hrtimer: fix rq->lock inversion (again)

It appears I inadvertly introduced rq->lock recursion to the
hrtimer_start() path when I delegated running already expired
timers to softirq context.

This patch fixes it by introducing a __hrtimer_start_range_ns()
method that will not use raise_softirq_irqoff() but
__raise_softirq_irqoff() which avoids the wakeup.

It then also changes schedule() to check for pending softirqs and
do the wakeup then, I'm not quite sure I like this last bit, nor
am I convinced its really needed.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus@samba.org
LKML-Reference: <20090313112301.096138802@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index bd37078c2d7d7ceb9f9f7b9bd2340724ae29c73d..0d2f7c8a33d6aa2983fd610bac1fb7b4ebc44cca 100644 (file)
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -336,6 +336,11 @@ extern int hrtimer_start(struct hrtimer *timer, ktime_t tim,
                         const enum hrtimer_mode mode);
 extern int hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim,
                        unsigned long range_ns, const enum hrtimer_mode mode);
+extern int
+__hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim,
+                        unsigned long delta_ns,
+                        const enum hrtimer_mode mode, int wakeup);
+
 extern int hrtimer_cancel(struct hrtimer *timer);
 extern int hrtimer_try_to_cancel(struct hrtimer *timer);
 

diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index c68bffd182bbe4d977848fd3507150da83be4d7b..4528bf70866a24d320dad035c63d88e16c25d95b 100644 (file)
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -294,6 +294,7 @@ extern void softirq_init(void);
 #define __raise_softirq_irqoff(nr) do { or_softirq_pending(1UL << (nr)); } while (0)
 extern void raise_softirq_irqoff(unsigned int nr);
 extern void raise_softirq(unsigned int nr);
+extern void wakeup_softirqd(void);
 
 /* This is the worklist that queues up per-cpu softirq work.
  *

diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index f394d2a42ca3442a048d6c9586ec8bfd6cfe4487..cb8a15c1958318c52c116d08e33003ad2ea0cc98 100644 (file)
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -651,14 +651,20 @@ static inline void hrtimer_init_timer_hres(struct hrtimer *timer)
  * and expiry check is done in the hrtimer_interrupt or in the softirq.
  */
 static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
-                                           struct hrtimer_clock_base *base)
+                                           struct hrtimer_clock_base *base,
+                                           int wakeup)
 {
        if (base->cpu_base->hres_active && hrtimer_reprogram(timer, base)) {
-               spin_unlock(&base->cpu_base->lock);
-               raise_softirq_irqoff(HRTIMER_SOFTIRQ);
-               spin_lock(&base->cpu_base->lock);
+               if (wakeup) {
+                       spin_unlock(&base->cpu_base->lock);
+                       raise_softirq_irqoff(HRTIMER_SOFTIRQ);
+                       spin_lock(&base->cpu_base->lock);
+               } else
+                       __raise_softirq_irqoff(HRTIMER_SOFTIRQ);
+
                return 1;
        }
+
        return 0;
 }
 
@@ -703,7 +709,8 @@ static inline int hrtimer_is_hres_enabled(void) { return 0; }
 static inline int hrtimer_switch_to_hres(void) { return 0; }
 static inline void hrtimer_force_reprogram(struct hrtimer_cpu_base *base) { }
 static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
-                                           struct hrtimer_clock_base *base)
+                                           struct hrtimer_clock_base *base,
+                                           int wakeup)
 {
        return 0;
 }
@@ -886,20 +893,9 @@ remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base)
        return 0;
 }
 
-/**
- * hrtimer_start_range_ns - (re)start an hrtimer on the current CPU
- * @timer:     the timer to be added
- * @tim:       expiry time
- * @delta_ns:  "slack" range for the timer
- * @mode:      expiry mode: absolute (HRTIMER_ABS) or relative (HRTIMER_REL)
- *
- * Returns:
- *  0 on success
- *  1 when the timer was active
- */
-int
-hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_ns,
-                       const enum hrtimer_mode mode)
+int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim,
+               unsigned long delta_ns, const enum hrtimer_mode mode,
+               int wakeup)
 {
        struct hrtimer_clock_base *base, *new_base;
        unsigned long flags;
@@ -940,12 +936,29 @@ hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_n
         * XXX send_remote_softirq() ?
         */
        if (leftmost && new_base->cpu_base == &__get_cpu_var(hrtimer_bases))
-               hrtimer_enqueue_reprogram(timer, new_base);
+               hrtimer_enqueue_reprogram(timer, new_base, wakeup);
 
        unlock_hrtimer_base(timer, &flags);
 
        return ret;
 }
+
+/**
+ * hrtimer_start_range_ns - (re)start an hrtimer on the current CPU
+ * @timer:     the timer to be added
+ * @tim:       expiry time
+ * @delta_ns:  "slack" range for the timer
+ * @mode:      expiry mode: absolute (HRTIMER_ABS) or relative (HRTIMER_REL)
+ *
+ * Returns:
+ *  0 on success
+ *  1 when the timer was active
+ */
+int hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim,
+               unsigned long delta_ns, const enum hrtimer_mode mode)
+{
+       return __hrtimer_start_range_ns(timer, tim, delta_ns, mode, 1);
+}
 EXPORT_SYMBOL_GPL(hrtimer_start_range_ns);
 
 /**
@@ -961,7 +974,7 @@ EXPORT_SYMBOL_GPL(hrtimer_start_range_ns);
 int
 hrtimer_start(struct hrtimer *timer, ktime_t tim, const enum hrtimer_mode mode)
 {
-       return hrtimer_start_range_ns(timer, tim, 0, mode);
+       return __hrtimer_start_range_ns(timer, tim, 0, mode, 1);
 }
 EXPORT_SYMBOL_GPL(hrtimer_start);
 

diff --git a/kernel/sched.c b/kernel/sched.c
index 196d48babbef87c088214e4f252b4397d2bf25be..63256e3ede2a3480cd9fe3b27122dea44171cb63 100644 (file)
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -231,13 +231,20 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b)
 
        spin_lock(&rt_b->rt_runtime_lock);
        for (;;) {
+               unsigned long delta;
+               ktime_t soft, hard;
+
                if (hrtimer_active(&rt_b->rt_period_timer))
                        break;
 
                now = hrtimer_cb_get_time(&rt_b->rt_period_timer);
                hrtimer_forward(&rt_b->rt_period_timer, now, rt_b->rt_period);
-               hrtimer_start_expires(&rt_b->rt_period_timer,
-                               HRTIMER_MODE_ABS);
+
+               soft = hrtimer_get_softexpires(&rt_b->rt_period_timer);
+               hard = hrtimer_get_expires(&rt_b->rt_period_timer);
+               delta = ktime_to_ns(ktime_sub(hard, soft));
+               __hrtimer_start_range_ns(&rt_b->rt_period_timer, soft, delta,
+                               HRTIMER_MODE_ABS, 0);
        }
        spin_unlock(&rt_b->rt_runtime_lock);
 }
@@ -1146,7 +1153,8 @@ static __init void init_hrtick(void)
  */
 static void hrtick_start(struct rq *rq, u64 delay)
 {
-       hrtimer_start(&rq->hrtick_timer, ns_to_ktime(delay), HRTIMER_MODE_REL);
+       __hrtimer_start_range_ns(&rq->hrtick_timer, ns_to_ktime(delay), 0,
+                       HRTIMER_MODE_REL, 0);
 }
 
 static inline void init_hrtick(void)

diff --git a/kernel/softirq.c b/kernel/softirq.c
index 4877516043000a42278eb13b2bef99e21696671a..accc85197c496940b587fb4b7037676f98f6fa6a 100644 (file)
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -58,7 +58,7 @@ static DEFINE_PER_CPU(struct task_struct *, ksoftirqd);
  * to the pending events, so lets the scheduler to balance
  * the softirq load for us.
  */
-static inline void wakeup_softirqd(void)
+void wakeup_softirqd(void)
 {
        /* Interrupts are disabled: no need to stop preemption */
        struct task_struct *tsk = __get_cpu_var(ksoftirqd);