2.4-ac task->cpu abstraction and optimization

diff -urN linux-2.4.20-pre5-ac4-rml/fs/proc/array.c linux/fs/proc/array.c
@@ -389,7 +389,7 @@
 		task->nswap,
 		task->cnswap,
 		task->exit_signal,
-		task->cpu);
+		task_cpu(task));
 	if(mm)
 		mmput(mm);
 	return res;

diff -urN linux-2.4.20-pre5-ac4-rml/include/linux/sched.h linux/include/linux/sched.h
@@ -976,6 +976,34 @@
 	return unlikely(current->need_resched);
 }
 
+/*
+ * Wrappers for p->cpu access. No-op on UP.
+ */
+#ifdef CONFIG_SMP
+
+static inline unsigned int task_cpu(struct task_struct *p)
+{
+	return p->cpu;
+}
+
+static inline void set_task_cpu(struct task_struct *p, unsigned int cpu)
+{
+	p->cpu = cpu;
+}
+
+#else
+
+static inline unsigned int task_cpu(struct task_struct *p)
+{
+	return 0;
+}
+
+static inline void set_task_cpu(struct task_struct *p, unsigned int cpu)
+{
+}
+
+#endif /* CONFIG_SMP */
+
 #endif /* __KERNEL__ */
 
 #endif

diff -urN linux-2.4.20-pre5-ac4-rml/kernel/sched.c linux/kernel/sched.c
@@ -148,7 +148,7 @@
 
 #define cpu_rq(cpu)		(runqueues + (cpu))
 #define this_rq()		cpu_rq(smp_processor_id())
-#define task_rq(p)		cpu_rq((p)->cpu)
+#define task_rq(p)		cpu_rq(task_cpu(p))
 #define cpu_curr(cpu)		(cpu_rq(cpu)->curr)
 #define rt_task(p)		((p)->prio < MAX_RT_PRIO)
 

@@ -311,8 +311,8 @@
 	need_resched = p->need_resched;
 	wmb();
 	set_tsk_need_resched(p);
-	if (!need_resched && (p->cpu != smp_processor_id()))
-		smp_send_reschedule(p->cpu);
+	if (!need_resched && (task_cpu(p) != smp_processor_id()))
+		smp_send_reschedule(task_cpu(p));
 #else
 	set_tsk_need_resched(p);
 #endif

@@ -391,10 +391,10 @@
 		 * currently. Do not violate hard affinity.
 		 */
 		if (unlikely(sync && (rq->curr != p) &&
-			(p->cpu != smp_processor_id()) &&
+			(task_cpu(p) != smp_processor_id()) &&
 			(p->cpus_allowed & (1UL << smp_processor_id())))) {
 
-			p->cpu = smp_processor_id();
+			set_task_cpu(p, smp_processor_id());
 			task_rq_unlock(rq, &flags);
 			goto repeat_lock_task;
 		}

@@ -437,7 +437,7 @@
 		p->sleep_avg = p->sleep_avg * CHILD_PENALTY / 100;
 		p->prio = effective_prio(p);
 	}
-	p->cpu = smp_processor_id();
+	set_task_cpu(p, smp_processor_id());
 	activate_task(p, rq);
 
 	rq_unlock(rq);

@@ -727,7 +727,7 @@
 	 */
 	dequeue_task(next, array);
 	busiest->nr_running--;
-	next->cpu = this_cpu;
+	set_task_cpu(next, this_cpu);
 	this_rq->nr_running++;
 	enqueue_task(next, this_rq->active);
 	if (next->prio < current->prio)

@@ -1718,7 +1718,7 @@
 
 void __init init_idle(task_t *idle, int cpu)
 {
-	runqueue_t *idle_rq = cpu_rq(cpu), *rq = cpu_rq(idle->cpu);
+	runqueue_t *idle_rq = cpu_rq(cpu), *rq = cpu_rq(task_cpu(idle));
 	unsigned long flags;
 
 	__save_flags(flags);

@@ -1730,7 +1730,7 @@
 	idle->array = NULL;
 	idle->prio = MAX_PRIO;
 	idle->state = TASK_RUNNING;
-	idle->cpu = cpu;
+	set_task_cpu(idle, cpu);
 	double_rq_unlock(idle_rq, rq);
 	set_tsk_need_resched(idle);
 	__restore_flags(flags);

@@ -1835,7 +1835,7 @@
 	 * Can the task run on the task's current CPU? If not then
 	 * migrate the process off to a proper CPU.
 	 */
-	if (new_mask & (1UL << p->cpu)) {
+	if (new_mask & (1UL << task_cpu(p))) {
 		task_rq_unlock(rq, &flags);
 		goto out;
 	}

@@ -1844,7 +1844,7 @@
 	 * it is sufficient to simply update the task's cpu field.
 	 */
 	if (!p->array && (p != rq->curr)) {
-		p->cpu = __ffs(p->cpus_allowed);
+		set_task_cpu(p, __ffs(p->cpus_allowed));
 		task_rq_unlock(rq, &flags);
 		goto out;
 	}

@@ -1914,18 +1914,18 @@
 		cpu_dest = __ffs(p->cpus_allowed);
 		rq_dest = cpu_rq(cpu_dest);
 repeat:
-		cpu_src = p->cpu;
+		cpu_src = task_cpu(p);
 		rq_src = cpu_rq(cpu_src);
 
 		local_irq_save(flags);
 		double_rq_lock(rq_src, rq_dest);
-		if (p->cpu != cpu_src) {
+		if (task_cpu(p) != cpu_src) {
 			double_rq_unlock(rq_src, rq_dest);
 			local_irq_restore(flags);
 			goto repeat;
 		}
 		if (rq_src == rq) {
-			p->cpu = cpu_dest;
+			set_task_cpu(p, cpu_dest);
 			if (p->array) {
 				deactivate_task(p, rq_src);
 				activate_task(p, rq_dest);