Initial checkin for rmlock (read mostly lock) a multi reader single writer

lock optimized for almost exclusive reader access. (see also rmlock.9)

TODO:
    Convert to per cpu variables linkerset as soon as it is available.
    Optimize UP (single processor)  case.

1 files changed, 37 insertions, 16 deletions

diff --git a/sys/kern/subr_smp.c b/sys/kern/subr_smp.c
index a884288..bc30b2a 100644
--- a/sys/kern/subr_smp.c
+++ b/sys/kern/subr_smp.c
@@ -104,10 +104,10 @@ SYSCTL_INT(_kern_smp, OID_AUTO, forward_roundrobin_enabled, CTLFLAG_RW,
 	   "Forwarding of roundrobin to all other CPUs");
 
 /* Variables needed for SMP rendezvous. */
-static void (*smp_rv_setup_func)(void *arg);
-static void (*smp_rv_action_func)(void *arg);
-static void (*smp_rv_teardown_func)(void *arg);
-static void *smp_rv_func_arg;
+static void (*volatile smp_rv_setup_func)(void *arg);
+static void (*volatile smp_rv_action_func)(void *arg);
+static void (* volatile smp_rv_teardown_func)(void *arg);
+static void * volatile smp_rv_func_arg;
 static volatile int smp_rv_waiters[3];
 
 /* 
@@ -286,6 +286,13 @@ restart_cpus(cpumask_t map)
 	return 1;
 }
 
+void smp_no_rendevous_barrier(void *dummy)
+{
+#ifdef SMP
+	KASSERT((!smp_started),("smp_no_rendevous called and smp is started"));
+#endif
+}
+
 /*
  * All-CPU rendezvous.  CPUs are signalled, all execute the setup function 
  * (if specified), rendezvous, execute the action function (if specified),
@@ -298,33 +305,41 @@ restart_cpus(cpumask_t map)
 void
 smp_rendezvous_action(void)
 {
-
+	void* local_func_arg = smp_rv_func_arg;
+	void (*local_setup_func)(void*)   = smp_rv_setup_func;
+	void (*local_action_func)(void*)   = smp_rv_action_func;
+	void (*local_teardown_func)(void*) = smp_rv_teardown_func;
+ 
 	/* Ensure we have up-to-date values. */
 	atomic_add_acq_int(&smp_rv_waiters[0], 1);
 	while (smp_rv_waiters[0] < mp_ncpus)
 		cpu_spinwait();
 
 	/* setup function */
-	if (smp_rv_setup_func != NULL)
-		smp_rv_setup_func(smp_rv_func_arg);
+	if (local_setup_func != smp_no_rendevous_barrier) {
+		if (smp_rv_setup_func != NULL)
+			smp_rv_setup_func(smp_rv_func_arg);
+		/* spin on entry rendezvous */
+		atomic_add_int(&smp_rv_waiters[1], 1);
+		while (smp_rv_waiters[1] < mp_ncpus)
+                	cpu_spinwait();
+	}
 
-	/* spin on entry rendezvous */
-	atomic_add_int(&smp_rv_waiters[1], 1);
-	while (smp_rv_waiters[1] < mp_ncpus)
-		cpu_spinwait();
 
 	/* action function */
-	if (smp_rv_action_func != NULL)
-		smp_rv_action_func(smp_rv_func_arg);
+	if (local_action_func != NULL)
+		local_action_func(local_func_arg);
+
 
 	/* spin on exit rendezvous */
 	atomic_add_int(&smp_rv_waiters[2], 1);
+	if (local_teardown_func == smp_no_rendevous_barrier)
+                return;
 	while (smp_rv_waiters[2] < mp_ncpus)
 		cpu_spinwait();
-
 	/* teardown function */
-	if (smp_rv_teardown_func != NULL)
-		smp_rv_teardown_func(smp_rv_func_arg);
+	if (local_teardown_func != NULL)
+		local_teardown_func(local_func_arg);
 }
 
 void
@@ -356,12 +371,18 @@ smp_rendezvous(void (* setup_func)(void *),
 	smp_rv_waiters[2] = 0;
 	atomic_store_rel_int(&smp_rv_waiters[0], 0);
 
+		
+
 	/* signal other processors, which will enter the IPI with interrupts off */
 	ipi_all_but_self(IPI_RENDEZVOUS);
 
 	/* call executor function */
 	smp_rendezvous_action();
 
+	if (teardown_func == smp_no_rendevous_barrier) { 
+		while (atomic_load_acq_int(&smp_rv_waiters[2]) < mp_ncpus)
+			cpu_spinwait();
+	}
 	/* release lock */
 	mtx_unlock_spin(&smp_ipi_mtx);
 }