[ltp] cloning 14G drives

[Mark Kennedy]

Mark Kennedy wrote:
> 
> Rob Mayoff wrote:
> >
> > | Seems like I hear a lot of suffering associated with 14G drives.  I
> > | wonder if there is a particular model that just sucks w.r.t. linux?
> > | These two drives: are they identical models?
> >
> > The problem with the 14 GB drive is that it does not correctly identify
> > itself to the system. The standard way for a drive to tell the system
> > that it's larger than 8 GB is to claim that it has a raw geometry of
> > 16383/16/63. The IBM 14 GB drive claims to have a raw geometry of
> > 16383/15/63. The kernel doesn't recognize this as the magic "big disk"
> > geometry so it assumes the drive is reporting its actual geometry.
> >
> > ----- The Linux ThinkPad mailing list -----
> > The linux-thinkpad mailing list home page is at:
> > http://www.bm-soft.com/~bm/tp_mailing.html
> 
> ok, between your last message, section 11 of the large disk HOWTO,
> and the output of hdparm, i'm beginning to understand.
> 
> so i should specify a logical CHS geometry of 16383/16/63 (instead
> of 1826/240/63) to trigger the large disk support and everything
> else will just work because the disk driver uses logical block
> addressing (ignores the CHS geometry) and the number of logical
> blocks reported by the disk (27,609,120 in this case or 13.1G)
> is correct?
> 
> if so, then bill needs to change the large disk hint on his
> linux/thinkpad page.
> 
> /mark
> ----- The Linux ThinkPad mailing list -----
> The linux-thinkpad mailing list home page is at:
> http://www.bm-soft.com/~bm/tp_mailing.html

i tried the dd to /dev/null on the train home last night with
16383/16/63 and it still stopped at the 8G barrier.  so there
is definitely a kernel or a kernel version dependence.
more when i discover it...

/mark
----- The Linux ThinkPad mailing list -----
The linux-thinkpad mailing list home page is at:
http://www.bm-soft.com/~bm/tp_mailing.html