On Wed, 19 Jun 2013, Robin Holt wrote: > cpusets was not for NUMA. It has no preference for "nodes" or anything like > that. It was for splitting a machine into layered smaller groups. Usually, > we see one cpuset with contains the batch scheduler. The batch scheduler then > creates cpusets for jobs it starts. Has nothing to do with nodes. That is > more an administrator issue. They set the minimum grouping of resources > for scheduled jobs. > I disagree with all of the above, it's not what Paul Jackson developed cpusets for, it's not what he wrote in Documentation/cgroups/cpusets.txt, and it's not why libnuma immediately supported it. Cpusets is for NUMA, like it or not. > > I'm saying there's absolutely no reason to have thp controlled by a > > cpuset, or ANY cgroup for that matter, since you chose not to respond to > > the question I asked: why do you want to control thp behavior for certain > > static binaries and not others? Where is the performance regression or > > the downside? Is it because of max_ptes_none for certain jobs blowing up > > the rss? We need information, and even if were justifiable then it > > wouldn't have anything to do with ANY cgroup but rather a per-process > > control. It has nothing to do with cpusets whatsoever. > > It was a request from our benchmarking group that has found some jobs > benefit from thp, while other are harmed. Let me ask them for more > details. > Yes, please, because if some jobs are harmed by thp then we need to fix that regression and not paper around with it with some cpuset-based solution. People should be able to run with CONFIG_TRANSPARENT_HUGEPAGE enabled and not be required to enable CONFIG_CPUSETS for optimal behavior. I'm suspecting that you're referring to enlarged rss because of khugepaged's max_ptes_none and because you're abusing the purpose of cpusets for containerization. -- To unsubscribe from this list: send the line "unsubscribe linux-doc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
