Re: [PATCH 03/16] GFS: core fs

[David Teigland <teigland@redhat.com>]

On Tue, Oct 11, 2005 at 02:15:25PM +0200, Jan Hudec wrote:
> On Mon, Oct 10, 2005 at 23:39:28 +0200, Pavel Machek wrote:

> > > + for (head = &ai->ai_ail1_list, tmp = head->prev, prev = tmp->prev;
> > > +      tmp != head;
> > > +      tmp = prev, prev = tmp->prev) {
> > 
> > 

> > > + for (head = &ai->ai_ail1_list, tmp = head->prev, prev = tmp->prev;
> > > +      tmp != head;
> > > +      tmp = prev, prev = tmp->prev) {
> > 
> > 
> > Can you get less creative in the for loops? [There are more examples
> > at other patches, for (i=something; i--; ) was "nicest" example].
> 
> The later two are good examples of where list_for_each_safe is
> appropriate.

There are multiple places like this that need either a
list_for_each_entry_reverse_safe or list_for_each_prev_safe, neither of
which exist.  I'll send a patch to add one.

I've just converted to a macro in ail2_empty() -- I'm not sure why I'd
left it out in that spot, maybe to be consistent with ail1_empty above.

Thanks,
Dave