[PATCH] CPU C-state breakage with PM Qos change

* [PATCH] CPU C-state breakage with PM Qos change
@ 2012-02-02 23:57 Venkatesh Pallipadi
  2012-02-03  0:14 ` Venkatesh Pallipadi
  0 siblings, 1 reply; 10+ messages in thread
From: Venkatesh Pallipadi @ 2012-02-02 23:57 UTC (permalink / raw)
  To: Rafael J. Wysocki
  Cc: Kevin Hilman, Jean Pihet, markgross, linux-kernel, Len Brown,
	Venkatesh Pallipadi

Looks like change "PM QoS: Move and rename the implementation files"
made pm_qos depend on CONFIG_PM which depends on
PM_SLEEP || PM_RUNTIME

That breaks CPU C-states with kernels not having these CONFIGs, causing CPUs
to spend time in Polling loop idle instead of going into deep C-states,
consuming way way more power. This is with either acpi idle or intel idle
enabled.

Either CONFIG_PM should be enabled with any pm_qos users or
the !CONFIG_PM pm_qos_request() should return sane defaults not to break
the existing users. Here's is the patch for the latter option.

Signed-off-by: Venkatesh Pallipadi <venki@google.com>
---
 include/linux/pm_qos.h |   14 +++++++++++++-
 1 files changed, 13 insertions(+), 1 deletions(-)

diff --git a/include/linux/pm_qos.h b/include/linux/pm_qos.h
index 83b0ea3..84c99f4 100644
--- a/include/linux/pm_qos.h
+++ b/include/linux/pm_qos.h
@@ -107,7 +107,19 @@ static inline void pm_qos_remove_request(struct pm_qos_request *req)
 			{ return; }
 
 static inline int pm_qos_request(int pm_qos_class)
-			{ return 0; }
+{
+	switch (pm_qos_class) {
+	case PM_QOS_CPU_DMA_LATENCY:
+		return PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE;
+	case PM_QOS_NETWORK_LATENCY:
+		return PM_QOS_NETWORK_LAT_DEFAULT_VALUE;
+	case PM_QOS_NETWORK_LATENCY:
+		return PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE;
+	default:
+		return PM_QOS_DEFAULT_VALUE;
+	}
+}
+
 static inline int pm_qos_add_notifier(int pm_qos_class,
 				      struct notifier_block *notifier)
 			{ return 0; }
-- 
1.7.7.3


^ permalink raw reply related	[flat|nested] 10+ messages in thread