Re: [RFC][PATCH 1/4] cgroup: support per cgroup subsys state ID (CSS ID)

[KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>]

On Fri, 09 Jan 2009 11:59:05 +0800
Li Zefan <lizf@cn.fujitsu.com> wrote:

> >  static struct inode *cgroup_new_inode(mode_t mode, struct super_block *sb)
> >  {
> >  	struct inode *inode = new_inode(sb);
> > @@ -2335,6 +2339,7 @@ static void init_cgroup_css(struct cgrou
> >  	css->cgroup = cgrp;
> >  	atomic_set(&css->refcnt, 1);
> >  	css->flags = 0;
> > +	css->id = NULL;
> >  	if (cgrp == dummytop)
> >  		set_bit(CSS_ROOT, &css->flags);
> >  	BUG_ON(cgrp->subsys[ss->subsys_id]);
> > @@ -2410,6 +2415,10 @@ static long cgroup_create(struct cgroup 
> >  			goto err_destroy;
> >  		}
> >  		init_cgroup_css(css, ss, cgrp);
> > +		if (ss->use_id)
> > +			if (alloc_css_id(ss, parent, cgrp))
> > +				goto err_destroy;
> > +		/* At error, ->destroy() callback has to free assigned ID. */
> 
> A bug here:
> 
> if alloc_css_id(ss, parent, cgrp) failed, then ss->destroy() called free_css_id(),
> then panic.
> 
> maybe check if (css->id == NULL) in free_css_id() ?
> 
Oh, thanks. it will be necessary.
I'll fix it.

But..Hmm...maybe it's useful to add fault injection feature to debug cgroup
for this kind of loop operation.

-Kame

> >  	}
> >  
> >  	cgroup_lock_hierarchy(root);
> > @@ -2699,6 +2708,8 @@ int __init cgroup_init(void)
> >  		struct cgroup_subsys *ss = subsys[i];
> >  		if (!ss->early_init)
> >  			cgroup_init_subsys(ss);
> > +		if (ss->use_id)
> > +			cgroup_subsys_init_idr(ss);
> >  	}
> >  
> >  	/* Add init_css_set to the hash table */
> > @@ -3231,3 +3242,260 @@ static int __init cgroup_disable(char *s
> >  	return 1;
> >  }
> >  __setup("cgroup_disable=", cgroup_disable);
> 

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>