RH 6.1, 2.2.12.. Strange mem=nnnM behavior!

[Paul Klissner <paulk@Eng.Sun.COM>]

I realize I only have a tier 2 support board, but I'm getting strange
behavior.  Hopefully you know something about this (or would like to)
and can steer me in the right direction.

Configuration:
        AMD 'Mainboard' mother board, 200Mhz bus...
        AMD 550Mhz Athlon processor
        2 x 128 Mb DRAM ECC
        RedHat Linux 6.1, 2.2.12-nnn

  First, I should mention, I've installed Linux and virtually everything
  works fine... sees my disks, compiled the drivers and used a SBLive!
  card, used the parallel port to talk to an HP Laserjet.


  The problem is, I can't make it utilize all 256Mb, it crashes.
  It only recognizes 64Mb by default (free command), I can up it
  to 128Mb with the mem=nnnM  either from the LILO boot prompt
  or in the lilo.conf file.

  When I start going upwards, the system crashes somewhere in
  the boot process, invariablly.  For example mem=180M or mem=256M
  will crash.

  Note that when I tried 3 x 64 Mb of RAM and tried mem=192M it
  crashed too, during boot.

  But mem=128M seems to work fine.

   Any ideas?

  Thanks,
  Paul



-
Linux-mm-www: builders list for the Linux-MM website
Archive:      http://humbolt.nl.linux.org/lists/
Web site:     http://www.linux.eu.org/Linux-MM/