[PATCH 2/2] cpuidle: coupled: fix decrementing ready count

From: Colin Cross
Date: Fri May 18 2012 - 14:12:24 EST


cpuidle_coupled_set_not_ready sometimes refuses to decrement the
ready count in order to prevent a race condition. This makes it
unsuitable for use when finished with idle. Add a new function
cpuidle_coupled_set_done that decrements both the ready count and
waiting count, and call it after idle is complete.

Signed-off-by: Colin Cross <ccross@xxxxxxxxxxx>
---
drivers/cpuidle/coupled.c | 18 ++++++++++++++++--
1 files changed, 16 insertions(+), 2 deletions(-)

diff --git a/drivers/cpuidle/coupled.c b/drivers/cpuidle/coupled.c
index b02810a..fc427fa 100644
--- a/drivers/cpuidle/coupled.c
+++ b/drivers/cpuidle/coupled.c
@@ -388,6 +388,21 @@ static void cpuidle_coupled_set_not_waiting(int cpu,
}

/**
+ * cpuidle_coupled_set_done - mark this cpu as leaving the ready loop
+ * @cpu: the current cpu
+ * @coupled: the struct coupled that contains the current cpu
+ *
+ * Marks this cpu as no longer in the ready and waiting loops. Decrements
+ * the waiting count first to prevent another cpu looping back in and seeing
+ * this cpu as waiting just before it exits idle.
+ */
+static void cpuidle_coupled_set_done(int cpu, struct cpuidle_coupled *coupled)
+{
+ cpuidle_coupled_set_not_waiting(cpu, coupled);
+ atomic_sub(MAX_WAITING_CPUS, &coupled->ready_waiting_counts);
+}
+
+/**
* cpuidle_coupled_clear_pokes - spin until the poke interrupt is processed
* @cpu - this cpu
*
@@ -501,8 +516,7 @@ int cpuidle_enter_state_coupled(struct cpuidle_device *dev,

entered_state = cpuidle_enter_state(dev, drv, next_state);

- cpuidle_coupled_set_not_waiting(dev->cpu, coupled);
- cpuidle_coupled_set_not_ready(coupled);
+ cpuidle_coupled_set_done(dev->cpu, coupled);

out:
/*
--
1.7.7.3

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/