Re: 2.5-bk AT_GID clash

Rusty Russell (rusty@rustcorp.com.au)
Wed, 13 Nov 2002 05:07:03 +1100


In message <1037122398.27014.43.camel@jen.americas.sgi.com> you write:
> On Tue, 2002-11-12 at 11:16, Rusty Russell wrote:
> > This might be kOK too, but in practice I don't think much will be in
> > moduleloader.h: asm/module.h only really defines struct
> > mod_arch_specific, which is embedded in struct module, and struct
> > module needs to be exposed for those inlines...
>
> But does everyone who wants to implement a module need to be exposed
> to all the details of the elf header?

Well, linux/module.h -> asm/module.h -> linux/elf.h. Although if you
use #define instead of typedef you can break the last link. Feel free
to send a patch to split it into moduleload.h or something, but I
think it'll look tiny.

But IMHO the nameclash needs to be fixed *anyway*, not hacked around,
or someone else will run over it one day. AFAICT, changing
fs/binfmt_elf.c and elf.h to AT_RGID is the simplest. Both should be
mildly chastised for using a prefix like AT_ publically.

Rusty.

--
  Anyone who quotes me in their sig is an idiot. -- Rusty Russell.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/