[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#80325: Installer misreports partition numbers

Christoph Paul <cpaul@augsburg.netsurf.de> writes:

> > > > Still kernels should have BSD disklable support as well i think
> Its neccessary for Solaris.
> > > I would be more comfortable if you checked it against the kernel (or
> > > fdisk, I guess) sources.
> > extended partition will always follow primary partitions, following this
> > will be BSD or solaris on a first come first serve basis.
> > The previous patch i did always put BSD before SOLARIS which is wrong
> But not quite- its ok with BSD 
> But look at this dmesg output:
> Partition check:
>  hda: hda1 hda2! hda3 <solaris: [s0] hda5 [s1] hda6 [s2] hda7 [s7] hda8 > hda4 < hda9 hda10 hda11 hda12 hda13 > < hda14 hda15 hda16 hda17 >                                                              
> I means what you see: BSD slices at the end, sol. after partition. 

Um, hum. According to the Linux+FreeBSD HOWTO, the slices and such are
numbered according to the order in which the primary partition that
holds them occurs.

Thus if you are on i386 and your DOS extended partitions are off hda4,
but you had *BSD partitons off hda3, then it would look something

  ... hda3 < hda5 hda6 hda7 > hda4 <hda8 hda9 hda10>
             ^^^^^^^^^^^^^^         ^^^^^^^^^^^^^^^
             *BSD partititions     extended partitions

> fdisk calls it /dev/hda7, mount /dev/hda11. 

Yah, not much I can do about that.  We've documented it at least.

So long as the kernel and dbootstrap agree, I'm happy.

.....Adam Di Carlo....adam@onShore.com.....<URL:http://www.onShore.com/>

Reply to: