Messages in this thread | | | Date | Thu, 29 Jan 2015 12:17:39 -0500 | From | Steven Rostedt <> | Subject | Re: [PATCH 5/5] sched/rt: Optimize find_lowest_rq() to select a cache hot cpu |
| |
On Fri, 30 Jan 2015 00:42:47 +0800 Xunlei Pang <pang.xunlei@linaro.org> wrote:
> I think the responsiveness is the most important feature for RT tasks, > so I think: > response latency > cache > SMT in significance.
Unfortunately, sometimes cache affects response latency.
> > I was wondering if we can take the cpuidle state into account like > current find_idlest_cpu() for CFS? > cpupri_find() can be easily modified to indicate the CPUPRI_IDLE case, > then we can select > an optimal idle cpu to improve RT tasks' responsiveness. For other > cases(mostly non-idle cpu),
Even if that idle cpu happens to be on another NUMA node?
-- Steve
> I think we can rely on the existent sched_domain iteraction to select > a cache-hot cpu without > caring too much about SMT.
| |