storing a value larger than UINT_MAX

storing a value larger than UINT_MAX

[Matthijs Möhlmann]

Hello ext4 developers,

I am trying to run a kernel with grsecurity with the size overflow
protection and am getting the following warnings / errors:

dmesg: http://pastebin.com/wr3UGLS9
config: http://pastebin.com/sr8M9bP0
mballoc.* (make fs/ext4/mballoc.o EXTRA_CFLAGS="-fdump-tree-all
-fdump-ipa-all") http://filebin.ca/3DMIChVw9lQM/mballoc.tgz

According to the grsecurity developers it seems to be a bug in ext4:
https://forums.grsecurity.net/viewtopic.php?f=1&t=4678&p=16971

The response from ephox (PAX team / grsecurity developer):
--
Thanks for the report. I think this is an upstream bug. Based on the
runtime values provided by you, ext4_mb_new_group_pa() tries to store a
value into pa->pa_lstart which larger than UINT_MAX which comes from
ext4_group_first_block_no().
Could you please report it to the ext4 developers?
--

I'll try to answer all the questions but I'm not an expert in this area.

I am also not subscribed to this mailinglist so please keep me in the CC.

Regards,

Matthijs Möhlmann

Re: storing a value larger than UINT_MAX

[Andreas Dilger]

[-- Attachment #1: Type: text/plain, Size: 1674 bytes --]

On Feb 24, 2017, at 3:38 AM, Matthijs Möhlmann <matthijs@cacholong.nl> wrote:
> 
> Hello ext4 developers,
> 
> I am trying to run a kernel with grsecurity with the size overflow
> protection and am getting the following warnings / errors:
> 
> dmesg: http://pastebin.com/wr3UGLS9
> config: http://pastebin.com/sr8M9bP0
> mballoc.* (make fs/ext4/mballoc.o EXTRA_CFLAGS="-fdump-tree-all
> -fdump-ipa-all") http://filebin.ca/3DMIChVw9lQM/mballoc.tgz
> 
> According to the grsecurity developers it seems to be a bug in ext4:
> https://forums.grsecurity.net/viewtopic.php?f=1&t=4678&p=16971

        pa->pa_pstart = ext4_grp_offs_to_block(sb, &ac->ac_b_ex);
        pa->pa_lstart = pa->pa_pstart;

pa_pstart is 64-bit, pa_lstart is 32-bit.  It isn't clear why pa_lstart isn't:

	pa->pa_lstart = ac->ac_b_ex.fe_logical;

as elsewhere in the code, but this _may_ be because the group prealloc is
for allocating multiple small files in the same group, so pa_lstart may not
make any sense as a per-file logical offset.

Cheers, Andreas

> The response from ephox (PAX team / grsecurity developer):
> --
> Thanks for the report. I think this is an upstream bug. Based on the
> runtime values provided by you, ext4_mb_new_group_pa() tries to store a
> value into pa->pa_lstart which larger than UINT_MAX which comes from
> ext4_group_first_block_no().
> Could you please report it to the ext4 developers?
> --
> 
> I'll try to answer all the questions but I'm not an expert in this area.
> 
> I am also not subscribed to this mailinglist so please keep me in the CC.
> 
> Regards,
> 
> Matthijs Möhlmann
> 


Cheers, Andreas






[-- Attachment #2: Message signed with OpenPGP --]
[-- Type: application/pgp-signature, Size: 195 bytes --]