Subject: NCR PCI problems still on 1.0
To: None <port-i386@NetBSD.ORG>
From: Michael L. VanLoon -- HeadCandy.com <michaelv@HeadCandy.com>
List: port-i386
Date: 03/15/1995 09:33:34
I posted a month or so ago about having problems getting the NCR810
PCI card to work well for me under 1.0.  Since then, a set of updated
files have been floating around.  I've been running the latest of
these for a week or two.  Lately, I've had some more problems.  I'm
getting media errors, which have caused the machine to panic twice.
Here's what they look like:

Mar 15 02:43:36 usenet1 /netbsd: sd2(ncr0:2:0): medium error, info = 2031300 (decimal)
Mar 15 02:50:19 usenet1 /netbsd: sd2(ncr0:2:0): medium error, info = 2031300 (decimal)
Mar 15 03:50:26 usenet1 /netbsd: sd2(ncr0:2:0): medium error, info = 2031204 (decimal)
Mar 15 04:06:05 usenet1 /netbsd: sd2(ncr0:2:0): medium error, info = 2031204 (decimal)
Mar 15 04:06:07 usenet1 /netbsd: sd2(ncr0:2:0): medium error, info = 2031172 (decimal)

The panic message I got was:

dev = 0x414, ino = 28, fs = /local2
panic: ifree: freeing free inode

When I did a stack trace, the last function that was running (which
called panic()) was _ffs_vfree.

So, is this just an FFS bug?  Or did the medium errors cause this to
happen?  Is this still an NCR driver bug, or do I just have a disk too
damaged for NetBSD to use it (the drive is only about a month old,
incidentally)?

More details as I understand what's going on...

-----------------------------------------------------------------------------
  Michael L. VanLoon                                 michaelv@HeadCandy.com
       --<  Free your mind and your machine -- NetBSD free un*x  >--
     NetBSD working ports: 386+PC, Mac, Amiga, HP300, Sun3, Sun4, PC532,
                           DEC pmax (MIPS R2k/3k), DEC/AXP (Alpha)
     NetBSD ports in progress: VAX and others...
-----------------------------------------------------------------------------