


David S. Miller says there is an issue around the kernel size -
it needs to be less than 4Mbyte ?
 http://marc.theaimsgroup.com/?l=linux-sparcm=109892515109346w=2
 http://marc.theaimsgroup.com/?l=linux-sparcm=110222004312949w=2

A bit more googling turned up these (in relation to solaris booting) -
 http://unix.derkeiler.com/Newsgroups/comp.sys.sun.admin/2004-05/0138.html
 http://unix.derkeiler.com/Newsgroups/comp.sys.sun.admin/2004-05/0107.html
These messages suggests the problem is the architecture that SILO is built
with.

Don't forget that the CPUs in these old (200MHz) boxes have some errors
in them which prevent them from working perfectly in 64-bit mode. In
32-bit mode they are fine however.

Where is SILO built for sarge - on what specific sparc platform?

Cheers
Vince





-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Bug#301478: [sparc][rc3][netinst] installation fails on u1/140

2005-03-26 Thread Vincent McIntyre
Package: installation-reports

Debian-installer-version: RC3 netinst
 
http://cdimage.debian.org/pub/cdimage-testing/sarge_d-i/sparc/rc3/sarge-sparc-netinst.iso
  md5 sum 95042ee7c74ce5a512ecaba41cea225d
uname -a: unable to boot. attempted to boot 2.4.27 (default)
Date: 26 Mar 2005
Method: Boot from netinst cdrom

Machine: Sun Microsystems Ultra1/140
Processor: UltraSparcI
Memory: 128M
Root Device:
Root Size/partition table:
Output of lspci:

Base System Installation Checklist:

Initial boot worked:[E]
Configure network HW:   [ ]
Config network: [ ]
Detect CD:  [ ]
Load installer modules: [ ]
Detect hard drives: [ ]
Partition hard drives:  [ ]
Create file systems:[ ]
Mount partitions:   [ ]
Install base system:[ ]
Install boot loader:[ ]
Reboot: [ ]
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it

Comments/Problems:

 Boot machine into solaris2.6, everything is ok.
 # halt
 ok setenv auto-boot? false
 ok reset-all
 [reboots]
 ok boot cdrom
 boot: return
 SIL0 1.4.9
 [ one line about booting vmlinuz, then... ]
 Fast Data Access MMU miss
 ok

 I tried a few more times, same result each time.

I thought I would just try. If this ancient h/w is no longer supported,
please close this report.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Bug#301478: [sparc][rc3][netinst] installation fails on u1/140

2005-03-26 Thread Steve Langasek
clone 301478 -1
reassign -1 silo
severity -1 important
title -1 Fast Data Access MMU Miss on Ultra 1/140
merge -1 261824
thanks

Hi Vince,

On Sat, Mar 26, 2005 at 08:06:53PM +1100, Vincent McIntyre wrote:
 Debian-installer-version: RC3 netinst
  
 http://cdimage.debian.org/pub/cdimage-testing/sarge_d-i/sparc/rc3/sarge-sparc-netinst.iso
   md5 sum 95042ee7c74ce5a512ecaba41cea225d
 uname -a: unable to boot. attempted to boot 2.4.27 (default)
 Date: 26 Mar 2005
 Method: Boot from netinst cdrom

 Comments/Problems:

  Boot machine into solaris2.6, everything is ok.
  # halt
  ok setenv auto-boot? false
  ok reset-all
  [reboots]
  ok boot cdrom
  boot: return
  SIL0 1.4.9
  [ one line about booting vmlinuz, then... ]
  Fast Data Access MMU miss
  ok

  I tried a few more times, same result each time.

Thanks for the report.  This seems to be an instance of bug #261824, which
has previously been seen on SunFire 280R machines and Sun Ultra1s.

Is there any chance you could try booting an RC2 CD on this machine as well,
just for the sake of confirming whether this is a regression?  (It probably
isn't, but the more information we have, the better.)

Cheers,
-- 
Steve Langasek
postmodern programmer


signature.asc
Description: Digital signature