Re: access_ok inside kernelspace

Horst von Brand (vonbrand@inf.utfsm.cl)
Tue, 19 Oct 1999 12:27:55 -0300

Tigran Aivazian <tigran@sco.COM> said:
> On Tue, 19 Oct 1999, Horst von Brand wrote:
> > I don't understand this. A module is linked into the running kernel, and
> > has access to all of it AFAIU, so I can't see any need for this contortion
> > for modules. Maybe I'm just plain blind today...

> The kernel only has access to symbols which base kernel chose to export
> (kernel/ksyms.c and arch/i386/kernel/i386_ksyms.c). However, you are not
> blind and neither was I - just slightly short-sighted :)

But you can export them if you wish. Is there anything against that idea?

> I forgot about the well-known fact that sys_call_table[] is exported so
> modules should call sys_XXX() functions indirectly via sys_call_table[].

That means an array lookup + a indirect function call, and also no argument
checking by gcc. Not exactly a overwhelming cost, but...

> I even wrote a little module ages ago that is based solely on this simple
> fact.

;-)

-- 
Dr. Horst H. von Brand                       mailto:vonbrand@inf.utfsm.cl
Departamento de Informatica                     Fono: +56 32 654431
Universidad Tecnica Federico Santa Maria              +56 32 654239
Casilla 110-V, Valparaiso, Chile                Fax:  +56 32 797513


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