On Fri, Jan 16, 2026 at 11:57:14AM +0000, Pratyush Yadav wrote: > Hi Evangelos, > > On Fri, Jan 16 2026, Evangelos Petrongonas wrote: > > > 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 introducing kho_mem_nodes_count() which counts only nodes > > that have memory (N_MEMORY state), and skip memoryless nodes in the > > per-node scratch allocation loop. > > > > Signed-off-by: Evangelos Petrongonas <[email protected]> > > --- > > kernel/liveupdate/kexec_handover.c | 23 ++++++++++++++++++++++- > > 1 file changed, 22 insertions(+), 1 deletion(-) > > > > diff --git a/kernel/liveupdate/kexec_handover.c > > b/kernel/liveupdate/kexec_handover.c > > index 9dc51fab604f..c970ed08b477 100644 > > --- a/kernel/liveupdate/kexec_handover.c > > +++ b/kernel/liveupdate/kexec_handover.c > > @@ -623,6 +623,23 @@ static phys_addr_t __init scratch_size_node(int nid) > > return round_up(size, CMA_MIN_ALIGNMENT_BYTES); > > } > > > > +/* > > + * Count online NUMA nodes that have memory. Memoryless nodes cannot have > > + * scratch memory and should be excluded. > > + */ > > +static unsigned int __init kho_mem_nodes_count(void) > > +{ > > + unsigned int cnt = 0; > > + int nid; > > + > > + for_each_online_node(nid) { > > + if (node_state(nid, N_MEMORY)) > > + cnt++; > > + } > > + > > + return cnt; > > +} > > + > > You don't need this. You can use nodes_weight(nodes_state[N_MEMORY]) > directly. Other than this, LGTM. > > > /** > > * kho_reserve_scratch - Reserve a contiguous chunk of memory for kexec > > * > > @@ -643,7 +660,7 @@ 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 = kho_mem_nodes_count() + 2; > > size = kho_scratch_cnt * sizeof(*kho_scratch); > > kho_scratch = memblock_alloc(size, PAGE_SIZE); > > if (!kho_scratch) > > @@ -674,6 +691,10 @@ static void __init kho_reserve_scratch(void) > > i++; > > > > for_each_online_node(nid) { > > + /* Skip memoryless nodes - we cannot allocate scratch memory > > there */ > > + if (!node_state(nid, N_MEMORY)) > > + continue; > > +
And here you can use 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, > > -- > Regards, > Pratyush Yadav -- Sincerely yours, Mike.
