mm/hotplug: correctly setup fallback zonelists when creating new pgdat
When hotadd_new_pgdat() is called to create new pgdat for a new node, a
fallback zonelist should be created for the new node. There's code to try
to achieve that in hotadd_new_pgdat() as below:
/*
* The node we allocated has no zone fallback lists. For avoiding
* to access not-initialized zonelist, build here.
*/
mutex_lock(&zonelists_mutex);
build_all_zonelists(pgdat, NULL);
mutex_unlock(&zonelists_mutex);
But it doesn't work as expected. When hotadd_new_pgdat() is called, the
new node is still in offline state because node_set_online(nid) hasn't
been called yet. And build_all_zonelists() only builds zonelists for
online nodes as:
for_each_online_node(nid) {
pg_data_t *pgdat = NODE_DATA(nid);
build_zonelists(pgdat);
build_zonelist_cache(pgdat);
}
Though we hope to create zonelist for the new pgdat, but it doesn't. So
add a new parameter "pgdat" the build_all_zonelists() to build pgdat for
the new pgdat too.
Signed-off-by: Jiang Liu <[email protected]>
Signed-off-by: Xishi Qiu <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Rusty Russell <[email protected]>
Cc: Yinghai Lu <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: KAMEZAWA Hiroyuki <[email protected]>
Cc: KOSAKI Motohiro <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Keping Chen <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 6c7e3bd9..9ad6866 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3032,7 +3032,7 @@
user_zonelist_order = oldval;
} else if (oldval != user_zonelist_order) {
mutex_lock(&zonelists_mutex);
- build_all_zonelists(NULL);
+ build_all_zonelists(NULL, NULL);
mutex_unlock(&zonelists_mutex);
}
}
@@ -3415,10 +3415,17 @@
{
int nid;
int cpu;
+ pg_data_t *self = data;
#ifdef CONFIG_NUMA
memset(node_load, 0, sizeof(node_load));
#endif
+
+ if (self && !node_online(self->node_id)) {
+ build_zonelists(self);
+ build_zonelist_cache(self);
+ }
+
for_each_online_node(nid) {
pg_data_t *pgdat = NODE_DATA(nid);
@@ -3463,7 +3470,7 @@
* Called with zonelists_mutex held always
* unless system_state == SYSTEM_BOOTING.
*/
-void __ref build_all_zonelists(void *data)
+void __ref build_all_zonelists(pg_data_t *pgdat, struct zone *zone)
{
set_zonelist_order();
@@ -3475,10 +3482,10 @@
/* we have to stop all cpus to guarantee there is no user
of zonelist */
#ifdef CONFIG_MEMORY_HOTPLUG
- if (data)
- setup_zone_pageset((struct zone *)data);
+ if (zone)
+ setup_zone_pageset(zone);
#endif
- stop_machine(__build_all_zonelists, NULL, NULL);
+ stop_machine(__build_all_zonelists, pgdat, NULL);
/* cpuset refresh routine should be here */
}
vm_total_pages = nr_free_pagecache_pages();
