rqspinlock: Use smp_cond_load_acquire_timeout()

Switch out the conditional load interfaces used by rqspinlock
to atomic_cond_read_acquire_timeout(), and
smp_cond_read_acquire_timeout().

Both these handle the timeout and amortize as needed, so use
check_timeout() directly.

Also, when using spin-wait implementations, redefine SMP_TIMEOUT_POLL_COUNT
to be 16k to be similar to the spin-count used in RES_CHECK_TIMEOUT().

Cc: Kumar Kartikeya Dwivedi <[email protected]>
Cc: Alexei Starovoitov <[email protected]>
Signed-off-by: Ankur Arora <[email protected]>

Author: terminus

kernel/bpf/rqspinlock.c

@@ -238,20 +238,14 @@ static noinline int check_timeout(rqspinlock_t *lock, u32 mask,
 }
 
 /*
- * Do not amortize with spins when res_smp_cond_load_acquire is defined,
- * as the macro does internal amortization for us.
+ * Amortize timeout check for busy-wait loops.
  */
-#ifndef res_smp_cond_load_acquire
 #define RES_CHECK_TIMEOUT(ts, ret, mask)                              \
 	({                                                            \
 		if (!(ts).spin++)                                     \
 			(ret) = check_timeout((lock), (mask), &(ts)); \
 		(ret);                                                \
 	})
-#else
-#define RES_CHECK_TIMEOUT(ts, ret, mask)			      \
-	({ (ret) = check_timeout((lock), (mask), &(ts)); })
-#endif
 
 /*
  * Initialize the 'spin' member.
@@ -265,6 +259,16 @@ static noinline int check_timeout(rqspinlock_t *lock, u32 mask,
  */
 #define RES_RESET_TIMEOUT(ts, _duration) ({ (ts).timeout_end = 0; (ts).duration = _duration; })
 
+/*
+ * Limit how often check_timeout() is invoked while spin-waiting in
+ * smp_cond_load_acquire_timeout() or atomic_cond_read_acquire_timeout().
+ * (ARM64, typically uses a waited implementation so we exclude that.)
+ */
+#ifndef CONFIG_ARM64
+#undef SMP_TIMEOUT_POLL_COUNT
+#define SMP_TIMEOUT_POLL_COUNT	(16*1024)
+#endif
+
 /*
  * Provide a test-and-set fallback for cases when queued spin lock support is
  * absent from the architecture.
@@ -310,12 +314,6 @@ EXPORT_SYMBOL_GPL(resilient_tas_spin_lock);
  */
 static DEFINE_PER_CPU_ALIGNED(struct qnode, rqnodes[_Q_MAX_NODES]);
 
-#ifndef res_smp_cond_load_acquire
-#define res_smp_cond_load_acquire(v, c) smp_cond_load_acquire(v, c)
-#endif
-
-#define res_atomic_cond_read_acquire(v, c) res_smp_cond_load_acquire(&(v)->counter, (c))
-
 /**
  * resilient_queued_spin_lock_slowpath - acquire the queued spinlock
  * @lock: Pointer to queued spinlock structure
@@ -415,7 +413,8 @@ int __lockfunc resilient_queued_spin_lock_slowpath(rqspinlock_t *lock, u32 val)
 	 */
 	if (val & _Q_LOCKED_MASK) {
 		RES_RESET_TIMEOUT(ts, RES_DEF_TIMEOUT);
-		res_smp_cond_load_acquire(&lock->locked, !VAL || RES_CHECK_TIMEOUT(ts, ret, _Q_LOCKED_MASK));
+		smp_cond_load_acquire_timeout(&lock->locked, !VAL,
+					      (ret = check_timeout(lock, _Q_LOCKED_MASK, &ts)));
 	}
 
 	if (ret) {
@@ -569,9 +568,8 @@ int __lockfunc resilient_queued_spin_lock_slowpath(rqspinlock_t *lock, u32 val)
 	 * us.
 	 */
 	RES_RESET_TIMEOUT(ts, RES_DEF_TIMEOUT * 2);
-	val = res_atomic_cond_read_acquire(&lock->val, !(VAL & _Q_LOCKED_PENDING_MASK) ||
-					   RES_CHECK_TIMEOUT(ts, ret, _Q_LOCKED_PENDING_MASK));
-
+	val = atomic_cond_read_acquire_timeout(&lock->val, !(VAL & _Q_LOCKED_PENDING_MASK),
+					       (ret = check_timeout(lock, _Q_LOCKED_PENDING_MASK, &ts)));
 waitq_timeout:
 	if (ret) {
 		/*