On Sun, Jun 04, 2017 at 11:09:42PM +0300, Vladimir Davydov wrote:
> On Sun, Jun 04, 2017 at 01:04:37PM -0700, Yu Zhao wrote:
> > @@ -2498,22 +2449,24 @@ static int mem_cgroup_resize_memsw_limit(struct 
> > mem_cgroup *memcg,
> >             }
> >  
> >             mutex_lock(&memcg_limit_mutex);
> > -           if (limit < memcg->memory.limit) {
> > +           inverted = memsw ? limit < memcg->memory.limit :
> > +                              limit > memcg->memsw.limit;
> > +           if (inverted)
> >                     mutex_unlock(&memcg_limit_mutex);
> >                     ret = -EINVAL;
> >                     break;
> >             }
> 
> For some reason, I liked this patch more without this extra variable :-)
Well, I'll refrain myself from commenting more because we are now at
the risk of starting a coding style war over this.

Reply via email to