Another alpha panic

[Andrew Gallatin]

John Baldwin writes:
 > On Sunday 11 April 2004 05:33 am, Kris Kennaway wrote:
 > > Not that I've had any luck getting someone to take a look at the last
 > > half-dozen alpha panics I've reported, but:
 > >
 > > panic: pmap_emulate_reference(0xfffffc0019436a00, 0xfffffe0014495e68, 0,
 > > 1): pa 0xb0a0000 not managed at line 2585 in file
 > > /a/asami/portbuild/alpha/src-client/sys/alpha/alpha/pmap.c Stack backtrace:
 > > db_print_backtrace() at db_print_backtrace+0x18
 > > backtrace() at backtrace+0x2c
 > > __panic() at __panic+0x150
 > > pmap_emulate_reference() at pmap_emulate_reference+0x15c
 > > trap() at trap+0x39c
 > > XentMM() at XentMM+0x2c
 > > --- memory management fault (from ipl 7) ---
 > > pmap_activate() at pmap_activate+0xb4
 > > Lcs1() at Lcs1+0x1c
 > > --- root of call graph ---
 > 
 > This is a VM-related panic.  Perhaps alc@ might have an idea.  There might be 
 > a reference fault on a K0SEG address or some other silly bug.

I just googled and saw that at least 2 of these have come with the
same stack (context-switch(), pmap_activate()).  What does
pmap_activate+0xb4 map to in your kernel?

What sort of machine?  Is this UP or SMP?

Drew