Current-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

rump load kmod segv



On a -current/amd64 system:

Reading symbols from /usr/bin/rump_allserver...(no debugging symbols 
found)...done.
(gdb) run -s -m /stand/amd64/6.99.32/modules/vnd/vnd.kmod unix:///tmp/sock
Starting program: /usr/bin/rump_allserver -s -m 
/stand/amd64/6.99.32/modules/vnd/vnd.kmod unix:///tmp/sock
WARNING: ZFS on NetBSD is under development

Program received signal SIGSEGV, Segmentation fault.
[Switching to LWP 1]
0x00000000e7805554 in ?? ()
(gdb) bt
#0  0x00000000e7805554 in ?? ()
#1  0x000000007fffd03d in ?? ()
#2  0x00007f7fe6c5328d in rumpns_kobj_reloc () from /usr/lib/librump.so.0
#3  0xffffffffffffffff in ?? ()
#4  0x0000000000000246 in ?? ()
#5  0x0000000000000000 in ?? ()


I still don't know what to expect from rump, so should that be doable?

On a tinkered-with -current/amd64 system:
(gdb) bt
#0  0x00000000e7806b22 in ?? ()
#1  0x000000007fffc03d in ?? ()
#2  0x00007f7ff7b43640 in ?? ()
#3  0xffffffffffffffff in ?? ()
#4  0x0000000000000246 in ?? ()
#5  0x0000000000000004 in ?? ()
#6  0x00007f7fffffd2a0 in ?? ()
#7  0x00007f7fe6ca7720 in module_do_load (
    name=0x7f7fe6c76430 <xc_wait+227> 
"\353\001\220\311\303UH\345H\203\354\060H}\350Hu\340HU\330HM\320H\215\005\354C(",
 isdep=127, flags=-11984, 
    props=0x7f7fe6efa840, modp=0x7f7fffffd100, class=32639, autoload=210)
    at /usr/src/lib/librump/../../sys/rump/../kern/kern_module.c:1118
Backtrace stopped: frame did not save the PC


The motivation for trying rump this time is: why do I get NOTTY
from DIOCGDISKINFO on a vnd (-current/amd64):
 
$ ./vndtest /dev/wd0g
size of /dev/wd0g: 500107862016 bytes
$ ./vndtest /dev/vnd0a
vndtest: DIOCGDISKINFO: Inappropriate ioctl for device

Is rump a good idea for debugging this?


Cheers,

Patrick


Home | Main Index | Thread Index | Old Index