missing frames in gdb -k on x86?

I'm running a ~2 week old -current, and when debugging some hacks of
mine, I noticed that gdb -k seems to be missing at least one frame
in its stack.  Eg, ddb shows:

panic: page fault
cpuid = 0; lapic.id = 00000000
Stack backtrace:
backtrace(c0380cf2,0,c036e757,d96d4bd4,100) at backtrace+0x17
panic(c036e757,c0395a7f,c5283164,1,1) at panic+0x14a
trap_fatal(d96d4c98,0,c0395c61,2cd,c5281e40) at trap_fatal+0x356
trap_pfault(d96d4c98,0,0,d0d0d0d0,0) at trap_pfault+0x1c2
trap(18,10,d96d0010,0,0) at trap+0x2ed
calltrap() at calltrap+0x5
--- trap 0xc, eip = 0xc52b3c5b, esp = 0xd96d4cd8, ebp = 0xd96d4cf4 ---
mi_go(0,0,c52b381f,c5283000,d96d4d34) at mi_go+0x2b
freebsd_mapper_thread(0,d96d4d48,c037e742,30e,d0d0d0d0) at freebsd_mapper_thread+0x49
fork_exit(c52b381f,0,d96d4d48) at fork_exit+0xcf
fork_trampoline() at fork_trampoline+0x8
--- trap 0x1, eip = 0, esp = 0xd96d4d7c, ebp = 0 ---
boot() called on cpu#0


But gdb -k shows:

# gdb -k /sys/i386/compile/DEBUG/kernel.debug /var/crash/vmcore.0
GNU gdb 5.2.1 (FreeBSD)
Copyright 2002 Free Software Foundation, Inc.
<....>

#0  doadump () at ../../../kern/kern_shutdown.c:240
#1  0xc020c3eb in boot (howto=256) at
../../../kern/kern_shutdown.c:372
#2  0xc020c7f6 in panic () at ../../../kern/kern_shutdown.c:550
#3  0xc03442e6 in trap_fatal (frame=0xd96d4c98, eva=0)
    at ../../../i386/i386/trap.c:821
#4  0xc0343f62 in trap_pfault (frame=0xd96d4c98, usermode=0, eva=0)
    at ../../../i386/i386/trap.c:735
#5  0xc0343b1d in trap (frame=
      {tf_fs = 24, tf_es = 16, tf_ds = -647167984, tf_edi = 0, tf_esi = 0, tf_ebp = -647148300, tf_isp = -647148348, tf_ebx = 0, tf_edx = 0, tf_ecx = -1065271296, tf_eax = -987394048, tf_trapno = 12, tf_err = 2, tf_eip = -987022245, tf_cs = 8, tf_eflags = 66118, tf_esp = -986972328, tf_ss = -986972233})
    at ../../../i386/i386/trap.c:420
#6  0xc032c318 in calltrap () at {standard input}:103
#7  0xc52b3868 in freebsd_mapper_thread (arg=0x0)
    at /home/gallatin/mx/mx/driver/freebsd/mi.c:60
#8  0xc01f785f in fork_exit (callout=0, arg=0x0, frame=0x0)
    at ../../../kern/kern_fork.c:790


The interesting function (mi_go) where the null pointer deref happened
is missing from the trace.  I thought only alpha had these kinds of
problems :(

Drew