[vox-tech] bad partition table

[Josh Parsons]

On Mon, 2005-02-14 at 16:19 -0800, Nick Schmalenberger wrote:

>  The whole geometry issue just confuses me.

As I understand it, you should be able to ignore all this stuff on a
modern linux system with a modern disk. I'm not sure what's causing your
problem, but I can tell you a bit about the different kinds of output
from hdparm:

hdparm -g is what linux thinks programs like fdisk or lilo, which need
to interoperate with the PC BIOS, should take the disk geometry to be.
It probably bears no relation to the physical geometry of the disk, and
it's munged in various ways (for example, "255 heads 63 sectors" is a
conventional way of saying LBA is in use). Very meaningless numbers.
This value isn't used in normal operation with a modern drive.

hdparm -i gives you a lot of information collected directly from the
disk, including what the physical layout of the disk hardware is
("RawCHS"), and what the layout that should be used for traditional
(i.e. non-LBA) addressing on the IDE bus is ("CurCHS").  Neither of
these are used in normal operation with a modern drive, either.

If the drive supports it, linux will put it into LBA mode, and ignore
whatever geometry (mis)information it was given by the BIOS.  If this is
happening (and it should be) you will see "LBA=yes" as part of the
output of hdparm -i, and "/255/63" as part of the output from hdparm -g

-- 
Josh Parsons
Philosophy Department
1238 Social Sciences and Humanities Bldg.
University of California
Davis, CA 95616-8673
USA
 
Please avoid sending me Word or PowerPoint attachments.
See http://www.fsf.org/philosophy/no-word-attachments.html