Re: PCI driver module unload race?

From: Rusty Russell (rusty@rustcorp.com.au)
Date: Sun Mar 16 2003 - 08:13:39 EST


In message <200303111707.09119.oliver@neukum.name> you write:
>
> > This means the module refcount must remain at 0, even after it's bound to
> > devices. Changing this would require a change in visible behavior, and
> > require an extra step by a user to disconnect the driver before they
> > unload the module.
>
> Yes, that would mean changing behaviour. On the other hand, we require
> new module utilities for 2.6 anyway, so why not?

Because I think it's silly, from a user point of view. Why are you
removing the module, after all?

Anyway, a pre-removal notifier in sys_module_delete would have the
same effect (with the same issues if the refcount isn't 0 after all,
and you then want to fail the rmmod).

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/



This archive was generated by hypermail 2b29 : Sun Mar 23 2003 - 22:00:20 EST