Linux: Make zfs_prune() fair on NUMA systems

Previous code evicted nr_to_scan items from each node. This not only multiplies the eviction by number of nodes, but may totally exhaust smaller ones, evicting inodes used by the iacive workload and requiring their immediate recreation. This patch spreads the requested eviction between all NUMA nodes proportionally to their evictable counts, which should be closer to expected LRU logic. Signed-off-by: Alexander Motin <[email protected]> Sponsored by: iXsystems, Inc.
truenas · Jul 29, 2024 · 790b3ce · 790b3ce
1 parent b157899
commit 790b3ce
Showing 1 changed file with 13 additions and 5 deletions.
diff --git a/module/os/linux/zfs/zfs_vfsops.c b/module/os/linux/zfs/zfs_vfsops.c
@@ -1282,14 +1282,22 @@ zfs_prune(struct super_block *sb, unsigned long nr_to_scan, int *objects)
 	defined(SHRINK_CONTROL_HAS_NID) && \
 	defined(SHRINKER_NUMA_AWARE)
 	if (shrinker->flags & SHRINKER_NUMA_AWARE) {
+		long tc = 1;
+		for_each_online_node(sc.nid) {
+			long c = shrinker->count_objects(shrinker, &sc);
+			if (c  == 0 || c == SHRINK_EMPTY)
+				continue;
+			tc += c;
+		}
 		*objects = 0;
 		for_each_online_node(sc.nid) {
+			long c = shrinker->count_objects(shrinker, &sc);
+			if (c  == 0 || c == SHRINK_EMPTY)
+				continue;
+			if (c > tc)
+				tc = c;
+			sc.nr_to_scan = mult_frac(nr_to_scan, c, tc) + 1;
 			*objects += (*shrinker->scan_objects)(shrinker, &sc);
-			/*
-			 * reset sc.nr_to_scan, modified by
-			 * scan_objects == super_cache_scan
-			 */
-			sc.nr_to_scan = nr_to_scan;
 		}
 	} else {
 			*objects = (*shrinker->scan_objects)(shrinker, &sc);