From f86582890bd7a72a75293e4abbf84c17bca4e754 Mon Sep 17 00:00:00 2001 From: Evangelos Petrongonas Date: Tue, 20 Jan 2026 17:59:11 +0000 Subject: [PATCH] kho: skip memoryless NUMA nodes when reserving scratch areas [ Upstream commit 427b2535f51342de3156babc6bdc3f3b7dd2c707 ] kho_reserve_scratch() iterates over all online NUMA nodes to allocate per-node scratch memory. On systems with memoryless NUMA nodes (nodes that have CPUs but no memory), memblock_alloc_range_nid() fails because there is no memory available on that node. This causes KHO initialization to fail and kho_enable to be set to false. Some ARM64 systems have NUMA topologies where certain nodes contain only CPUs without any associated memory. These configurations are valid and should not prevent KHO from functioning. Fix this by only counting nodes that have memory (N_MEMORY state) and skip memoryless nodes in the per-node scratch allocation loop. Link: https://lkml.kernel.org/r/20260120175913.34368-1-epetron@amazon.de Fixes: 3dc92c311498 ("kexec: add Kexec HandOver (KHO) generation helpers"). Signed-off-by: Evangelos Petrongonas Reviewed-by: Pratyush Yadav Reviewed-by: Mike Rapoport (Microsoft) Reviewed-by: Pasha Tatashin Cc: Alexander Graf Cc: Signed-off-by: Andrew Morton Signed-off-by: Sasha Levin --- kernel/kexec_handover.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/kernel/kexec_handover.c b/kernel/kexec_handover.c index 03d12e27189f..2da4bd994322 100644 --- a/kernel/kexec_handover.c +++ b/kernel/kexec_handover.c @@ -619,7 +619,7 @@ static void __init kho_reserve_scratch(void) scratch_size_update(); /* FIXME: deal with node hot-plug/remove */ - kho_scratch_cnt = num_online_nodes() + 2; + kho_scratch_cnt = nodes_weight(node_states[N_MEMORY]) + 2; size = kho_scratch_cnt * sizeof(*kho_scratch); kho_scratch = memblock_alloc(size, PAGE_SIZE); if (!kho_scratch) @@ -649,7 +649,11 @@ static void __init kho_reserve_scratch(void) kho_scratch[i].size = size; i++; - for_each_online_node(nid) { + /* + * Loop over nodes that have both memory and are online. Skip + * memoryless nodes, as we can not allocate scratch areas there. + */ + for_each_node_state(nid, N_MEMORY) { size = scratch_size_node(nid); addr = memblock_alloc_range_nid(size, CMA_MIN_ALIGNMENT_BYTES, 0, MEMBLOCK_ALLOC_ACCESSIBLE,