Re: Tmpfs panic in -current

Hello Kevin,

May you rebuild your kernel with some options that will make the dump of
the crash easier to send to people?
Try to follow these steps:
1) Configure KERNEL with the folowing options:
options         KDB
options         GDB
options         DDB

2) Configure rc.conf:
dumpdev="/dev/ad0s1b" #Your swap slice.
dumpdir="/var/crash"  #By default.
crashinfo_enable="YES"
crashinfo_program="/usr/sbin/crashinfo"

3) When the FreeBSD crash, it won't rebut, but will be stop on a gdb prompt.
        3.1) gdb> continue
        3.2) gdb> dump

4) kgdb /boot/kernel/kernel.symbols /var/crash/vmcore.0
5) kgdb> backtrace
6) kgdb> up 4
7) kgdb> up

6 --> Up to line number 4.
7 --> Up to the next line.


Best Regards,
- Araujo


2012/6/25 Konstantin Belousov <kostikbel_at_gmail.com>

> On Mon, Jun 25, 2012 at 10:03:28AM +0800, Kevin Lo wrote:
> > I've observed a panic in recent -current several times but I only
> > have a picture of the backtrace:
> > http://people.freebsd.org/~kevlo/panic_tmpfs.jpg
> >
> > Does this look at all familiar to anyone?
>
> Can you look up the line corresponding to tmpfs_reg_resize + 0x627 address
> in your kernel ?
>
> The screenshot looks strange. The instruction on which the kernel trapped
> is int 0x28 which should not appear in the compiled code.
>



-- 
Marcelo Araujo
araujo_at_FreeBSD.org