Messages in this thread | | | Subject | Re: [patch] sched: fix SMT scheduler regression in find_busiest_queue() | From | Peter Zijlstra <> | Date | Tue, 16 Feb 2010 19:48:55 +0100 |
| |
On Tue, 2010-02-16 at 23:55 +0530, Vaidyanathan Srinivasan wrote: > The sched_domain will be rebuilt with the sched_group_cpus() > representing only online siblings right? sched_group_cpus(group) will > always be a subset of cpu_active_mask. Can please explain your > comment.
__build_*_sched_domain() seems to only rebuild the sd->span, not the sched_group's mask, cpu_to_*_group() only picks an existing group based on the cpumask passed in, it doesn't change sg->cpumask afaict.
That is also the reason we drag load_balance_tmpmask all through load_balance() afaict.
| |