Current-Users archive

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

Re: tmpfs panic



On Sat, Apr 20, 2013 at 01:05:32AM +0200, Thomas Klausner wrote:
> On Fri, Apr 19, 2013 at 07:07:44PM +0100, Patrick Welche wrote:
> > Good news is I have a core file, though something seems odd:
> > 
> > (gdb) target kvm netbsd.1.core
> > #0  0x0000000000000000 in ?? ()
> > (gdb) bt
> > #0  0x0000000000000000 in ?? ()
> > #1  0x0000000000000000 in ?? ()
> > (gdb) print panicstr
> > $1 = 0xffffffff80d43b00 "kernel diagnostic assertion 
> > \"(node)->tn_spec.tn_dir.tn_readdir_lastp == NULL || 
> > tmpfs_dircookie((node)->tn_spec.tn_dir.tn_readdir_lastp) == 
> > (node)->tn_spec.tn_dir.tn_readdir_lastn\" failed: file \"../.."...
> 
> Reminds me of my PR 47739. Good that you have a core file, I don't
> have one.

It looks exactly like your PR: I was doing a pkg_rr with a tmpfs WRKOBJDIR,
so excercising tmpfs. Unfortunately the corefile doesn't seem to be of
any use?

# crash -M netbsd.1.core -N netbsd.gdb
Crash version 6.99.19, image version 6.99.19.
System panicked: kernel diagnostic assertion 
"(node)->tn_spec.tn_dir.tn_readdir_lastp == NULL || 
tmpfs_dircookie((node)->tn_spec.tn_dir.tn_readdir_lastp) == 
(node)->tn_spec.tn_dir.tn_readdir_lastn" failed: file 
"../../../../fs/tmpfs/tmpfs_subr.c", line 643 
Backtrace from time of crash is available.
crash> bt
_KERNEL_OPT_NVGA_RASTERCONSOLE() at 0
_KERNEL_OPT_NVGA_RASTERCONSOLE() at 0
crash> ps
PID    LID S CPU     FLAGS       STRUCT LWP *               NAME WAIT
crash> 

... and this was the DIAGNOSTIC+LOCKDEBUG+UVMHIST box of uvn_fp2 fame...

Cheers,

Patrick


Home | Main Index | Thread Index | Old Index