Comment 31 for bug 10712

Jean-Marc, this make it a processor without NX bit, so it should not be affected by the bug I was speaking earlier.

Also I found this information, that I am not sure is relevant:
"Starting with 2.4.19 and continuing in 2.4.20-pre8, the order the kernel
associates with the two IDE controllers (one VIA vt8233 and one PDC20265 intended for RAID use) on the A7V266-E has been reversed. The BIOS and GRUB consider the VIA to be first, so root(hd0,0) loads the kernel from the first device on the VIA controller. Prior to 2.4.19, the OS then booted with that drive identified as hda. Beginning with 2.4.19, however, the kernel instead identifies the PDC as ide0 and ide1, and puts the VIA at ide2 and ide3, resulting in the boot drive being hde.
I found an earlier mention of this on the mailing list, but no solution or workaround was suggested. We are using a workaround where 2.4.19 and later kernels are booted with root=/dev/hde1 and earlier with hda1, and fstab lists both hda2 and hde2 as swap partitions, simply failing to insert one.
This works, but the general ugliness and maintenance headaches since this is different than the typical machine config we use around here make it difficult to use in the long run. "