isp panic after printing the uptime

[Florian Smeets]

Hi,

i have a 100% reproducible panic on sparc64 with an isp controller.
Every time i reboot/shutdown the machine it panics after printing
"Uptime: XXXX". I'm not sure if this is more scsi or sparc64 related but
as the trace includes isp_* functions i went for scsi at .

The controller used is:

isp0: <Qlogic ISP 2200 PCI FC-AL Adapter> port 0x300-0x3ff mem
0x100000-0x100fff at device 4.0 on pci2
isp0: [ITHREAD]
isp0: Board Type 2200, Chip Revision 0x5, loaded F/W Revision 2.2.6
isp0: invalid NVRAM header
isp0: invalid NVRAM header

The messages prior to the panic and the panic look like this:

Sayncing diisks, vnotdes remaiining...n2 g (max 60 seconds) for system
process `syncer' to stop...2 1 0 1 0 0 done
All buffers synced.
zfs_umount:969[0]: Force unmount is not supported, removing FORCE flag.
zfs_umount:969[0]: Force unmount is not supported, removing FORCE flag.
zfs_umount:969[0]: Force unmount is not supported, removing FORCE flag.
lock order reversal:
  1st 0xfffff800038394e0 ufs (ufs) @ /usr/src/sys/kern/vfs_mount.c:1202
  2nd 0xfffff80003839c40 devfs (devfs) @ /usr/src/sys/kern/vfs_subr.c:2092
KDB: stack backtrace:
_witness_debugger() at _witness_debugger+0x38
witness_checkorder() at witness_checkorder+0xcf8
__lockmgr_args() at __lockmgr_args+0x874
vop_stdlock() at vop_stdlock+0x38
VOP_LOCK1_APV() at VOP_LOCK1_APV+0x110
_vn_lock() at _vn_lock+0x80
vget() at vget+0x120
devfs_allocv() at devfs_allocv+0x114
devfs_root() at devfs_root+0x3c
dounmount() at dounmount+0x490
vfs_unmountall() at vfs_unmountall+0x3c
boot() at boot+0x674
reboot() at reboot+0x58
syscall() at syscall+0x2bc
-- syscall (55, FreeBSD ELF64, reboot) %o7=0x102a88 --
userland() at 0x12b348
user trace: trap %o7=0x102a88
pc 0x12b348, sp 0x7fdffffdd91
pc 0x101778, sp 0x7fdffffdf01
pc 0x1001d0, sp 0x7fdffffe531
pc 0, sp 0x7fdffffe5f1
done
Uptime: 6m9s
panic: trap: fast data access mmu miss
cpuid = 0
KDB: enter: panic
[thread pid 1 tid 100001 ]
Stopped at      kdb_enter+0x80: ta              %xcc, 1
db> where
Tracing pid 1 tid 100001 td 0xfffff80002082000
panic() at panic+0x20c
trap() at trap+0x570
-- fast data access mmu miss tar=0x1454156000 %o7=0xc040e7a4 --
_mtx_lock_spin_flags() at _mtx_lock_spin_flags+0x5c
callout_lock() at callout_lock+0x50
untimeout() at untimeout+0xc
isp_done() at isp_done+0x140
isp_intr() at isp_intr+0x3eb8
isp_poll() at isp_poll+0x38
xpt_polled_action() at xpt_polled_action+0xf0
dashutdown() at dashutdown+0x130
boot() at boot+0x8ac
reboot() at reboot+0x58
syscall() at syscall+0x2bc
-- syscall (55, FreeBSD ELF64, reboot) %o7=0x102a88 --
userland() at 0x12b348
user trace: trap %o7=0x102a88
pc 0x12b348, sp 0x7fdffffdd91
pc 0x101778, sp 0x7fdffffdf01
pc 0x1001d0, sp 0x7fdffffe531
pc 0, sp 0x7fdffffe5f1
done
db>

Anything i can do, aside from not rebooting the machine :-) ?

Cheers,
Florian