5 files changed, 64 insertions, 35 deletions

diff --git a/arch/arm/kernel/ecard.c b/arch/arm/kernel/ecard.c
index 4dd0edab6a65..1651d4950744 100644
--- a/arch/arm/kernel/ecard.c
+++ b/arch/arm/kernel/ecard.c
@@ -242,6 +242,7 @@ static void ecard_init_pgtables(struct mm_struct *mm)
 
         memcpy(dst_pgd, src_pgd, sizeof(pgd_t) * (EASI_SIZE / PGDIR_SIZE));
 
+        vma.vm_flags = VM_EXEC;
         vma.vm_mm = mm;
 
         flush_tlb_range(&vma, IO_START, IO_START + IO_SIZE);
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index 5bb91bf3d47f..b2abfa18f137 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -180,7 +180,7 @@ armpmu_event_set_period(struct perf_event *event,
 u64
 armpmu_event_update(struct perf_event *event,
                     struct hw_perf_event *hwc,
-                    int idx, int overflow)
+                    int idx)
 {
         struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
         u64 delta, prev_raw_count, new_raw_count;
@@ -193,13 +193,7 @@ again:
                              new_raw_count) != prev_raw_count)
                 goto again;
 
-        new_raw_count &= armpmu->max_period;
-        prev_raw_count &= armpmu->max_period;
-
-        if (overflow)
-                delta = armpmu->max_period - prev_raw_count + new_raw_count + 1;
-        else
-                delta = new_raw_count - prev_raw_count;
+        delta = (new_raw_count - prev_raw_count) & armpmu->max_period;
 
         local64_add(delta, &event->count);
         local64_sub(delta, &hwc->period_left);
@@ -216,7 +210,7 @@ armpmu_read(struct perf_event *event)
         if (hwc->idx < 0)
                 return;
 
-        armpmu_event_update(event, hwc, hwc->idx, 0);
+        armpmu_event_update(event, hwc, hwc->idx);
 }
 
 static void
@@ -232,7 +226,7 @@ armpmu_stop(struct perf_event *event, int flags)
         if (!(hwc->state & PERF_HES_STOPPED)) {
                 armpmu->disable(hwc, hwc->idx);
                 barrier(); /* why? */
-                armpmu_event_update(event, hwc, hwc->idx, 0);
+                armpmu_event_update(event, hwc, hwc->idx);
                 hwc->state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
         }
 }
@@ -518,7 +512,13 @@ __hw_perf_event_init(struct perf_event *event)
         hwc->config_base            |= (unsigned long)mapping;
 
         if (!hwc->sample_period) {
-                hwc->sample_period  = armpmu->max_period;
+                /*
+                 * For non-sampling runs, limit the sample_period to half
+                 * of the counter width. That way, the new counter value
+                 * is far less likely to overtake the previous one unless
+                 * you have some serious IRQ latency issues.
+                 */
+                hwc->sample_period  = armpmu->max_period >> 1;
                 hwc->last_period    = hwc->sample_period;
                 local64_set(&hwc->period_left, hwc->sample_period);
         }
@@ -680,6 +680,28 @@ static void __init cpu_pmu_init(struct arm_pmu *armpmu)
 }
 
 /*
+ * PMU hardware loses all context when a CPU goes offline.
+ * When a CPU is hotplugged back in, since some hardware registers are
+ * UNKNOWN at reset, the PMU must be explicitly reset to avoid reading
+ * junk values out of them.
+ */
+static int __cpuinit pmu_cpu_notify(struct notifier_block *b,
+                                        unsigned long action, void *hcpu)
+{
+        if ((action & ~CPU_TASKS_FROZEN) != CPU_STARTING)
+                return NOTIFY_DONE;
+
+        if (cpu_pmu && cpu_pmu->reset)
+                cpu_pmu->reset(NULL);
+
+        return NOTIFY_OK;
+}
+
+static struct notifier_block __cpuinitdata pmu_cpu_notifier = {
+        .notifier_call = pmu_cpu_notify,
+};
+
+/*
  * CPU PMU identification and registration.
  */
 static int __init
@@ -730,6 +752,7 @@ init_hw_perf_events(void)
                 pr_info("enabled with %s PMU driver, %d counters available\n",
                         cpu_pmu->name, cpu_pmu->num_events);
                 cpu_pmu_init(cpu_pmu);
+                register_cpu_notifier(&pmu_cpu_notifier);
                 armpmu_register(cpu_pmu, "cpu", PERF_TYPE_RAW);
         } else {
                 pr_info("no hardware support available\n");
diff --git a/arch/arm/kernel/perf_event_v6.c b/arch/arm/kernel/perf_event_v6.c
index 533be9930ec2..b78af0cc6ef3 100644
--- a/arch/arm/kernel/perf_event_v6.c
+++ b/arch/arm/kernel/perf_event_v6.c
@@ -467,23 +467,6 @@ armv6pmu_enable_event(struct hw_perf_event *hwc,
         raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
-static int counter_is_active(unsigned long pmcr, int idx)
-{
-        unsigned long mask = 0;
-        if (idx == ARMV6_CYCLE_COUNTER)
-                mask = ARMV6_PMCR_CCOUNT_IEN;
-        else if (idx == ARMV6_COUNTER0)
-                mask = ARMV6_PMCR_COUNT0_IEN;
-        else if (idx == ARMV6_COUNTER1)
-                mask = ARMV6_PMCR_COUNT1_IEN;
-
-        if (mask)
-                return pmcr & mask;
-
-        WARN_ONCE(1, "invalid counter number (%d)\n", idx);
-        return 0;
-}
-
 static irqreturn_t
 armv6pmu_handle_irq(int irq_num,
                     void *dev)
@@ -513,7 +496,8 @@ armv6pmu_handle_irq(int irq_num,
                 struct perf_event *event = cpuc->events[idx];
                 struct hw_perf_event *hwc;
 
-                if (!counter_is_active(pmcr, idx))
+                /* Ignore if we don't have an event. */
+                if (!event)
                         continue;
 
                 /*
@@ -524,7 +508,7 @@ armv6pmu_handle_irq(int irq_num,
                         continue;
 
                 hwc = &event->hw;
-                armpmu_event_update(event, hwc, idx, 1);
+                armpmu_event_update(event, hwc, idx);
                 data.period = event->hw.last_period;
                 if (!armpmu_event_set_period(event, hwc, idx))
                         continue;
diff --git a/arch/arm/kernel/perf_event_v7.c b/arch/arm/kernel/perf_event_v7.c
index 6933244c68f9..4d7095af2ab3 100644
--- a/arch/arm/kernel/perf_event_v7.c
+++ b/arch/arm/kernel/perf_event_v7.c
@@ -809,6 +809,11 @@ static inline int armv7_pmnc_disable_intens(int idx)
 
         counter = ARMV7_IDX_TO_COUNTER(idx);
         asm volatile("mcr p15, 0, %0, c9, c14, 2" : : "r" (BIT(counter)));
+        isb();
+        /* Clear the overflow flag in case an interrupt is pending. */
+        asm volatile("mcr p15, 0, %0, c9, c12, 3" : : "r" (BIT(counter)));
+        isb();
+
         return idx;
 }
 
@@ -955,6 +960,10 @@ static irqreturn_t armv7pmu_handle_irq(int irq_num, void *dev)
                 struct perf_event *event = cpuc->events[idx];
                 struct hw_perf_event *hwc;
 
+                /* Ignore if we don't have an event. */
+                if (!event)
+                        continue;
+
                 /*
                  * We have a single interrupt for all counters. Check that
                  * each counter has overflowed before we process it.
@@ -963,7 +972,7 @@ static irqreturn_t armv7pmu_handle_irq(int irq_num, void *dev)
                         continue;
 
                 hwc = &event->hw;
-                armpmu_event_update(event, hwc, idx, 1);
+                armpmu_event_update(event, hwc, idx);
                 data.period = event->hw.last_period;
                 if (!armpmu_event_set_period(event, hwc, idx))
                         continue;
diff --git a/arch/arm/kernel/perf_event_xscale.c b/arch/arm/kernel/perf_event_xscale.c
index 3b99d8269829..71a21e6712f5 100644
--- a/arch/arm/kernel/perf_event_xscale.c
+++ b/arch/arm/kernel/perf_event_xscale.c
@@ -255,11 +255,14 @@ xscale1pmu_handle_irq(int irq_num, void *dev)
                 struct perf_event *event = cpuc->events[idx];
                 struct hw_perf_event *hwc;
 
+                if (!event)
+                        continue;
+
                 if (!xscale1_pmnc_counter_has_overflowed(pmnc, idx))
                         continue;
 
                 hwc = &event->hw;
-                armpmu_event_update(event, hwc, idx, 1);
+                armpmu_event_update(event, hwc, idx);
                 data.period = event->hw.last_period;
                 if (!armpmu_event_set_period(event, hwc, idx))
                         continue;
@@ -592,11 +595,14 @@ xscale2pmu_handle_irq(int irq_num, void *dev)
                 struct perf_event *event = cpuc->events[idx];
                 struct hw_perf_event *hwc;
 
-                if (!xscale2_pmnc_counter_has_overflowed(pmnc, idx))
+                if (!event)
+                        continue;
+
+                if (!xscale2_pmnc_counter_has_overflowed(of_flags, idx))
                         continue;
 
                 hwc = &event->hw;
-                armpmu_event_update(event, hwc, idx, 1);
+                armpmu_event_update(event, hwc, idx);
                 data.period = event->hw.last_period;
                 if (!armpmu_event_set_period(event, hwc, idx))
                         continue;
@@ -663,7 +669,7 @@ xscale2pmu_enable_event(struct hw_perf_event *hwc, int idx)
 static void
 xscale2pmu_disable_event(struct hw_perf_event *hwc, int idx)
 {
-        unsigned long flags, ien, evtsel;
+        unsigned long flags, ien, evtsel, of_flags;
         struct pmu_hw_events *events = cpu_pmu->get_hw_events();
 
         ien = xscale2pmu_read_int_enable();
@@ -672,26 +678,31 @@ xscale2pmu_disable_event(struct hw_perf_event *hwc, int idx)
         switch (idx) {
         case XSCALE_CYCLE_COUNTER:
                 ien &= ~XSCALE2_CCOUNT_INT_EN;
+                of_flags = XSCALE2_CCOUNT_OVERFLOW;
                 break;
         case XSCALE_COUNTER0:
                 ien &= ~XSCALE2_COUNT0_INT_EN;
                 evtsel &= ~XSCALE2_COUNT0_EVT_MASK;
                 evtsel |= XSCALE_PERFCTR_UNUSED << XSCALE2_COUNT0_EVT_SHFT;
+                of_flags = XSCALE2_COUNT0_OVERFLOW;
                 break;
         case XSCALE_COUNTER1:
                 ien &= ~XSCALE2_COUNT1_INT_EN;
                 evtsel &= ~XSCALE2_COUNT1_EVT_MASK;
                 evtsel |= XSCALE_PERFCTR_UNUSED << XSCALE2_COUNT1_EVT_SHFT;
+                of_flags = XSCALE2_COUNT1_OVERFLOW;
                 break;
         case XSCALE_COUNTER2:
                 ien &= ~XSCALE2_COUNT2_INT_EN;
                 evtsel &= ~XSCALE2_COUNT2_EVT_MASK;
                 evtsel |= XSCALE_PERFCTR_UNUSED << XSCALE2_COUNT2_EVT_SHFT;
+                of_flags = XSCALE2_COUNT2_OVERFLOW;
                 break;
         case XSCALE_COUNTER3:
                 ien &= ~XSCALE2_COUNT3_INT_EN;
                 evtsel &= ~XSCALE2_COUNT3_EVT_MASK;
                 evtsel |= XSCALE_PERFCTR_UNUSED << XSCALE2_COUNT3_EVT_SHFT;
+                of_flags = XSCALE2_COUNT3_OVERFLOW;
                 break;
         default:
                 WARN_ONCE(1, "invalid counter number (%d)\n", idx);
@@ -701,6 +712,7 @@ xscale2pmu_disable_event(struct hw_perf_event *hwc, int idx)
         raw_spin_lock_irqsave(&events->pmu_lock, flags);
         xscale2pmu_write_event_select(evtsel);
         xscale2pmu_write_int_enable(ien);
+        xscale2pmu_write_overflow_flags(of_flags);
         raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }