Sachin recently reported that linux-next was no more bootable on few systems. https://lore.kernel.org/linux-next/3381CD91-AB3D-4773-BA04-E7A072A63968@xxxxxxxxxxxxxxxxxx/ # numactl -H available: 2 nodes (0-1) node 0 cpus: node 0 size: 0 MB node 0 free: 0 MB node 1 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 node 1 size: 35247 MB node 1 free: 30907 MB node distances: node 0 1 0: 10 40 1: 40 10 # Sachin bisected the problem to Commit a75056fc1e7c ("mm/memcontrol.c: allocate shrinker_map on appropriate NUMA node") The root cause analysis showed that mm/slub and powerpc/numa had some shortcomings with respect to offline nodes. This patch series is on top of patches posted at https://lore.kernel.org/linuxppc-dev/20200311110237.5731-1-srikar@xxxxxxxxxxxxxxxxxx/t/#u Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> Cc: linux-mm@xxxxxxxxx Cc: Mel Gorman <mgorman@xxxxxxx> Cc: Michael Ellerman <mpe@xxxxxxxxxxxxxx> Cc: Sachin Sant <sachinp@xxxxxxxxxxxxxxxxxx> Cc: Michal Hocko <mhocko@xxxxxxxxxx> Cc: Christopher Lameter <cl@xxxxxxxxx> Cc: linuxppc-dev@xxxxxxxxxxxxxxxx Cc: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx> Cc: Kirill Tkhai <ktkhai@xxxxxxxxxxxxx> Cc: Vlastimil Babka <vbabka@xxxxxxx> Cc: Srikar Dronamraju <srikar@xxxxxxxxxxxxxxxxxx> Cc: Bharata B Rao <bharata@xxxxxxxxxxxxx> Srikar Dronamraju (4): mm: Check for node_online in node_present_pages mm/slub: Use mem_node to allocate a new slab mm: Implement reset_numa_mem powerpc/numa: Set fallback nodes for offline nodes arch/powerpc/mm/numa.c | 11 ++++++++++- include/asm-generic/topology.h | 3 +++ include/linux/mmzone.h | 6 ++++-- include/linux/topology.h | 7 +++++++ mm/slub.c | 19 ++++++++----------- 5 files changed, 32 insertions(+), 14 deletions(-) -- 2.18.1
