Re: [PATCH, v3 2/2] cgroups: introduce timer slack subsystem
From: Kirill A. Shutemov
Date: Mon Feb 07 2011 - 04:48:56 EST
On Sat, Feb 05, 2011 at 06:49:51PM -0800, Matt Helsley wrote:
> On Thu, Feb 03, 2011 at 11:41:38AM +0200, Kirill A. Shutemov wrote:
> > On Wed, Feb 02, 2011 at 09:46:16PM -0800, Matt Helsley wrote:
> > > On Wed, Feb 02, 2011 at 10:47:36PM +0200, Kirill A. Shutsemov wrote:
> > > > From: Kirill A. Shutemov <kirill@xxxxxxxxxxxxx>
>
> <snip>
>
> > > > diff --git a/kernel/cgroup_timer_slack.c b/kernel/cgroup_timer_slack.c
> > > > new file mode 100644
> > > > index 0000000..a343a50
> > > > --- /dev/null
> > > > +++ b/kernel/cgroup_timer_slack.c
>
> <snip>
>
> > > > +static int tslack_write_set_slack_ns(struct cgroup *cgroup, struct cftype *cft,
> > > > + u64 val)
> > > > +{
> > > > + struct timer_slack_cgroup *tslack_cgroup;
> > > > + struct cgroup_iter it;
> > > > + struct task_struct *task;
> > > > +
> > > > + tslack_cgroup = cgroup_to_tslack_cgroup(cgroup);
> > > > + if (!val || val < tslack_cgroup->min_slack_ns ||
> > >
> > > Why is a val of 0 disallowed? I know having slack is good, but for
> > > an administrator or tool that doesn't care about number of wakeups
> > > and cares more about wringing out performance a slack of
> > > 0 seems acceptable. Is this just here to be consistent with the
> > > values passed in via prctl?
> >
> > Yes, it's to consistent with the prctl(). I don't think that it's good
> > idea to allow to set timer_slack outside of range prctl() allows. It may
> > lead to interface abuse.
>
> Hmm, I was just thinking that 0 timer slack might be useful. But I
> suppose you could just as easily set it to 1 and nobody would notice.
I've rechecked once again. it lookes cleaner to allow 0 as timer slack
value.
I allowed it in version 4 of the patchset.
--
Kirill A. Shutemov
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/