Kernel panic in DSL

[cornhenge]

I'm trying to bring up Damn Small Linux on an older i386 with 128 MB RAM, using GRUB. The kernel seems to panic, and prints out several addresses like "[<c01c0880>]" to screen before saying it's "Unable to handle kernel paging request at virtual address
c8000000.":

(...) [<c020c7fd>]<1>Unable to handle kernel paging request at virtual address c8000000
 printing eip: 
c0108bf8
*pde=00000000

I did some digging and found that the `eip' value actually coincides with EOF for the kernel, if that makes any difference, and I've been able to reproduce this problem with the following commands passed to GRUB:

kernel /boot/linux24 --no-mem-option root=/dev/hda2
kernel /boot/linux24 mem=128M root=/dev/hda2

My swap partition (hda1) is 500MB; hda2 contains the boot image (well, hopefully) along with an ext2fs filesystem. I'm specifying `mem=' explicitly on the second line because GRUB _may_ be reading from grub.conf and setting mem=16M.

My machine seems to be in an unbootable state, since neither my SuSE or Slackware live CD can bootstrap the system (there's just a blinking cursor). fdisk is unavailable. I can't spawn init (runlevel 1,2, 3, 4, or 5), period. Any suggestions?

[cornhenge]

Solved.

I was able to use a SuSE live CD to overwrite the partition table with the same boundaries that I wrote in my install diary prior to the panic. This was enough to put DSL in a bootable state again, but the problem is I don't know why this worked. Any clues?