At 5:12 PM -0800 3/10/98, Bill Studenmund wrote:
>Here's the patch to try and stop kernels from hanging when they go to set
>ttyflags.

Well, I tried the patch.  The patch program didn't like your email (the
@@'s specifically) so I applied it by hand.  (Some line feeds may have
gotten scrambled in transit from Eudora to desktop to fetch to NetBSD.  Not
sure where.)

Anyway I built a 1.3 kernel with the patch.  (A simple make after applying
it should work, right?  It seemed to redo the right things.)  A copy of the
config file and resulting kernel is in
ftp://macbsd.jpl.nasa.gov/pub/users/hotz.  HOTZ#0 was without patch, HOTZ#1
with.

Result:  no apparent change.  Sorry.

Dave Huang has been looking at my other crashes and some of his own and
speculates that the DMA may not be disabled during boot, just the DMA
completion interrupt.  If there is a problem like that around then who
knows what failures we should expect to see.

Signature failed Preliminary Design Review.
Feasibility of a new signature is currently being evaluated.
h.b.hotz@jpl.nasa.gov, or hbhotz@oxy.edu