Re: simple handling of module removals Re: [OKS] Module removal

From: Daniel Gryniewicz (dang@fprintf.net)
Date: Mon Jul 08 2002 - 10:48:19 EST


Okay, maybe this is a bit naive, but isn't this problem already solved?
Couldn't we just put a read/write lock on the module, where using is
reading, and removing is writing? As I understand it, this should
impose little overhead on the use (read) case, and ensure that, when a
context has the remove (write) lock there are no no users (readers) and
cannot be any?

Daniel

On Mon, 2002-07-08 at 09:58, Thunder from the hill wrote:
> Hi,
>
> Still, we shouldn't lock everything. I could do awful lots of interesting
> things while the only thing that is being done is to remove a module. It
> doesn't make sense IMO to lock things that are completely unrelated to
> modules.
>
> And BTW, what's so much of an overhead if we tell everyone who tries to
> mess around with a certain module that he'd better wait until we unloaded
> it? It could be done like your schedule hack, but cleaner in that respect
> that those who got nothing to do with the module can keep on running.
>
> > Good point. Member usecount could be anything. A 'long' isn't the
> > correct pad for all types, but it will probably handle everything that
> > was intended.
>
> But as I mentioned - atomic_t is changed (e.g. to long long) ->
> module->pad blows up, because the sizeof(struct module) is different,
> depending on which part of the union we're using.
>
> Regards,
> Thunder
> --
> (Use http://www.ebb.org/ungeek if you can't decode)
> ------BEGIN GEEK CODE BLOCK------
> Version: 3.12
> GCS/E/G/S/AT d- s++:-- a? C++$ ULAVHI++++$ P++$ L++++(+++++)$ E W-$
> N--- o? K? w-- O- M V$ PS+ PE- Y- PGP+ t+ 5+ X+ R- !tv b++ DI? !D G
> e++++ h* r--- y-
> ------END GEEK CODE BLOCK------

-- 
Recursion n.:
        See Recursion.
                        -- Random Shack Data Processing Dictionary

- 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 : Mon Jul 15 2002 - 22:00:13 EST