[tip: locking/core] locking/osq: Save the CPU number for 'prev', not the node address

[tip-bot2 for David Laight]

The following commit has been merged into the locking/core branch of tip:

Commit-ID:     72c440a4e225879063809c3d0bcdd7e6544069fe
Gitweb:        https://git.kernel.org/tip/72c440a4e225879063809c3d0bcdd7e6544069fe
Author:        David Laight <david.laight.linux@xxxxxxxxx>
AuthorDate:    Mon, 07 Sep 2026 09:41:26 +01:00
Committer:     Ingo Molnar <mingo@xxxxxxxxxx>
CommitterDate: Fri, 06 Nov 2026 11:16:24 +01:00

locking/osq: Save the CPU number for 'prev', not the node address

The CPU number of node->prev is needed for both the vcpu_is_preempted()
test and to update lock->tail. This saves reading the cache line for
the other CPU's per-CPU data.

The CPU member of optimistic_spin_node is no longer needed.

[ peterz: Renamed stuff ]

Signed-off-by: David Laight <david.laight.linux@xxxxxxxxx>
Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
Reviewed-by: Waiman Long <longman@xxxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Link: https://patch.msgid.link/20260907084133.3696-3-david.laight.linux@xxxxxxxxx
---
 kernel/locking/osq_lock.c | 29 ++++++++++++-----------------
 1 file changed, 12 insertions(+), 17 deletions(-)

diff --git a/kernel/locking/osq_lock.c b/kernel/locking/osq_lock.c
index 984fa11..ee1f69b 100644
--- a/kernel/locking/osq_lock.c
+++ b/kernel/locking/osq_lock.c
@@ -13,9 +13,9 @@
  */
 
 struct optimistic_spin_node {
-	struct optimistic_spin_node *next, *prev;
+	struct optimistic_spin_node *next;
 	int locked; /* 1 if lock acquired */
-	int cpu; /* encoded CPU # + 1 value */
+	int prev_cpu; /* CPU number offset by 1 */
 };
 
 static DEFINE_PER_CPU_SHARED_ALIGNED(struct optimistic_spin_node, osq_node);
@@ -29,11 +29,6 @@ static inline int encode_cpu(int cpu_nr)
 	return cpu_nr + 1;
 }
 
-static inline int node_cpu(struct optimistic_spin_node *node)
-{
-	return node->cpu - 1;
-}
-
 static inline struct optimistic_spin_node *decode_cpu(int encoded_cpu_val)
 {
 	int cpu_nr = encoded_cpu_val - 1;
@@ -137,11 +132,10 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 	struct optimistic_spin_node *node = this_cpu_ptr(&osq_node);
 	struct optimistic_spin_node *prev, *next;
 	int curr = encode_cpu(smp_processor_id());
-	int old;
+	int prev_cpu;
 
 	node->locked = 0;
 	node->next = NULL;
-	node->cpu = curr;
 
 	/*
 	 * We need both ACQUIRE (pairs with corresponding RELEASE in
@@ -149,12 +143,12 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 	 * the node fields we just initialised) semantics when updating
 	 * the lock tail.
 	 */
-	old = atomic_xchg(&lock->tail, curr);
-	if (old == OSQ_UNLOCKED_VAL)
+	prev_cpu = atomic_xchg(&lock->tail, curr);
+	if (prev_cpu == OSQ_UNLOCKED_VAL)
 		return true;
 
-	prev = decode_cpu(old);
-	node->prev = prev;
+	prev = decode_cpu(prev_cpu);
+	node->prev_cpu = prev_cpu;
 	osq_link_next(prev, node);
 
 	/*
@@ -173,7 +167,7 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 	 * polling, be careful.
 	 */
 	if (smp_cond_load_relaxed(&node->locked, VAL || need_resched() ||
-				  vcpu_is_preempted(node_cpu(node->prev))))
+				  vcpu_is_preempted(node->prev_cpu - 1)))
 		return true;
 
 	/* unqueue */
@@ -208,7 +202,8 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 		 * Or we race against a concurrent unqueue()'s step-B, in which
 		 * case its step-C will write us a new @node->prev pointer.
 		 */
-		prev = READ_ONCE(node->prev);
+		prev_cpu = READ_ONCE(node->prev_cpu);
+		prev = decode_cpu(prev_cpu);
 	}
 
 	/*
@@ -218,7 +213,7 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 	 * back to @prev.
 	 */
 
-	next = osq_wait_next(lock, node, prev->cpu);
+	next = osq_wait_next(lock, node, prev_cpu);
 	if (!next)
 		return false;
 
@@ -230,7 +225,7 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 	 * it will wait in Step-A.
 	 */
 
-	WRITE_ONCE(next->prev, prev);
+	WRITE_ONCE(next->prev_cpu, prev_cpu);
 	osq_link_next(prev, next);
 
 	return false;