Hi Peter Zijlstra, On 2022/1/12 22:43, Peter Zijlstra wrote: >> Set per-process numa balancing: >> prctl(PR_NUMA_BALANCING, PR_SET_NUMAB_DISABLE); //disable >> prctl(PR_NUMA_BALANCING, PR_SET_NUMAB_ENABLE); //enable >> prctl(PR_NUMA_BALANCING, PR_SET_NUMAB_DEFAULT); //follow global > > This seems to imply you can prctl(ENABLE) even if the global is > disabled, IOW sched_numa_balancing is off. I have discussed the semantics of this API with Mel Gorman, we both agree that we can prctl(ENABLE) even if the global is disabled. On 2021/11/10 00:26, Mel Gorman wrote: [1] > For symmetry and consistency of the tuning. Either there is per-process > control or there is not. Right now, there is only the ability to turn > off NUMA balancing via prctl if globally enabled. There is no option to > turn NUMA balancing on for a single task if globally disabled. On 11/18/21 4:58 PM, Mel Gorman wrote: [2] > On Thu, Nov 18, 2021 at 11:26:30AM +0800, Gang Li wrote: >> 3. prctl(PR_NUMA_BALANCING, PR_SET_NUMAB_ENABLE); //enable > > If PR_SET_NUMAB_ENABLE enables numa balancing for a task when > kernel.numa_balancing == 0 instead of returning an error then sure. [1] Link: https://lore.kernel.org/lkml/20211109162647.GY3891@xxxxxxx/ [2] Link: https://lore.kernel.org/lkml/20211118085819.GD3301@xxxxxxx/ -- Thanks Gang Li
