Messages in this thread | | | From | Paul Turner <> | Date | Tue, 27 Aug 2013 02:05:06 -0700 | Subject | Re: [PATCH 09/10] sched, fair: Fix the sd_parent_degenerate() code |
| |
On Mon, Aug 26, 2013 at 2:49 PM, Rik van Riel <riel@surriel.com> wrote: > On 08/26/2013 08:09 AM, Peter Zijlstra wrote: >> On Sat, Aug 24, 2013 at 03:45:57AM -0700, Paul Turner wrote: >>>> @@ -5157,6 +5158,13 @@ cpu_attach_domain(struct sched_domain *s >>>> tmp->parent = parent->parent; >>>> if (parent->parent) >>>> parent->parent->child = tmp; >>>> + /* >>>> + * Transfer SD_PREFER_SIBLING down in case of a >>>> + * degenerate parent; the spans match for this >>>> + * so the property transfers. >>>> + */ >>>> + if (parent->flags & SD_PREFER_SIBLING) >>>> + tmp->flags |= SD_PREFER_SIBLING; >>>> destroy_sched_domain(parent, cpu); >>>> } else >>>> tmp = tmp->parent; >>>> >>> >>> Reviewed-by: Paul Turner <pjt@google.com> >> >> BTW, did that comment make sense to you or would you suggest something >> different? I had/am having a hard time with that comment. Somehow it >> leaves me wanting. I know I understand the issue now, but I'll doubt the >> comment will suffice in a years time :/ > > The comment made sense to me :)
It makes sense once you read the code. That we push down is somehow counter intuitive in the reading.
> > -- > All rights reversed.
| |