On Fri, Oct 02, 1998 at 06:18:55 -0700, Erik E. Fair wrote:
> Just out of curiosity, are you using many SunOS 4.1.x binaries in
> emulation? The kernel stack revealed that my panics were almost always
> triggered by a getdents() call out of the emulation layer...

I'll go for a dump next time it happens (actually I thought that I had at
least one dump rotating on the disk...but that must be some other machine...).

I may even have a traceback laying somewhere on a piece of paper...

Yes, this machine has COMPAT_{43,12,SUNOS} in the kernel, but only runs NetBSD
binaries compiled on the same machine.

We have another SS (a SS5) running NetBSD 1.3.2 as well, but with KAME IPv6, and this
machine has never seen this panic. This second machine has COMPAT_{12,43,SUNOS,SVR4}
and TCP_COMPAT_42. Also no sunos binaries on this machine.

Both machines have been running NetBSD since late summer this year, and the SS20
has seen the pv_unlink0 panics on the average at least once a week, while the SS5
has seen none of them.

I'm not new to NetBSD, but very new to NetBSD/sparc, and I added COMPAT_SUNOS
as an experiment to run the Solaris version of Netscape, but have since read
on this mailing list, that the needed COMPAT is SVR4 rather than for SUNOS.

I'll probably remove most (if not all) of the compat options, including COMPAT_SUNOS,
as I have no Sunos4 binaries to emulate... I'll probably keek the current setup
at least for one more panic to get the traceback.

regards
- Erik Bertelsen