@@ -245,5 +245,13 @@ static inline const struct cpumask *cpu_cpu_mask(int cpu)
return cpumask_of_node(cpu_to_node(cpu));
}
+#ifdef CONFIG_NUMA
+int sched_numa_find_nth_cpu(const struct cpumask *cpus, int cpu, int node);
+#else
+static __always_inline int sched_numa_find_nth_cpu(const struct cpumask *cpus, int cpu, int node)
+{
+ return cpumask_nth(cpu, cpus);
+}
+#endif /* CONFIG_NUMA */
#endif /* _LINUX_TOPOLOGY_H */
@@ -1764,6 +1764,8 @@ bool find_numa_distance(int distance)
* there is an intermediary node C, which is < N hops away from both
* nodes A and B, the system is a glueless mesh.
*/
+#include <linux/bsearch.h>
+
static void init_numa_topology_type(int offline_node)
{
int a, b, c, n;
@@ -2067,6 +2069,61 @@ int sched_numa_find_closest(const struct cpumask *cpus, int cpu)
return found;
}
+struct __cmp_key {
+ const struct cpumask *cpus;
+ struct cpumask ***masks;
+ int node;
+ int cpu;
+ int w;
+};
+
+static int cmp(const void *a, const void *b)
+{
+ struct cpumask **prev_hop = *((struct cpumask ***)b - 1);
+ struct cpumask **cur_hop = *(struct cpumask ***)b;
+ struct __cmp_key *k = (struct __cmp_key *)a;
+
+ if (cpumask_weight_and(k->cpus, cur_hop[k->node]) <= k->cpu)
+ return 1;
+
+ k->w = (b == k->masks) ? 0 : cpumask_weight_and(k->cpus, prev_hop[k->node]);
+ if (k->w <= k->cpu)
+ return 0;
+
+ return -1;
+}
+
+/*
+ * sched_numa_find_nth_cpu() - given the NUMA topology, find the Nth next cpu
+ * closest to @cpu from @cpumask.
+ * cpumask: cpumask to find a cpu from
+ * cpu: Nth cpu to find
+ *
+ * returns: cpu, or nr_cpu_ids when nothing found.
+ */
+int sched_numa_find_nth_cpu(const struct cpumask *cpus, int cpu, int node)
+{
+ struct __cmp_key k = { .cpus = cpus, .node = node, .cpu = cpu };
+ struct cpumask ***hop_masks;
+ int hop, ret = nr_cpu_ids;
+
+ rcu_read_lock();
+
+ k.masks = rcu_dereference(sched_domains_numa_masks);
+ if (!k.masks)
+ goto unlock;
+
+ hop_masks = bsearch(&k, k.masks, sched_domains_numa_levels, sizeof(k.masks[0]), cmp);
+ hop = hop_masks - k.masks;
+
+ ret = hop ?
+ cpumask_nth_and_andnot(cpu - k.w, cpus, k.masks[hop][node], k.masks[hop-1][node]) :
+ cpumask_nth_and(cpu, cpus, k.masks[0][node]);
+unlock:
+ rcu_read_unlock();
+ return ret;
+}
+EXPORT_SYMBOL_GPL(sched_numa_find_nth_cpu);
#endif /* CONFIG_NUMA */
static int __sdt_alloc(const struct cpumask *cpu_map)
The function finds Nth set CPU in a given cpumask starting from a given node. Leveraging the fact that each hop in sched_domains_numa_masks includes the same or greater number of CPUs than the previous one, we can use binary search on hops instead of linear walk, which makes the overall complexity of O(log n) in terms of number of cpumask_weight() calls. Signed-off-by: Yury Norov <yury.norov@gmail.com> --- include/linux/topology.h | 8 ++++++ kernel/sched/topology.c | 57 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 65 insertions(+)