In message <20021120071743.28899ddd.ggm@apnic.net>, George Michaelson writes:
>
>I have an Archos MP3 player, using the ISD-200 USB-ATA bridge.
>
>When I put this into my NetBSD-current box, it failed with a mass
>of BBB failures. Scanning mailing lists, I see this as a very common failure
>in NetBSD with random CF devices and other things emulating scsi disks.
>
>When I don't get BBB failures, I get I/O errors on the mount request even
>if I do see a disklabel.
>
>On the advice of the rockbox Archos hackers, I added USB debugging, and I
>found that my kernel now recognized the device.
>
>Does this help identify umass_quirks or scsi_quirks which relate to
>time-critical behaviour? I posit/hypothicate that the kernel debug printfs
>slow something down enough to make USB happier.
>
>the device is USB2.0, the PC is a Dell L400, USB1.0 only.
>

Based on some advice I had received, I tried deleting the atapibus
over USB line.  But I've seen the failure since then.

I wouldn't be too convinced that your problem is solved.  For me, at 
least, it's a heisenbug.

		--Steve Bellovin, http://www.research.att.com/~smb (me)
		http://www.wilyhacker.com ("Firewalls" book)