Re: [RFC] Raceless module interface

[Daniel Phillips <phillips@arcor.de>]

On Friday 13 September 2002 18:39, Thunder from the hill wrote:
> Hi,
> 
> On Fri, 13 Sep 2002, Daniel Phillips wrote:
> > That's debatable.  Arguably, a failed ->module_cleanup() should be
> > retried on every rmmod -a, but expecting module.c to just keep
> > retrying mindlessly on its own sounds too much like a busy wait.
> 
> Hmmm. You might as well give it back to the user.
> 
> rmmod: remove failed: do it again!

That's what happens now.  We can certainly improve the error message,
and actually, that just falls out, from properly adding an error
return code to ->cleanup_module()

> So the cleanup code could as well just do it on its own.

???

> > > Why is that sloppy? E.g. kfree() happily accepts NULL pointers as well.
> > 
> > That is sloppy.  Different discussion.
> 
> What should kfree do in your opinion? BUG()?

Yuppers:

static inline void kfree_test(void *object)
{
	if (object)
		kfree(object);
}

#define kfree_sloppy kfree_test

s/kfree/kfree_sloppy/g

(But see "different discussion" above.)

> doodle.c:12: attempted to free NULL pointer, as you know it already is.

Um.  You know it's NULL and you freed it anyway?

(But see "different discussion" above.)

> > I take it that the points you didn't reply to are points that you
> > agree with?  (The main point being, that we both advocate a simple,
> > two-method interface for module load/unload.)
> 
> You could even do it using three methods.

Yes, or two, my favorite.

-- 
Daniel