On Thu, Nov 11, 2004 at 06:33:32PM +0200, Frank Wille wrote:

> My main problem is that many (most? all?) NetBSD programs tend
> to cause a segmentation fault.

I have seen this on a r4000 Indigo and a r5000 O2.  I _strongly_ believe that
the problem is caused by CPU-specific cache botches in/around pmap.c -- swapping
the CPU modules with r4400 and r10000 make the problem go away.

>   Checking for botched superblock upgrades
>   Reparing partition /dev/rsd0a
>   Alternate super block location: 16
>   DOWNGRADING TO OLD SUPERBLOCK LAYOUT

Interesting.  Are you sure this drive is good?

> - The NG1 driver places the cursor one position to the left
>   besides the correct one. When I type a character I can't see
>   it, because the cursor will appear immediately over it. ;)

You have a revision 5 board.  From newport.c:

                /* Work around bug in some board revisions */
                if (dc->dc_boardrev < 6)
                        x_offset = 21;
                else if (dc->dc_vc2rev == 0) 
                        x_offset = 29;
                else
                        x_offset = 31;

My rev 4 NG1 works with the 21-pixel offset.  Perhaps it should be used for
board revisions less than five rather than six?

> - "poweroff" works, but the Indy's shutdown-melody is destroyed.
>   I hear a loud random noise.

Very odd.

-- Chris
	GPG key FEB9DE7F (91AF 4534 4529 4BCC 31A5  938E 023E EEFB FEB9 DE7F)