[Ocfs2-devel] [PATCH 1/2] Introduce ocfs2_recover_node

[Wengang Wang]

On 10-11-17 09:49, Wengang Wang wrote:
> On 10-11-16 16:59, Goldwyn Rodrigues wrote:
> 
> <snip>
> > @@ -212,8 +198,8 @@ static int ocfs2_recovery_thread_running(struct
> > ocfs2_super *osb)
> > 
> >  void ocfs2_recovery_exit(struct ocfs2_super *osb)
> >  {
> > -	struct ocfs2_recovery_map *rm;
> > -
> > +	struct list_head *iter;
> > +	struct ocfs2_recover_node *rn;
> >  	/* disable any new recovery threads and wait for any currently
> >  	 * running ones to exit. Do this before setting the vol_state. */
> >  	mutex_lock(&osb->recovery_lock);
> > @@ -226,75 +212,66 @@ void ocfs2_recovery_exit(struct ocfs2_super *osb)
> >  	 * complete. */
> >  	flush_workqueue(ocfs2_wq);
> > 
> > -	/*
> > -	 * Now that recovery is shut down, and the osb is about to be
> > -	 * freed,  the osb_lock is not taken here.
> > -	 */
> > -	rm = osb->recovery_map;
> > -	/* XXX: Should we bug if there are dirty entries? */
> > -
> > -	kfree(rm);
> > +	spin_lock(&osb->osb_lock);
> > +	list_for_each(iter, &osb->s_recovery_nodes) {
> 
> list_for_each_safe()?
> 
> regards,
> wengang.
> > +			rn = list_entry(iter, struct ocfs2_recover_node,
> > +					rn_list);
> > +			list_del(&rn->rn_list);

If you need to do this here, I guess you may also want to free rm.

regards,
wengang.
> > +	}
> > +	spin_unlock(&osb->osb_lock);
> >  }
> >