Re: GRUB bug

[Derrik Pates]

On Fri, 12 Jan 2001, OKUJI Yoshinori wrote:

> You have't still told us how you booted Linux and what GRUB said after
> each command execution.

Ok. hd0 on the system I'm testing on is a Compaq SMART3200 PCI RAID
controller, with 5 Ultra2/Wide 18 GB SCSI hard drives hanging off it in a
RAID-5 configuration.

These are the commands I used to boot it:

grub> root (hd0,1)
 Filesystem type is ext2fs, partition type 0x83

grub> kernel /boot/vmlinuz root=/dev/ida/disc0/part2 ro
   [Linux-bzImage, setup=0x1200, size=0xbd262]

grub> boot

I also tried calling 'impsprobe' to detect and initialize the 2 CPUs, but
that had no effect.

> This is not what you judge; If you know what is important, you have no
> reason to ask us what is wrong. Don't omit any facts, which is a
> general rule when you report a bug.

I try to think of what's relevant to the problem I'm seeing. I doubt
things like drive geometry are an issue considering I can boot just fine
from lilo - the problem is that, on these systems, something is happening
that causes kmem_alloc_cache() to return an obviously-wrong pointer when
grub is used to boot the machine. I could (certainly) provide you with
more info than you can stand, but you don't need that - that's why I'm
trying to provide just the essentials.

Derrik Pates      | Sysadmin, Douglas School|    _   #linuxOS on EFnet
address@hidden |  District (dsdk12.net)  |   | |   and now OPN too!
   Student @ South Dakota School of Mines   | __| |___ _ _ _   ___ _ _   ____
       & Technology (www.sdsmt.edu)         |/ _  / -_) ' \ '\/ _ \ ' \ (____)
UNIX: Because you want to USE your computer.|\___,\___|_||_||_\___/_||_|