Re: [PATCH] protect remove_proc_entry

[Arjan van de Ven]

On Sat, 2005-12-31 at 01:58 -0500, Steven Rostedt wrote:
> On Fri, 30 Dec 2005, Andrew Morton wrote:
> 
> > Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
> > >
> > > +static DEFINE_SPINLOCK(remove_proc_lock);
> > >
> >
> > I'll take a closer look at this next week.
> >
> > The official way of protecting the contents of a directory from concurrent
> > lookup or modification is to take its i_sem.  But procfs is totally weird
> > and that approach may well not be practical here.  We'd certainly prefer
> > not to rely upon lock_kernel().
> >
> 
> Yeah, I thought about using the sem (or mutex ;) but remove_proc_entry is
> used so darn much around the kernel, I wasn't sure it's not used in irq
> context. 

it can't be; "anything-VFS" like this can sleep if the file is busy etc
etc.


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/