On Sun, 28 Oct 2001, Roland McGrath wrote:

> It seems likely that this crash is due to improper reentrance.  I am
> suspicious of your other patch, because the situation it disables the check
> for is just the kind of situation I would suspect in this kind of lossage.

As you have foreseen, attacking the problem probebly, solved most of
the crashes I have seen. The only one which is left know seems to be a
kind of a race condition. I have the normal pfinet translater attachet
to /servers/socket/2. Then I start my own program which tries to answer
any incomming ping packet (I'm still working on the program for testing
the low level net functions). Now I started pinging the hurd box. And
after some time the kernel crashes.

Here the ping output from the linux box:

newton:~$ ping gauss
PING gauss.wlab.debian (192.168.2.3): 56 data bytes
64 bytes from 192.168.2.3: icmp_seq=0 ttl=255 time=2.3 ms
64 bytes from 192.168.2.3: icmp_seq=1 ttl=255 time=0.8 ms
64 bytes from 192.168.2.3: icmp_seq=2 ttl=255 time=0.8 ms
64 bytes from 192.168.2.3: icmp_seq=3 ttl=255 time=126.0 ms
64 bytes from 192.168.2.3: icmp_seq=4 ttl=255 time=2.1 ms
64 bytes from 192.168.2.3: icmp_seq=5 ttl=255 time=0.9 ms
[gauss has crashed]


And here is the bt from gdb:


(gdb) bt
#0  panic (fmt=0x20710a "vm_map_copyout: wiring 0x%x") at
../../../oskit-20010214/libc/gen/panic.c:31
#1  0x001224a9 in vm_map_copyout (dst_map=0x408b250, dst_addr=0x59e4f4c, 
copy=0x4087058) at ../../oskit-mach/vm/vm_map.c:2656
#2  0x00103531 in ipc_kmsg_copyout_body (saddr=68354088, eaddr=68354128, 
space=0x408b860, map=0x408b250)
    at ../../oskit-mach/ipc/ipc_kmsg.c:2518
#3  0x0010d5fb in mach_msg_trap (msg=0x10217cc, option=3, send_size=24, rcv_size=64, 
rcv_name=39, time_out=0, notify=0)
    at ../../oskit-mach/ipc/mach_msg.c:1244
#4  0x0013e253 in mach_call_call ()
#5  0x04113d80 in ?? ()


I know this kind of thing doing to the kernel is nasty, but
still the kernel shouldn't not crash, I think. 
Now I'm trying to find out more about this problem and of course how
this can be fixed.


wagi

-- 
Daniel Wagner                              "use quit to exit"
email: [EMAIL PROTECTED]

GnuPG: 1024D/DCDE890A (public key available on any keyserver)

_______________________________________________
Bug-hurd mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-hurd

Reply via email to