Re: [OOPS] 2.5.63 - NULL pointer dereference in loop device

Andrew Morton (akpm@digeo.com)
Wed, 26 Feb 2003 02:52:16 -0800

Hugh Dickins <hugh@veritas.com> wrote:
>
> On Mon, 24 Feb 2003, Jonah Sherman wrote:
>
> > I have come across a bug in the loop driver. To reproduce this bug,
> > simply do:
> ...
> I can't reproduce this, and I don't understand it: please help me!

Well I can't make it happen either now. It went pop first time I tried it
yesterday.

That being said, I can still trigger it by mmapping /dev/loop0 MAP_SHARED and
dirtying it all. That triggers the problematic PF_MEMALLOC path much more easily.

mem=256M
losetup /dev/loop0 /dev/hda5
usemem -m 300 -f /dev/loop0
<oops>

(gdb) p/x lo->lo_flags
$3 = 0x0

Userspace is passing in lo_encrypt_type == 0, so loop_init_xfer() never calls the transfer
init function.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/