gr@eclipsed.net

>I got the following privately, and am posting it here with
>Emmanuel's permission:
>
>On Tue, Apr 17, 2001 at 09:59:37AM +0200, Emmanuel Pineau wrote:
>> I'm trying to install your snapshot on a G3 B&W OF3. I use
>> netbsd.ram.gz the installation kernel that contain all devices. It
>> is very interesting for me because it's the only installation kernel
>> that recognize my two IDE hard disks. But I have a problem. The
>> system boot correctly and then stop after displaying "Terminal
>> vt100?". I hit the shift key then return key but nothing happen.
>> What can I do? Is there a problem with USB device?
>
>Basically, I'm not fortunate enough to have had this problem. (I
>have no B&W machines available for testing, rather just aging 7x00s
>and 6500s.)
>
>I *think* this USB-dies-after-boot thing is a live issue, but I'm
>not sure, and I haven't really been paying attention. (Remember, I
>have no machines with USB, so I couldn't be working to fix the
>problem even if I were paying attentiont to it, so...) Anyone care
>to comment? Bob Nestor?
>
>Another suggestion I made for Emmanuel is that he could try a more
>recent (1.5U) snapshot I've made (but have not yet used even on my
>development machine, though I don't anticipate problems and, as it
>turned out, didn't even have to lobotimize the floppy to get it to
>fit). This snapshot is now posted on http://eclipsed.net/netbsd/.
>(It's the 20010404 snapshot.)
>
>Bob, when there are reports of this snapshot working at least as
>well as the 20010323 one, 20010404 should probably replace 20010323
>on the ftp site, since it has a complete boot.fs. (I may put this
>snapshot on my devel machine later this week. I'll let you know if
>no one else has reported to port-macppc before then.)

This is a long-standing problem with some of the early model G3s and 
iMacs.  This happened late last year when some changes went in to clean 
up USB and/or console handling.  Until recently there were two problems; 
no keyboard during boot and no keyboard after boot.  The first problem 
was recently fixed in -current but the second hasn't been found and fixed 
yet.

Those of us who have systems that can't run 1.5, 1.5.1 or -current are 
still able to run 1.4.3A I think, at least my early G3 does.  I can now 
build -current kernels and test changes but have no idea what is wrong or 
where the problem might be.  I've tried enabling some of the debugging 
stuff but it didn't help much.

If anyone has some ideas of what to try I'd be happy to try them out, but 
like most others in this situation I'm at a loss as to what to do to fix 
the problem.

-bob