Re: Encoding page size in the ELF header

[Florian Weimer <fweimer at redhat dot com>]

On 09/25/2015 04:53 PM, Szabolcs Nagy wrote:
> On 25/09/15 14:04, Florian Weimer wrote:
>> Not sure if this the right mailing list for this discussion.
>>
>> I think we should record the expected page size in the ELF header in
>> some way.  It affects the behavior of the static linker, the dynamic
>> linker, and even some applications.  (For example, an application might
>> want to make a part of .data read-only after initialization, as a
>> hardening measure.)
>>
> 
> you mean to fail when the runtime page size does
> not match the expected page size of the binary?
> (that would make binaries non-portable so each
> page size would be a different abi).

Such binaries already exist, see âsuhosin_write_protect_configurationâ.
 This is quite a useful feature, and it would be nice to be able to do
this with greater reliability.  (Something similar would have helped
Exim earlier this year, for instance.)

> or do you want a hint that makes things more
> efficient/secure in the normal case, but there
> is still some fall back?
> (i'm not sure how this could work)

That's the current state, ld.so will make additional segments writable
if the system page size is larger than what the static linker expected.

-- 
Florian Weimer / Red Hat Product Security