Re: Not a bug but an enquiry about CHS

[roger]

Thanks for your advice, I went ahead and partitioned and all turned
out well :-) I'm very impressed with (parted), it really is
user-friendly. One point, when I ran mkpartfs on the new Linux
partition, I found that the ext2 filesystem hadn't been created (or
wasn't recognized by Linux), and I had to use mke2fs later to complete
the job...

Just a further question about CHS: the kernel CHS (2480/255/63) is the
"logical value". Linux also gives the "physical CHS" as (16383/16/63)
which is the same as that given by the Seagate diagnostic program 
(my HDD is Seagate 20Gb). So what is the BIOS CHS (39535/16/63)
measuring? How can the cylinder count be so different? Is this
something to do with LBA?

Regards,
Roger Young. 
 


.............................................................................
Roger Young wrote:
> 
> In the documentation it is recommended that the kernel CHS (2480/255/63)
> should
> be in sync with the BIOS CHS (39535/16/63) by changing the former. When
> I do this
> I get an error "Partition does not end on cylinder boundary".
> Furthermore
> a diagnostic program shows a third CHS (16383/16/63) for the drive.
> There is some
> confusion here. Should I worry about it, or just proceed and hope for
> the best?

I would "proceed, and hope for the best"...  BTW: the BIOS is always
right TM.  (The disk might say something different to what the BIOS
says, but this is irrelevant.  It's the BIOS that talks to boot
loaders, etc., not the disk)

Having the geometry not matching, etc., might screw up boot loaders,
or DOS/Windows, if they're not using LBA.  (In both cases, you should
enable LBA)

Andrew Clausen