PE> I went to look at my partition table with linux, and as well
PE> as giving me strange numbers for my partition sizes, it has also
PE> determined that the number of cylinders on the HD is 4969, which is
PE> what the BIOS setup determined it to be, but since I have LBA, isn't
PE> all this meant to have been remapped, so it can't tell the difference?

Thats a bit complicated. The short story is that with LBA enabled,
yes, commands down the ribbon cable to the drive are purely in terms
of the logical block number. *BUT* with plain booted DOS, *DOS* doesnt
know about that stuff, so the bios fakes it up so that DOS can still
see cylinders/heads/sectors that it understands. The bios then maps
those into the LBA block numbers for commands down the cable.

Now superficially that aint relevant for Linux, but the trick is that
WHILE BOOTING, you are using the bios to boot whatever is bootable, at
that stage it doesnt even know WHAT its booting, and that stuff is done
at the very low level, initially just looking at the first physical sector
on the drive where the partition table and master boot record reside.

You also have the question of whether the boot manager is operating in LBA
mode and other complexitys like use on a motherboard which does not support
LBA mode in the motherboard bios, but where the drive is used in LBA mode
once you are into the boot phase, like once the boot manager is in control etc.

Gets surprisingly complex.

PE> There is a possible problem that LBA is not actually doing a bloody
PE> thing on my system.  How can I confirm whether LBA is activated properly?

Even thats rather tricky because you need to be able to
power up the system and effectively tell the drive to operate
in LBA mode, which it wont normally have started up in.

There should be some utes around which can tell you that stuff,
but I cant tell you of one that does that specifically, I have been
deferring a full comb thru the net for the best currently around until
the local call charge internet was fully up here. Havent done it yet.

The Begley had chased some of that stuff up, but more for SCSI from
memory. Presumably he had noticed what the IDE info was tho too.

PE> BTW Rod, you asked me why I had 5 meg partitions.
PE> I actually define them and then delete them, so that I can
PE> add hard drive letters easily if my configuration changes.

OK, that does make some sense.
@EOT:

---