On Wed, 28 Nov 2007 14:20:33 +0300
Pavel Emelyanov <xemul@openvz.org> wrote:

> > +static ssize_t mem_cgroup_write_limit(struct cgroup *cont, struct cftype *cft,
> > +				struct file *file, const char __user *userbuf,
> > +				size_t nbytes, loff_t *ppos)
> > +{
> > +	ssize_t ret;
> > +	ret = mem_cgroup_write(cont, cft, file, userbuf, nbytes, ppos);
> > +	if (ret > 0)
> > +		mem_cgroup_init_watermark(cont);
> 
> No, please, no! I'd be very disappointed if I tune high and low watermarks
> carefully and then they are silently re-set after I tune the limit. Better
> (see my comment to patch #3) return -EINVAL in case I try to set limit 
> below hwmark. Please :)
>
Ok, I'll drop this.

Thanks,
-Kame

_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers