Re: [PATCH] Make aout executables work again

[Parag Warudkar]

> > Is there a reason why we specifically want to make a distinction
> > between binfmts having and not having VDSOs and do we want aout to
> > have the possibility of NOT having VDSO when a) It used to have it
> > unconditionally b) nothing special is needed in arch or common code to
> > have it and c) Not letting have it requires special arch-specific
> > code* ?
>
> I mainly did it this way because it was quite a lot simpler.
> I'm not worrying too much about other strange binfmts to be honest

Well with your suggestion of using weak definition (BTW, I find that
thing cool :))  for arch_setup_additional_pages() I think my patch is
simpler and it preserves existing behavior - it just boils down to
actually calling the already existing function in binfmt_aout.c.


> > [*] I see the patch handles i386 and x86_64 - but I am not sure if
> > something similar will be needed for the other arches to allow aout
> > executables in absence of VDSO (powerpc, sh).
>
> Neither powerpc nor sh support a.out.

Ok, that's not a concern then. Not that I care about aout but if a
simple fix can let it have VDSO and prevent special casing, I would
prefer to let it have VDSO and still work - at least in the interest
of preserving existing behavior.

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