mm: export fragmentation index via debugfs

The fragmentation fragmentation index, is only meaningful if an allocation would fail and indicates what the failure is due to. A value of -1 such as in many of the examples above states that the allocation would succeed. If it would fail, the value is between 0 and 1. A value tending towards 0 implies the allocation failed due to a lack of memory. A value tending towards 1 implies it failed due to external fragmentation. For the most part, the huge page size will be the size of interest but not necessarily so it is exported on a per-order and per-zo basis via /sys/kernel/debug/extfrag/extfrag_index > cat /sys/kernel/debug/extfrag/extfrag_index Node 0, zone DMA -1.000 -1.000 -1.000 -1.000 -1.000 -1.000 -1.000 -1.00 Node 0, zone Normal -1.000 -1.000 -1.000 -1.000 -1.000 -1.000 -1.000 0.954 Signed-off-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: Minchan Kim <minchan.kim@gmail.com> Acked-by: Rik van Riel <riel@redhat.com> Reviewed-by: Christoph Lameter <cl@linux-foundation.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: export fragmentation index via debugfs
The fragmentation fragmentation index, is only meaningful if an allocation would fail and indicates what the failure is due to. A value of -1 such as in many of the examples above states that the allocation would succeed. If it would fail, the value is between 0 and 1. A value tending towards 0 implies the allocation failed due to a lack of memory. A value tending towards 1 implies it failed due to external fragmentation. For the most part, the huge page size will be the size of interest but not necessarily so it is exported on a per-order and per-zo basis via /sys/kernel/debug/extfrag/extfrag_index > cat /sys/kernel/debug/extfrag/extfrag_index Node 0, zone DMA -1.000 -1.000 -1.000 -1.000 -1.000 -1.000 -1.000 -1.00 Node 0, zone Normal -1.000 -1.000 -1.000 -1.000 -1.000 -1.000 -1.000 0.954 Signed-off-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: Minchan Kim <minchan.kim@gmail.com> Acked-by: Rik van Riel <riel@redhat.com> Reviewed-by: Christoph Lameter <cl@linux-foundation.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Mel Gorman · Linus Torvalds
1 parent d7a5752c0c
Showing 1 changed file with 84 additions and 0 deletions Side-by-side Diff
mm/vmstat.c
@@ -16,6 +16,7 @@
 #include <linux/cpu.h>
 #include <linux/vmstat.h>
 #include <linux/sched.h>
+#include <linux/math64.h>
  
 #ifdef CONFIG_VM_EVENT_COUNTERS
 DEFINE_PER_CPU(struct vm_event_state, vm_event_states) = {{0}};
@@ -420,6 +421,33 @@
 						(order - suitable_order);
 	}
 }
+
+/*
+ * A fragmentation index only makes sense if an allocation of a requested
+ * size would fail. If that is true, the fragmentation index indicates
+ * whether external fragmentation or a lack of memory was the problem.
+ * The value can be used to determine if page reclaim or compaction
+ * should be used
+ */
+int fragmentation_index(unsigned int order, struct contig_page_info *info)
+{
+	unsigned long requested = 1UL << order;
+
+	if (!info->free_blocks_total)
+		return 0;
+
+	/* Fragmentation index only makes sense when a request would fail */
+	if (info->free_blocks_suitable)
+		return -1000;
+
+	/*
+	 * Index is between 0 and 1 so return within 3 decimal places
+	 *
+	 * 0 => allocation would fail due to lack of memory
+	 * 1 => allocation would fail due to fragmentation
+	 */
+	return 1000 - div_u64( (1000+(div_u64(info->free_pages * 1000ULL, requested))), info->free_blocks_total);
+}
 #endif
  
 #if defined(CONFIG_PROC_FS) || defined(CONFIG_COMPACTION)
@@ -1087,6 +1115,58 @@
 	.release	= seq_release,
 };
  
+static void extfrag_show_print(struct seq_file *m,
+					pg_data_t *pgdat, struct zone *zone)
+{
+	unsigned int order;
+	int index;
+
+	/* Alloc on stack as interrupts are disabled for zone walk */
+	struct contig_page_info info;
+
+	seq_printf(m, "Node %d, zone %8s ",
+				pgdat->node_id,
+				zone->name);
+	for (order = 0; order < MAX_ORDER; ++order) {
+		fill_contig_page_info(zone, order, &info);
+		index = fragmentation_index(order, &info);
+		seq_printf(m, "%d.%03d ", index / 1000, index % 1000);
+	}
+
+	seq_putc(m, '\n');
+}
+
+/*
+ * Display fragmentation index for orders that allocations would fail for
+ */
+static int extfrag_show(struct seq_file *m, void *arg)
+{
+	pg_data_t *pgdat = (pg_data_t *)arg;
+
+	walk_zones_in_node(m, pgdat, extfrag_show_print);
+
+	return 0;
+}
+
+static const struct seq_operations extfrag_op = {
+	.start	= frag_start,
+	.next	= frag_next,
+	.stop	= frag_stop,
+	.show	= extfrag_show,
+};
+
+static int extfrag_open(struct inode *inode, struct file *file)
+{
+	return seq_open(file, &extfrag_op);
+}
+
+static const struct file_operations extfrag_file_ops = {
+	.open		= extfrag_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
 static int __init extfrag_debug_init(void)
 {
 	extfrag_debug_root = debugfs_create_dir("extfrag", NULL);
@@ -1095,6 +1175,10 @@
  
 	if (!debugfs_create_file("unusable_index", 0444,
 			extfrag_debug_root, NULL, &unusable_file_ops))
+		return -ENOMEM;
+
+	if (!debugfs_create_file("extfrag_index", 0444,
+			extfrag_debug_root, NULL, &extfrag_file_ops))
 		return -ENOMEM;
  
 	return 0;
...	...	@@ -16,6 +16,7 @@
16	16	#include <linux/cpu.h>
17	17	#include <linux/vmstat.h>
18	18	#include <linux/sched.h>
	19	+#include <linux/math64.h>
19	20
20	21	#ifdef CONFIG_VM_EVENT_COUNTERS
21	22	DEFINE_PER_CPU(struct vm_event_state, vm_event_states) = {{0}};
...	...	@@ -420,6 +421,33 @@
420	421	(order - suitable_order);
421	422	}
422	423	}
	424	+
	425	+/*
	426	+ * A fragmentation index only makes sense if an allocation of a requested
	427	+ * size would fail. If that is true, the fragmentation index indicates
	428	+ * whether external fragmentation or a lack of memory was the problem.
	429	+ * The value can be used to determine if page reclaim or compaction
	430	+ * should be used
	431	+ */
	432	+int fragmentation_index(unsigned int order, struct contig_page_info *info)
	433	+{
	434	+ unsigned long requested = 1UL << order;
	435	+
	436	+ if (!info->free_blocks_total)
	437	+ return 0;
	438	+
	439	+ /* Fragmentation index only makes sense when a request would fail */
	440	+ if (info->free_blocks_suitable)
	441	+ return -1000;
	442	+
	443	+ /*
	444	+ * Index is between 0 and 1 so return within 3 decimal places
	445	+ *
	446	+ * 0 => allocation would fail due to lack of memory
	447	+ * 1 => allocation would fail due to fragmentation
	448	+ */
	449	+ return 1000 - div_u64( (1000+(div_u64(info->free_pages * 1000ULL, requested))), info->free_blocks_total);
	450	+}
423	451	#endif
424	452
425	453	#if defined(CONFIG_PROC_FS) \|\| defined(CONFIG_COMPACTION)
...	...	@@ -1087,6 +1115,58 @@
1087	1115	.release = seq_release,
1088	1116	};
1089	1117
	1118	+static void extfrag_show_print(struct seq_file *m,
	1119	+ pg_data_t pgdat, struct zone zone)
	1120	+{
	1121	+ unsigned int order;
	1122	+ int index;
	1123	+
	1124	+ /* Alloc on stack as interrupts are disabled for zone walk */
	1125	+ struct contig_page_info info;
	1126	+
	1127	+ seq_printf(m, "Node %d, zone %8s ",
	1128	+ pgdat->node_id,
	1129	+ zone->name);
	1130	+ for (order = 0; order < MAX_ORDER; ++order) {
	1131	+ fill_contig_page_info(zone, order, &info);
	1132	+ index = fragmentation_index(order, &info);
	1133	+ seq_printf(m, "%d.%03d ", index / 1000, index % 1000);
	1134	+ }
	1135	+
	1136	+ seq_putc(m, '\n');
	1137	+}
	1138	+
	1139	+/*
	1140	+ * Display fragmentation index for orders that allocations would fail for
	1141	+ */
	1142	+static int extfrag_show(struct seq_file m, void arg)
	1143	+{
	1144	+ pg_data_t pgdat = (pg_data_t )arg;
	1145	+
	1146	+ walk_zones_in_node(m, pgdat, extfrag_show_print);
	1147	+
	1148	+ return 0;
	1149	+}
	1150	+
	1151	+static const struct seq_operations extfrag_op = {
	1152	+ .start = frag_start,
	1153	+ .next = frag_next,
	1154	+ .stop = frag_stop,
	1155	+ .show = extfrag_show,
	1156	+};
	1157	+
	1158	+static int extfrag_open(struct inode inode, struct file file)
	1159	+{
	1160	+ return seq_open(file, &extfrag_op);
	1161	+}
	1162	+
	1163	+static const struct file_operations extfrag_file_ops = {
	1164	+ .open = extfrag_open,
	1165	+ .read = seq_read,
	1166	+ .llseek = seq_lseek,
	1167	+ .release = seq_release,
	1168	+};
	1169	+
1090	1170	static int __init extfrag_debug_init(void)
1091	1171	{
1092	1172	extfrag_debug_root = debugfs_create_dir("extfrag", NULL);
...	...	@@ -1095,6 +1175,10 @@
1095	1175
1096	1176	if (!debugfs_create_file("unusable_index", 0444,
1097	1177	extfrag_debug_root, NULL, &unusable_file_ops))
	1178	+ return -ENOMEM;
	1179	+
	1180	+ if (!debugfs_create_file("extfrag_index", 0444,
	1181	+ extfrag_debug_root, NULL, &extfrag_file_ops))
1098	1182	return -ENOMEM;
1099	1183
1100	1184	return 0;