Re: [PATCH] EDAC: core EDAC support code

[Greg KH <greg@kroah.com>]

On Mon, Mar 06, 2006 at 10:24:00PM +0000, Al Viro wrote:
> On Mon, Mar 06, 2006 at 01:53:44PM -0800, Greg KH wrote:
> > > 	rmmod your_turd </sys/spew/from/your_turd
> > > and there you go.  rmmod can _NOT_ wait for sysfs references to go away.
> > 
> > To be fair, the only part of the kernel that supports the above process,
> > is the network stack.  And they implemented a special kind of lock to
> > handle just this kind of thing.
> > 
> > That is not something that I want the rest of the kernel to have to use.
> > If your code blocks when doing the above thing, that's fine with me.
> 
> One word: fail.  With -EBUSY.
>  
> > Note, you better have the module owner reference right for the above to
> > not oops the kernel, deadlock is fine.
> 
> Never is.

My apologies, you are right, for some reason I thought rmmod would just
wait for the reference count to go away.  I just tested this on a lot of
different things in sysfs and it works properly and rmmod will return an
error saying the module is in use at this time.

> > There is no rule that we _have_
> > to allow rmmod to always succeed.
> 
> Quite so, which means we can have it fail saying that module removal has
> failed.  Deadlock is not the same thing.

Agreed.

thanks,

greg k-h