Re: [PATCH 4/5] cgroup: remove extra calls to find_existing_css_set

[Mandeep Singh Baines <msb-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>]

Tejun Heo (tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org) wrote:
> Hello,
> 
> On Tue, Dec 20, 2011 at 03:14:32PM -0800, Mandeep Singh Baines wrote:
> > @@ -1942,9 +1917,17 @@ int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk)
> >  		}
> >  	}
> >  
> > -	retval = cgroup_task_migrate(cgrp, oldcgrp, tsk, false);
> > -	if (retval)
> > +	task_lock(tsk);
> > +	oldcg = tsk->cgroups;
> > +	task_unlock(tsk);
> 
> Probably this is patch order issue w/ Frederic's patch but we don't
> need task_lock here, right?
> 

Yeah, avoided removing the locks because Frederic's patch already did that.
Once his changes get in, I'll rebase and resend.

> > @@ -2171,17 +2088,18 @@ int cgroup_attach_proc(struct cgroup *cgrp, struct task_struct *leader)
> >  	 * step 2: make sure css_sets exist for all threads to be migrated.
> >  	 * we use find_css_set, which allocates a new one if necessary.
> >  	 */
> > -	INIT_LIST_HEAD(&newcg_list);
> >  	for (i = 0; i < group_size; i++) {
> >  		tc = flex_array_get(group, i);
> >  		/* get old css_set pointer */
> >  		task_lock(tc->task);
> >  		oldcg = tc->task->cgroups;
> >  		task_unlock(tc->task);
> > -		/* if we don't already have it in the list get a new one */
> > -		if (!css_set_fetched(cgrp, tc->task, oldcg, &newcg_list))
> > -			if (retval = css_set_prefetch(cgrp, oldcg, &newcg_list))
> > -				goto out_list_teardown;
> > +		tc->cg = find_css_set(oldcg, cgrp);
> > +		if (!tc->cg) {
> > +			retval = -ENOMEM;
> > +			css_set_refs = i + 1;
> > +			goto out_list_teardown;
> > +		}
> 
> Nice cleanup, but can't the above be merged into the loop which builds
> the flex array?
> 

Good idea. That would remove a for loop. Will fix in next version.

> Thanks.
> 
> -- 
> tejun