Re: [PATCH] sched: fix infinity loop in update_blocked_averages
From: Vincent Guittot
Date: Fri Dec 28 2018 - 13:04:38 EST
On Fri, 28 Dec 2018 at 18:46, Tejun Heo <tj@xxxxxxxxxx> wrote:
>
> On Fri, Dec 28, 2018 at 06:25:37PM +0100, Vincent Guittot wrote:
> > > done without extra space as long as each node has the parent pointer,
> > > which they do. Is the dedicated list an optimization?
> >
> > It prevents to parse and walk all task group struct every time.
> > Instead, you just have to follow a linked list
>
> Hmmm... I'm having a bit of a hard time imagining doing an actual
> traversal being a meaningful optimization. It may require more
> branches but that shouldn't be expensive at all, especially compared
> to walking all idle groups in the system each time which the code used
> to do. Anyways, this is tangential.
>
> Thanks for the explanation and happy new year!
Happy new year for you too
Thanks
Vincent
>
> --
> tejun