Subject: Re: i386 1.5_BETA binaries available
To: Bang Jun-Young <bjy@mogua.org>
From: Manuel Bouyer <bouyer@antioche.lip6.fr>
List: port-i386
Date: 10/23/2000 21:28:00
On Mon, Oct 23, 2000 at 10:07:27PM +0900, Bang Jun-Young wrote:
> > Please test them, especially the installation floppies.
> > The install documentation still needs some updating to
> > match the world today, but if you have comments on it,
> > they are welcome.
> 
> I've just installed 1.5_BETA. 
> 
> It could be booted from CD-ROM drive without problem. But
> I have found some problems during installation:
> 1. Try dmesg | more when you're exited from sysinst. It prints innumerable
> ^@ characters before the actual dmesg.
> 
> 2. Reported drive cylinder value is not the same as the real one. Look at 
> this:
> ...
> wd0 at pciide0 channel 0 drive 0: <QUANTUM FIREBALLlct10 30>
> wd0: drive supports 16-sector pio transfers, lba addressing
> wd0: 28629 MB, 16383 cyl, 16 head, 63 sec, 512 bytes/sect x 58633344 sectors
>                ^^^^^ 
> wd0: 32-bit data port
> wd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 4
> ...
> 
> I know, of course, it's not a bug since the maximum logical cylinder
> value is 16383 for drive which has capacity larger than 8.4GB.  
> But on the other hand, FreeBSD and Linux reports the real values. 

There is no "real value" as disks have variable zone records now.
I'd prefer to keep the values reported by the drives here.

> 
> 3. Sysinst has the similar problem. It sets the cylinder limit to 1024 
> no matter what the real value is.  

For the BIOS geometry this is normal, the MBR doesn't support more.

--
Manuel Bouyer <bouyer@antioche.eu.org>
--