Red Hat Bugzilla – Bug 35820
lilo can't parse append="mem=256M"
Last modified: 2007-04-18 12:32:39 EDT
From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows NT)
lilo complains about the line
removing the second '=' sign does not produce the error message,
but is incorrect.
Putting spaces around the first '=' sign doesn't help.
Steps to Reproduce:
1.Edit lilo.conf to include append="mem=256M"
3.Will fail, giving the line where the above occurs
Actual Results: MBR got corrupted, now boots with just LI
Expected Results: Should work like it did with SuSE :-)
At least shouldn't have fried MBR.
Ran up2date, got a little newer lilo, but still has this problem.
Reinstalled woverine from CDs, problem disappeared.
Thinking about "what changed", I think that this (mis)behavior may be
manifesting itself when either
1. There is <=16MB of RAM installed or
2. There is more physical RAM than what either BIOS or kernel sees, and perhaps
the amount that either BIOS or the kernel sees is 16MB or less.
Could somebody check this? That is, feed mem=16M as a boot parameter
to LILO as a system boots up, then see if it still can parse
append="mem=<whatever>" correctly. I currently don't have a system that I'm in
a hurry to destroy. . .
FWIW, I can't see this happening at all with my 128 MB system (RHL 7.1), whether
I specify 16 or 128 MB of memory.