Messages in this thread | | | Date | Mon, 21 Aug 2006 15:54:30 +0200 (CEST) | From | Simon Derr <> | Subject | Re: cpusets not cpu hotplug aware |
| |
On Mon, 21 Aug 2006, Anton Blanchard wrote:
> > Hi, > > We have a bug report where sched_setaffinity fails for cpus that are > hotplug added after boot. Nathan found this suspicious piece of code: > > void __init cpuset_init_smp(void) > { > top_cpuset.cpus_allowed = cpu_online_map; > top_cpuset.mems_allowed = node_online_map; > } > > Any reason we statically set the top level cpuset to the boot time cpu > online map? >
Are you suggesting that all possible cpus should be in the top cpuset, instead of all existing cpus ?
I'd prefer the hotplugged CPUs being added to the top cpuset afterwards.
Simon.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |