On 2023/4/11 21:12, Michal Hocko wrote:
On Tue 11-04-23 21:04:18, Gang Li wrote:
On 2023/4/11 20:23, Michal Koutný wrote:
Hello.
On Tue, Apr 11, 2023 at 02:58:15PM +0800, Gang Li <ligang.bdlg@xxxxxxxxxxxxx> wrote:
+ cpuset_for_each_descendant_pre(cs, pos_css, &top_cpuset) {
+ if (nodes_equal(cs->mems_allowed, task_cs(current)->mems_allowed)) {
+ css_task_iter_start(&(cs->css), CSS_TASK_ITER_PROCS, &it);
+ while (!ret && (task = css_task_iter_next(&it)))
+ ret = fn(task, arg);
+ css_task_iter_end(&it);
+ }
+ }
+ rcu_read_unlock();
+ cpuset_read_unlock();
+ return ret;
+}
I see this traverses all cpusets without the hierarchy actually
mattering that much. Wouldn't the CONSTRAINT_CPUSET better achieved by
globally (or per-memcg) scanning all processes and filtering with:
Oh I see, you mean scanning all processes in all cpusets and scanning
all processes globally are equivalent.
Why cannot you simple select a process from the cpuset the allocating
process belongs to? I thought the whole idea was to handle well
partitioned workloads.
Yes I can :) It's much easier.
nodes_intersect(current->mems_allowed, p->mems_allowed
Perhaps it would be better to use nodes_equal first, and if no suitable
victim is found, then downgrade to nodes_intersect?
How can this happen?
NUMA balancing mechanism tends to keep memory on the same NUMA node, and
if the selected victim's memory happens to be on a node that does not
intersect with the current process's node, we still won't be able to
free up any memory.
AFAIR NUMA balancing doesn't touch processes with memory policies.
