Hi all,
Currently, in the process of initialization or offline memory, memoryless
nodes will still be built into the fallback list of itself or other nodes.
This is not what we expected, so this patch series removes memoryless
nodes from the fallback list entirely.
This series is based on the next-20231018.
Comments and suggestions are welcome.
Thanks,
Qi
Changlog in v3 -> v4:
- fix typo in the commit messages and comment (pointed by Ingo Molnar)
- collect Acked-bys
Changlog in v2 -> v3:
- add a comment in [PATCH v2 2/2] (suggested by David Hildenbrand)
- collect Acked-bys
Changlog in v1 -> v2:
- modify the commit message in [PATCH 1/2], mention that it can also fix the
specific crash. (suggested by Ingo Molnar)
- rebase onto the next-20231018
Qi Zheng (2):
mm: page_alloc: skip memoryless nodes entirely
mm: memory_hotplug: drop memoryless node from fallback lists
mm/memory_hotplug.c | 6 +++++-
mm/page_alloc.c | 7 +++++--
2 files changed, 10 insertions(+), 3 deletions(-)
--
2.30.2
In offline_pages(), if a node becomes memoryless, we
will clear its N_MEMORY state by calling node_states_clear_node().
But we do this after rebuilding the zonelists by calling
build_all_zonelists(), which will cause this memoryless node to
still be in the fallback list of other nodes. This will incur
some runtime overhead.
To drop memoryless nodes from fallback lists in this case, just
call node_states_clear_node() before calling build_all_zonelists().
Signed-off-by: Qi Zheng <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Acked-by: Ingo Molnar <[email protected]>
---
mm/memory_hotplug.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index d4a364fdaf8f..f019f7d6272c 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -2036,12 +2036,16 @@ int __ref offline_pages(unsigned long start_pfn, unsigned long nr_pages,
/* reinitialise watermarks and update pcp limits */
init_per_zone_wmark_min();
+ /*
+ * Make sure to mark the node as memory-less before rebuilding the zone
+ * list. Otherwise this node would still appear in the fallback lists.
+ */
+ node_states_clear_node(node, &arg);
if (!populated_zone(zone)) {
zone_pcp_reset(zone);
build_all_zonelists(NULL);
}
- node_states_clear_node(node, &arg);
if (arg.status_change_nid >= 0) {
kcompactd_stop(node);
kswapd_stop(node);
--
2.30.2
In find_next_best_node(), we skipped memoryless nodes
when building the zonelists of other normal nodes (N_NORMAL),
but did not skip the memoryless node itself when building
the zonelist. This will cause it to be traversed at runtime.
For example, say we have node0 and node1, node0 is memoryless
node, then the fall back order of node0 and node1 as follows:
[ 0.153005] Fallback order for Node 0: 0 1
[ 0.153564] Fallback order for Node 1: 1
After this patch, we skip memoryless node0 entirely, then
the fall back order of node0 and node1 as follows:
[ 0.155236] Fallback order for Node 0: 1
[ 0.155806] Fallback order for Node 1: 1
So it becomes completely invisible, which will reduce runtime
overhead.
And in this way, we will not try to allocate pages from memoryless
node0, then the panic mentioned in [1] will also be fixed. Even though
this problem has been solved by dropping the NODE_MIN_SIZE constrain
in x86 [2], it would be better to fix it in the core MM as well.
[1]. https://lore.kernel.org/all/[email protected]/
[2]. https://lore.kernel.org/all/[email protected]/
Signed-off-by: Qi Zheng <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Acked-by: Ingo Molnar <[email protected]>
---
mm/page_alloc.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index ee392a324802..1f852929709f 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5052,8 +5052,11 @@ int find_next_best_node(int node, nodemask_t *used_node_mask)
int min_val = INT_MAX;
int best_node = NUMA_NO_NODE;
- /* Use the local node if we haven't already */
- if (!node_isset(node, *used_node_mask)) {
+ /*
+ * Use the local node if we haven't already, but for memoryless local
+ * node, we should skip it and fall back to other nodes.
+ */
+ if (!node_isset(node, *used_node_mask) && node_state(node, N_MEMORY)) {
node_set(node, *used_node_mask);
return node;
}
--
2.30.2