PCU panic on V240

Running -current on a Sparc V240 I get a diagnostic assertion running
the script attached:

panic: kernel diagnostic assertion "!cpu_intr_p()" failed: file "/work/build/src/sys/kern/subr_xcall.c", line 351 
Stopped in pid 0.20 (system) at netbsd:cpu_Debugger+0x4:        nop
db{1}> bt
db{1}> call stackdump
Frame pointer is at 0x2b6d6ea41
Call traceback:
 netbsd:db_command+0x94(ff00, 0, ffffffffffffffff, 2b6d6f428, 1c19c00, 69) fp = 2b6d6eb71
 netbsd:db_command_loop+0x110(1c16c80, 1c16cb8, 0, 0, 6, 73) fp = 2b6d6ecb1
 netbsd:db_trap+0x100(10ef028, 0, 14, 1c6feb9, 0, 1000000) fp = 2b6d6ed91
 netbsd:kdb_trap+0x134(101, 0, 0, 0, 10d06a0, 1857de0) fp = 2b6d6ee51
 netbsd:trap+0x398(101, 2b6d6f950, 0, fefefefefefefeff, ffffffffffffffff, ff000000) fp = 2b6d6ef01
 netbsd:1008d2c+0(2b6d6f950, 101, 10ef020, 1d0006, 2b6d6fb08, 73) fp = 2b6d6f0a1
 netbsd:vpanic+0x16c(1868da0, 1d32fa8, 18168c0, 2b6d6fc78, 100a080, 2b6d58000) fp = 2b6d6f281
 netbsd:kern_assert+0x34(18168c0, 2b6d6fc78, 1624520, 1624524, 28, 14c8e80) fp = 2b6d6f331
 netbsd:xc__highpri_intr+0xa8(18168c0, 18168f8, 1820218, 188ce90, 15f, 1c93800) fp = 2b6d6f3f1
 netbsd:softint_dispatch+0x104(0, 1c95580, 1d2a290, 0, 18533d8, 1853ce0) fp = 2b6d6f4a1
 netbsd:softint_fastintr+0x8c(1856f60, 4, 1856f38, 18168c0, 18168f8, 1856d48) fp = 2b6d6f571
 netbsd:softint_dispatch+0x248(109428880, 4, 1857000, 18168c0, 18168f8, 1857de0) fp = 2b6d6f621
 netbsd:100030+0(f005eab8, 113800, 113c00, 1118e0, 111d78, 111858) fp = fedc9651
0

This is the high priority xcall from subr_pcu.c and
curcpu()->ci_idepth is zero.

What is going wrong here?

--
J. Hannken-Illjes - [email protected] - TU Braunschweig (Germany)