Re: Re: fr1-2.16 patch for 2.6.10 kernel

"Also sprach denis bonnenfant:"
[Charset ISO-8859-1 unsupported, filtering to ASCII...]
> denis bonnenfant a _crit :
> >>
> >> It appears the test for the appropriate space assumed too much about
> >> the quantity of data left engraved in the space. p->rdev->bdev
> >> apparently is sometimes erased, so its attributes cannot be tested.
> >>
> >> Maybe the above will help. Maybe it won't.
> > 
> > It goes a little bit farther :
> > 
> > raid1 1091: ** mddev=f7fc8480 rdev=f7131880 mirror=0
> > RAID1 conf printout:
> > 
> > so it's in print_conf ...
> > so i comment it...
> 
> now it's
> 
> Oops: 0000 [#1]
> PREEMPT
> Modules linked in: raid1 bitmap md epic100 usbkbd parport_pc lp parport
> CPU:    0
> EIP:    0060:[<f88a4858>]    Not tainted
> EFLAGS: 00010082   (2.6.8-enbd-fr1-debug1)
> EIP is at unplug_slaves+0x48/0xa8 [raid1]

unplug_slaves is a new function in 2.6.8.1. It's likely this line, a
bit less than half way through:

   request_queue_t *r_queue = bdev_get_queue(rdev->bdev);

I suspect that the bdev struct has simply been handed back to the kernel
and has been reused, so it contains nonsense. Can you see in the dmesg
output a

   "md: export_rdev(%s)\n"

?? I'd like you to put a printk before every call to export_rdev() in
md.c, so that we find out which call to export_rdev it is.

Peter