Correct pernodesize calculation.
PATCH 1/2 Correct pernodesize calculation.
A simple fix. The existing pernodesize reservation is not taking into account a second array of pg_data_t structures. This is normally not important because the PAGE_ALIGN macro reserves adequate space.
I made the compute_pernodesize steps in the same order as the fill_pernode steps to make the correlation more clear.
Index: per_cpu_mca_v1/arch/ia64/mm/discontig.c
=================================--- per_cpu_mca_v1.orig/arch/ia64/mm/discontig.c 2008-02-13 06:56:45.000000000 -0600
+++ per_cpu_mca_v1/arch/ia64/mm/discontig.c 2008-02-13 07:12:44.847819507 -0600
@@ -124,6 +124,7 @@ static unsigned long __meminit compute_p
pernodesize += node * L1_CACHE_BYTES;
pernodesize += L1_CACHE_ALIGN(sizeof(pg_data_t)); pernodesize += L1_CACHE_ALIGN(sizeof(struct ia64_node_data));
+ pernodesize += L1_CACHE_ALIGN(sizeof(pg_data_t));
pernodesize = PAGE_ALIGN(pernodesize);
return pernodesize;
}