memblock: avoid some repeat when add new range

The worst case is that the new memory range overlaps all existing
regions, which requires type->cnt + 1 empty struct memblock_region slots in
the type->regions array.
So if type->cnt + 1 + type->cnt is less than type->max, we can insert
regions directly rather than calculate the needed amount before the
insertion.
And becase of merge operation in the end of function, tpye->cnt will
increase slowly for many cases.

This change allows to avoid unnecessary repeat of memblock ranges traversal
for many cases when adding new memory range.

Signed-off-by: Jinyu Tang <tjytimi@163.com>
[rppt: massaged comment and changelog text]
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>

1 files changed, 11 insertions, 0 deletions

diff --git a/mm/memblock.c b/mm/memblock.c
index e4f03a6e8e56..16f006521afa 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -593,6 +593,17 @@ static int __init_memblock memblock_add_range(struct memblock_type *type,
 		type->total_size = size;
 		return 0;
 	}
+
+	/*
+	 * The worst case is when new range overlaps all existing regions,
+	 * then we'll need type->cnt + 1 empty regions in @type. So if
+	 * type->cnt * 2 + 1 is less than type->max, we know
+	 * that there is enough empty regions in @type, and we can insert
+	 * regions directly.
+	 */
+	if (type->cnt * 2 + 1 < type->max)
+		insert = true;
+
 repeat:
 	/*
 	 * The following is executed twice.  Once with %false @insert and