[PATCH v5 7/7] locking/lockdep: Add a fast path for chain_hlocks allocation

[Waiman Long]

When alloc_chain_hlocks() is called, the most likely scenario is
to allocate from the primordial chain block which holds the whole
chain_hlocks[] array initially. It is the primordial chain block if its
size is bigger than MAX_LOCK_DEPTH. As long as the number of entries left
after splitting is still bigger than MAX_CHAIN_BUCKETS it will remain
in bucket 0. By splitting out a sub-block at the end, we only need to
adjust the size without changing any of the existing linkage information.
This optimized fast path can reduce the latency of allocation requests.

This patch does change the order by which chain_hlocks entries are
allocated. The original code allocates entries from the beginning of
the array. Now it will be allocated from the end of the array backward.

Signed-off-by: Waiman Long <longman@xxxxxxxxxx>
---
 kernel/locking/lockdep.c | 24 ++++++++++++++++++++----
 1 file changed, 20 insertions(+), 4 deletions(-)

diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
index 70288bb2331d..d05799872882 100644
--- a/kernel/locking/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -2701,15 +2701,19 @@ static inline int chain_block_size(int offset)
 	return (chain_hlocks[offset + 2] << 16) | chain_hlocks[offset + 3];
 }
 
+static inline void init_chain_block_size(int offset, int size)
+{
+	chain_hlocks[offset + 2] = size >> 16;
+	chain_hlocks[offset + 3] = (u16)size;
+}
+
 static inline void init_chain_block(int offset, int next, int bucket, int size)
 {
 	chain_hlocks[offset] = (next >> 16) | CHAIN_BLK_FLAG;
 	chain_hlocks[offset + 1] = (u16)next;
 
-	if (bucket == 0) {
-		chain_hlocks[offset + 2] = size >> 16;
-		chain_hlocks[offset + 3] = (u16)size;
-	}
+	if (bucket == 0)
+		init_chain_block_size(offset, size);
 }
 
 static inline void add_chain_block(int offset, int size)
@@ -2809,6 +2813,18 @@ static int alloc_chain_hlocks(int req)
 			return curr;
 		}
 
+		/*
+		 * Fast path: splitting out a sub-block at the end of the
+		 * primordial chain block.
+		 */
+		if (likely((size > MAX_LOCK_DEPTH) &&
+			   (size - req > MAX_CHAIN_BUCKETS))) {
+			size -= req;
+			nr_free_chain_hlocks -= req;
+			init_chain_block_size(curr, size);
+			return curr + size;
+		}
+
 		if (size > max_size) {
 			max_prev = prev;
 			max_curr = curr;
-- 
2.18.1