Date: Fri, 9 Feb 1996 14:36:02 +1100 From: Bruce Evans <bde@zeta.org.au> To: hsu@clinet.fi, joerg_wunsch@uriah.heep.sax.de Cc: bde@zeta.org.au, freebsd-bugs@freebsd.org Subject: Re: bin/984 Message-ID: <199602090336.OAA18315@godzilla.zeta.org.au>
next in thread | raw e-mail | index | archive | help
> > > Actually it should be possible to fix it by running fdisk to set the new > > > C/H/S values. > > > > Ok, this might be possible. Anyway, the least we would need to know > > is whether my suggested manual fix did really fix the problem. > > ... >... >About the suggested fdisk fix, if C/H/S values are changed wouldn't this >break the partition limits, or are the partitions encoded just as block >offsets? The partition limits are encoded as both block offsets and as C/H/S values, and of course you should not change the offsets. > > (No, i'm not going to buy just another motherboard to see your > > problems fixed. Even if i did -- i wouldn't had the problem in the > > first place, or whaddaya think why i've been inventing the > > ``dangerously dedicated'' mode? :) For ideological reasons. >Install an Intel Plato system in dangerously dedicated mode with BT ISA >controller and replace it with a NCR PCI controller, and it may/will not >boot (too large cylinder number 1040 > 1024, BIOS limit). This is actually >quite weird, as the root partition is about 200Mbytes which should be below >1024 cylinder, no matter what geometry is used. This isn't that serious, With a geometry of 1 sector/track and 1 track/cylinder, anything larger than 1MB will be above the 1024 cylinder mark. Perhaps the NCR controller is confused about the geometry. The BT445C controller guesses the geometry from the partition table so it is safest to have a non-garbage geometry there. Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199602090336.OAA18315>