More info [was Re: Sony PCG-GRX570 laptop, panic on boot w/ 5.1R...]

George Hartzell writes:
 > 
 > I've been trying to install something 5-ish on a Sony PCG-GRX570
 > laptop.
 > 
 > I started off trying to boot off of the 5.1 release CD, normally,
 > w/out acpi, and safe.  Every option panic-ed, with essentially the
 > same message (see below), although it followed a different driver
 > depending on how it was booted.
 > 
 > Then I installed 4.7 (since I had the CD), cvsup-ed my repository, and
 > cvs up'ed /usr/src to the 5-current.  I followed the section on moving
 > from 4 to 5-current in UPDATING to build the world, etc....  I had to
 > work around a bit of previously reported 4.7/5 weirdness in
 > /usr/include, but it went w/out any trouble.
 > 
 > When I reached the point where I was supposed to boot the new kernel
 > in single user mode, the 5-current kernel paniced:
 > 
 >   miibus0: <MII bux> in fxp0
 >   inphy0: <i82562ET 10/100 media interface> on miibus0
 >   inphy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
 >   
 >   
 >   
 >   Fatal trap 12: page fault while in kernel mode
 >   cpuid = 0; apic id = 00
 >   fault virtual address	= 0x63696d20
 >   fault code		= supervisor write, page not present
 >   instruction pointer	= 0x8:0xc0659df3
 >   stack pointer		= 0x10:0xc0c217ac
 >   frame pointer		= 0x10:0xc0c217cc
 >   code segment		= base 0x0, limit 0xfffff, type 0x1b
 >   			= DPL 0, pres 1, def32 1, gran 1
 >   processor eflags	= interrupt enabled, resume, IOPL = 0
 >   current process		= 0 (swapper)
 >   kernel: type 12 trap, code=0
 >   Stopped at	ithread_add_handler+0x163:	movl	%ebx,0(%eax)
 >   db>
 > 
 > I've seen several similar reports in the archives for late last
 > summer.  The general answer seemed to be that people were having
 > hardware trouble.  I don't think that is the case in my case, unless
 > -current is doing something very strange, since the same machine runs
 > well enough under 4.7 to buildworld and buildkernel, and the same
 > hardware has been running Suse and Win2000.
 > 
 > How can I help get this solved?

It turns out that the 5.0 release CD also boots successfully, so it
seems to be something that's happened in -current since then.

Help?

g.