Hi
Within one hour I got the next one, kdb log in attachment.
first line of kdb_log:
------------------------------------------------
Entering kdb (current=0xc2168000, pid 1593) kdb: Debugger re-entered on cpu 1, new reason = 10
on processor 0 Strange, cpu 1 should not be running
Not executing a kdb command
No longjmp available for recovery
due to WatchDog Interrupt @ 0xc0108f05
Cannot recover, allowing event to proceed
eax = 0xc0111eb4 ebx = 0xc03de880 ecx = 0x00000004 edx = 0xc03de888
esi = 0xc6d01640 edi = 0x00000012 esp = 0xc2169e9c eip = 0xc0108f05
ebp = 0xc2169eac xss = 0x00000018 xcs = 0x00000010 eflags = 0x00000002
xds = 0xc2160018 xes = 0xc01c0018 origeax = 0xc0111eb4 ®s = 0xc2169e68
[0]kdb> o2o6o8o-o9o4ooooo''5'i'a'd'kdb: expected enter got 0xe9 status 0x14
Unknown kdb command: 'o2o6o8o'
*** input to kdb ^^^^^^^ is rubbish because something felt on keyboard...
[0]kdb> cpu
Currently on cpu 0
Available cpus: 0, 1*
[0]kdb> bt
EBP EIP Function(args)
0xc2169eac 0xc0108f05 free_irq+0x71 (0x4, 0xc03fc3d0)
kernel .text 0xc0100000 0xc0108e94 0xc0108f2c
0xc2169ec8 0xc01d2141 shutdown+0xf9 (0xc12af680)
kernel .text 0xc0100000 0xc01d2048 0xc01d22b0
0xc2169ee0 0xc01d3aa0 rs_close+0x138 (0xc3f71000, 0xc6062d40)
kernel .text 0xc0100000 0xc01d3968 0xc01d3b3c
0xc2169f50 0xc01c07ef release_dev+0x26f (0xc6062d40)
------------------------------------------------------------------------
Observations:
- compiling the kernel as UP keeps it from crashing
- the bios RTC is most of the times disturbed after a crash
The application 'hart' does basically the following:
for(;;)
{
open port (ttyS0)
write a few bytes
ioctl toggle RTS line
read a few bytes (with poll)
close port
}
If some information is required, I'll try to provide it.
Kees
This archive was generated by hypermail 2b29 : Wed May 23 2001 - 21:00:41 EST