Port-amd64 archive

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

Re: installation problem



Robert Elz <kre%munnari.OZ.AU@localhost> writes:

>     Date:        Fri, 3 Jul 2015 17:09:15 +0100 (BST)
>     From:        steve%prd.co.uk@localhost (Steve Blinkhorn)
>     Message-ID:  <20150703160916.5D2031E488%body.prd.co.uk@localhost>
>
> What's this supposed to be?
>
>   | 16 partitions:
>   | #        size    offset     fstype [fsize bsize cpg/sgs]
>   |  a: 976773167         1     4.2BSD   4096 32768     0  # (Cyl.      0*- 1033622*)
>   |  c: 976773167         1     unused      0     0        # (Cyl.      0*- 1033622*)
>   |  d: 976773168         0     unused      0     0        # (Cyl.      0 - 1033622*)
>
> I am fairly sure that you need to allow more than 1 block for the 2nd stage
> boot loader to fit.   Your other disc used the common, but meaningless, 63
> as the start offset for a & c partitions.   Better (though it might not matter
> on those drives) would be a power of 2 (or at least, a multiple of 8).  32
> should work fine.   

You do need more, nominally 14, but the 2nd stage loader goes in blocks
2-15 of the root filesystem partition.  Still, the use of offset 1 is
irregular and risks testing unused code paths, which is recommended only
if you want to encounter and fix new obscure bugs for the greater good.

In all seriousness, on x86, use 64 for a filesystem or raid at a these
days, in order to avoid misalignment of 4K native sectors.

Attachment: pgph8jccLjpYt.pgp
Description: PGP signature



Home | Main Index | Thread Index | Old Index