Re: [module-init-tools] fix weak symbol handling

Richard Henderson (rth@twiddle.net)
Thu, 16 Jan 2003 18:09:13 -0800

On Fri, Jan 17, 2003 at 12:57:03PM +1100, Rusty Russell wrote:
> > No. The semantics I need is if A references a weak symbol S
> > and *no one* implements it, then S resolves to NULL.
>
> Sorry, I was unclear. I want to know the dependency semantics:
>
> If B exports S, should depmod believe A needs B, or not? Your patch
> leaves that semantic (all it does is suppress the errors).

Well, that depends on whether A defines S or not. If A does
define S, then I don't care. I'd say "no", A does not depend
on B. If A does not define S, then most definitely "yes", as
with any other definition.

> I'm not sure what semantics are "right", since I don't know what
> you're trying to do, or what is wrong with get_symbol().

I just told you. Quoted again above. Perhaps the following
dummy module will make things even clearer.

---
#include <linux/module.h>
#include <linux/init.h>


extern int not_defined __attribute__((weak));


static int init(void)
{
  return &not_defined ? -EINVAL : 0;
}


static void fini(void)
{
}


module_init(init);
module_exit(fini);

---


You should be able to load this module.




r~
-
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/