Fatal trap 12: page fault panic with recent kernel with ZFS

I'm not sure if this is related to recent kernel memory code changes
or what, but it hasn't happened with code from earlier than a couple
days ago.  I had this happen twice, I think the first time was with 
arc max set to 1024M, the second time was when arc max was unset in
loader.conf and the system had been up a few hours but the arc hadn't
been squeezed down to a small number yet:
Mem: 719M Active, 12G Inact, 3413M Wired, 2608K Cache, 24M Buf, 3741M Free

I've deleted the kernel since then but I did not change my sources,
I could build a new one and check where the pointers point to I think?
If needed.  Or I could reproduce the panic if needed.  It doesn't dump,
it just gets stuck after printing the panic.

Transcribed by hand to save bandwidth:

Fatal trap 12: page fault while in kernel mode
cpuid = 3; apic id = 03
fault virtual address = 0xffff804000000000
fault code = supervisor read data, page not present
instruction pointer = 0x20:0xffffffff804a4d4e
stack pointer = 0x28:0xffffff80000c3900
frame pointer = 0x28:0xffffff80000c3910
code segment = base 0x0, limit 0xfffff, type 0x1b
= DPL 0, pres 1, long 1, def32 0, gran 1
processor eflags = interrupt enabled, resume, IOPL = 0
current process = 4 (g_down)
trap number = 12
panic: page fault
cpuid = 0
Uptime: 2h40m14s