[ announce of BOOTX kernel ]
> > 
> The SCSI part works a lot better:
> 	Before (with BOOTFPE netbsd 1.1) i had trouble with a SCSI2 drive
> (Quantum lps52s), it was sometimes stopping with busy led still on, so the 
> kernel also stop. Now i've got error messages instead of stopping the kernel,
> there are like: dmascsi(1:0:0): ...data stuck in fifo...
Mmmm, yet another problem. I'll think about it. If I think I'm on
to something, I let you know.

If there are more people with scsi-problems, please let me know. The
following problems are known to me:
   * (General)
      Some (if not all) zip drives report data overruns when transfering
      large amounts of data. This also happens on the mac68k-port and 
      Allen Briggs is doing some tests on it.
   * (Falcon)
      Markus Kilbinger reports "spurious interrups" and sometimes systems
      hangs. I am currently trying to fix this.
   * (TT+Falcon)
      Also reported by Markus Kilbinger is the fact the sometimes all
      activity seems to stop. When you hit any key, the disks come to
      life again. I saw this once on my TT too. It seems like an interrupt
      is being missed.

> 	This disk is the only scsi2 in the chain (i've got also a Quantum LP40,
> a Conner CP30080, and a Maxtor 7120SCS).
> 
> 	But there is trouble with the mount_msdos option in fstab. This appear,
> in NetBSD 1.1B. The command execute, but the directories where the msdos disk
> are linked are empty.
Did you get the current.tar.gz file? It contains a new mount_msdos binary.
You also need to update your /etc/fstab to include the '-G' option.
Like:
       /dev/sd0g /transfer msdos rw,-uleo,-G 1 2

Leo.