On Fri, Apr 18, 2014 at 10:26:28AM -0700, Andi Kleen wrote: > Mel Gorman <mgorman@xxxxxxx> writes: > > > zone_reclaim_mode causes processes to prefer reclaiming memory from local > > node instead of spilling over to other nodes. This made sense initially when > > NUMA machines were almost exclusively HPC and the workload was partitioned > > into nodes. The NUMA penalties were sufficiently high to justify reclaiming > > the memory. On current machines and workloads it is often the case that > > zone_reclaim_mode destroys performance but not all users know how to detect > > this. > > Non local memory also often destroys performance. > True, but if they are sophisticated enough to detect it, they should also know about the tunable. > > Favour the common case and disable it by default. Users that are > > sophisticated enough to know they need zone_reclaim_mode will detect it. > > While I'm not totally against this change, it will destroy many > carefully tuned configurations as the default NUMA behavior may be completely > different now. So it seems like a big hammer, and it's not even clear > what problem you're exactly solving here. > It's a sysctl entry for them to add. The problem is that many users do not know or cannot detect why page reclaim is happening early. They do not have the people on staff to detect it where as the NUMA people appear to generally do. I see bugs semi-regularly on the problem albeit generally against the distribution rather than upstream. -- Mel Gorman SUSE Labs -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
