[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200511130327.GA2978@hirez.programming.kicks-ass.net>
Date: Mon, 11 May 2020 15:03:27 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: bsegall@...gle.com
Cc: Huaixin Chang <changhuaixin@...ux.alibaba.com>,
chiluk+linux@...eed.com, linux-kernel@...r.kernel.org,
mingo@...hat.com, pauld@...head.com, vincent.guittot@...aro.org
Subject: Re: [PATCH v2] sched: Defend cfs and rt bandwidth quota against
overflow
On Mon, Apr 27, 2020 at 11:29:30AM -0700, bsegall@...gle.com wrote:
> Huaixin Chang <changhuaixin@...ux.alibaba.com> writes:
>
> > When users write some huge number into cpu.cfs_quota_us or
> > cpu.rt_runtime_us, overflow might happen during to_ratio() shifts of
> > schedulable checks.
> >
> > to_ratio() could be altered to avoid unnecessary internal overflow, but
> > min_cfs_quota_period is less than 1 << BW_SHIFT, so a cutoff would still
> > be needed. Set a cap MAX_BW for cfs_quota_us and rt_runtime_us to
> > prevent overflow.
>
> Reviewed-by: Ben Segall <bsegall@...gle.com>
Thanks!
Powered by blists - more mailing lists