Re: [kernel-hardening] [PATCH 2/2] arm: apply more __ro_after_init

[Greg KH]

On Fri, Jun 03, 2016 at 02:26:54PM -0700, Kees Cook wrote:
> On Fri, Jun 3, 2016 at 11:51 AM, Greg KH <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> > On Fri, Jun 03, 2016 at 11:40:24AM -0700, Kees Cook wrote:
> >> Guided by grsecurity's analogous __read_only markings in arch/arm,
> >> this applies several uses of __ro_after_init to structures that are
> >> only updated during __init.
> >>
> >> Signed-off-by: Kees Cook <keescook@xxxxxxxxxxxx>
> >> ---
> >>  arch/arm/kernel/cpuidle.c |  2 +-
> >>  arch/arm/kernel/setup.c   | 10 +++++-----
> >>  arch/arm/kernel/smp.c     |  2 +-
> >>  arch/arm/lib/delay.c      |  2 +-
> >>  arch/arm/mm/mmu.c         |  9 ++-------
> >>  arch/x86/mm/ioremap.c     |  3 +--
> >
> > I don't think this x86 file is an arm-specific one :)
> 
> Hah, whooops. :)
> 
> > That minor nit aside, these patches are a great step forward, are you
> > going to take them and work to push them upstream, or do you want/need
> > others to do this?
> 
> I'll collect more like these and carry a tree for -next and push them for v4.8.

Sounds good!

Is there any "problem" with applying these markings to code that could
be built as a module?  I'm thinking of lots of buses and drivers that
have structures like this, but can be a module or not, depending on the
configuration selected.  It would be nice to get the "benefit" of
protection if the code is built into the kernel image.

thanks,

greg k-h