Ever since supping a few days ago, my machine gets a page fault trap
when I try to connect to my ISP via ppp.  A kernel built 2/11/96
doesn't do this.  I supped on 2/15 (started 16:28 PST; finished 17:18
PST), and built a new kernel.  I just tried it today, and had this
problem.  I resupped today (started 22:50 PST; finished 23:23), built
a new kernel, and had exactly the same failure.  My 2/11 kernel still
works correctly.  The ppp sources have been touched recently, which
leads me to believe that may be the source of the trouble.

The modem dials correctly, and ppp even negotiates and completes the
connection, bringing IP up fully.  Shortly (a few seconds) after that,
the machine panics with this:

vm_fault(f87e7600, 0, 1, 0) -> 1
kernel: page fault trap, code = 0
Stopped at	_idp_ctlinput + 0x37:	cmpb $0x6, 0x1(%edx)

Trace revealed:
_idp_ctlinput(0, f875ef24, 0) at _idp_ctlinput + 0x37
_pfctlinput(0, f875ef24) at _pfctlinput + 0x2c
_ifioctl(f87e8100, 80206910, f9c29ec0, f87e8800, 0) at _ifioctl + 0x12a
_soo_ioctl(...) + 0x128
_sys_ioctl(...) + 0x355
_syscall() + 0x297
--- syscall (number 54) ---
0x10081c4b:

My kernel is built with gcc 2.7.2 (stock current gcc), and with "-O6
-fno-strength-reduce -m486 -pipe" options.

I noticed that the ppp files in /sys/net/ have been touched recently.
Can anyone shed any light on what was changed, and why they might have
broken?  Thanks.

-----------------------------------------------------------------------------
  Michael L. VanLoon                                 michaelv@HeadCandy.com
       --<  Free your mind and your machine -- NetBSD free un*x  >--
     NetBSD working ports: 386+PC, Mac 68k, Amiga, HP300, Sun3, Sun4,
                           DEC PMAX (MIPS), DEC Alpha, PC532
     NetBSD ports in progress: VAX, Atari 68k, others...
-----------------------------------------------------------------------------