All of lore.kernel.org
 help / color / mirror / Atom feed
* idle patches for 2.6.36.merge
@ 2010-08-15  5:22 Len Brown
  2010-08-15  5:22   ` Len Brown
  0 siblings, 1 reply; 11+ messages in thread
From: Len Brown @ 2010-08-15  5:22 UTC (permalink / raw)
  To: linux-pm; +Cc: linux-kernel

The following patches are queued for 2.6.36.merge
Please let me know if you see problems with any of them.

thanks,
Len Brown, Intel Open Source Technology Center

^ permalink raw reply	[flat|nested] 11+ messages in thread

* [PATCH 1/5] intel_idle: delete power_policy modparam, and choose substate functions
  2010-08-15  5:22 idle patches for 2.6.36.merge Len Brown
@ 2010-08-15  5:22   ` Len Brown
  0 siblings, 0 replies; 11+ messages in thread
From: Len Brown @ 2010-08-15  5:22 UTC (permalink / raw)
  To: linux-pm; +Cc: linux-kernel, Len Brown

From: Len Brown <len.brown@intel.com>

The idea behind power policy was that it would start off as a modparam,
and then hook into the new "global" in-kernel power vs energy tunable.
But that tunable isn't happening, so delete the hook here.

With the policy hook gone, the sub-state choice functions
do not do anything useful, so delete them from the critical path.

To handle sub-states in the future, we will advertise them
with dedicated cpuidle_state entries.  That is necessary
because some of the sub-states will have substantially different
properties than their peer sub-states.

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/intel_idle.c |   43 -------------------------------------------
 1 files changed, 0 insertions(+), 43 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 82c9a58..6d43bc3 100755
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -77,10 +77,8 @@ static struct cpuidle_driver intel_idle_driver = {
 };
 /* intel_idle.max_cstate=0 disables driver */
 static int max_cstate = MWAIT_MAX_NUM_CSTATES - 1;
-static int power_policy = 7; /* 0 = max perf; 15 = max powersave */
 
 static unsigned int mwait_substates;
-static int (*choose_substate)(int);
 
 /* Reliable LAPIC Timer States, bit 1 for C1 etc.  */
 static unsigned int lapic_timer_reliable_states;
@@ -168,42 +166,6 @@ static struct cpuidle_state atom_cstates[MWAIT_MAX_NUM_CSTATES] = {
 		.enter = NULL },	/* disabled */
 };
 
-/*
- * choose_tunable_substate()
- *
- * Run-time decision on which C-state substate to invoke
- * If power_policy = 0, choose shallowest substate (0)
- * If power_policy = 15, choose deepest substate
- * If power_policy = middle, choose middle substate etc.
- */
-static int choose_tunable_substate(int cstate)
-{
-	unsigned int num_substates;
-	unsigned int substate_choice;
-
-	power_policy &= 0xF;	/* valid range: 0-15 */
-	cstate &= 7;	/* valid range: 0-7 */
-
-	num_substates = (mwait_substates >> ((cstate) * 4))
-				& MWAIT_SUBSTATE_MASK;
-
-	if (num_substates <= 1)
-		return 0;
-
-	substate_choice = ((power_policy + (power_policy + 1) *
-				(num_substates - 1)) / 16);
-
-	return substate_choice;
-}
-
-/*
- * choose_zero_substate()
- */
-static int choose_zero_substate(int cstate)
-{
-	return 0;
-}
-
 /**
  * intel_idle
  * @dev: cpuidle_device
@@ -221,8 +183,6 @@ static int intel_idle(struct cpuidle_device *dev, struct cpuidle_state *state)
 
 	cstate = (((eax) >> MWAIT_SUBSTATE_SIZE) & MWAIT_CSTATE_MASK) + 1;
 
-	eax = eax + (choose_substate)(cstate);
-
 	local_irq_disable();
 
 	if (!(lapic_timer_reliable_states & (1 << (cstate))))
@@ -301,13 +261,11 @@ static int intel_idle_probe(void)
 	case 0x25:	/* Westmere */
 	case 0x2C:	/* Westmere */
 		cpuidle_state_table = nehalem_cstates;
-		choose_substate = choose_tunable_substate;
 		break;
 
 	case 0x1C:	/* 28 - Atom Processor */
 		lapic_timer_reliable_states = (1 << 2) | (1 << 1); /* C2, C1 */
 		cpuidle_state_table = atom_cstates;
-		choose_substate = choose_zero_substate;
 		break;
 #ifdef FUTURE_USE
 	case 0x17:	/* 23 - Core 2 Duo */
@@ -447,7 +405,6 @@ static void __exit intel_idle_exit(void)
 module_init(intel_idle_init);
 module_exit(intel_idle_exit);
 
-module_param(power_policy, int, 0644);
 module_param(max_cstate, int, 0444);
 
 MODULE_AUTHOR("Len Brown <len.brown@intel.com>");
-- 
1.7.2.1.95.g3d045


^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 1/5] intel_idle: delete power_policy modparam, and choose substate functions
@ 2010-08-15  5:22   ` Len Brown
  0 siblings, 0 replies; 11+ messages in thread
From: Len Brown @ 2010-08-15  5:22 UTC (permalink / raw)
  To: linux-pm; +Cc: Len Brown, linux-kernel

From: Len Brown <len.brown@intel.com>

The idea behind power policy was that it would start off as a modparam,
and then hook into the new "global" in-kernel power vs energy tunable.
But that tunable isn't happening, so delete the hook here.

With the policy hook gone, the sub-state choice functions
do not do anything useful, so delete them from the critical path.

To handle sub-states in the future, we will advertise them
with dedicated cpuidle_state entries.  That is necessary
because some of the sub-states will have substantially different
properties than their peer sub-states.

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/intel_idle.c |   43 -------------------------------------------
 1 files changed, 0 insertions(+), 43 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 82c9a58..6d43bc3 100755
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -77,10 +77,8 @@ static struct cpuidle_driver intel_idle_driver = {
 };
 /* intel_idle.max_cstate=0 disables driver */
 static int max_cstate = MWAIT_MAX_NUM_CSTATES - 1;
-static int power_policy = 7; /* 0 = max perf; 15 = max powersave */
 
 static unsigned int mwait_substates;
-static int (*choose_substate)(int);
 
 /* Reliable LAPIC Timer States, bit 1 for C1 etc.  */
 static unsigned int lapic_timer_reliable_states;
@@ -168,42 +166,6 @@ static struct cpuidle_state atom_cstates[MWAIT_MAX_NUM_CSTATES] = {
 		.enter = NULL },	/* disabled */
 };
 
-/*
- * choose_tunable_substate()
- *
- * Run-time decision on which C-state substate to invoke
- * If power_policy = 0, choose shallowest substate (0)
- * If power_policy = 15, choose deepest substate
- * If power_policy = middle, choose middle substate etc.
- */
-static int choose_tunable_substate(int cstate)
-{
-	unsigned int num_substates;
-	unsigned int substate_choice;
-
-	power_policy &= 0xF;	/* valid range: 0-15 */
-	cstate &= 7;	/* valid range: 0-7 */
-
-	num_substates = (mwait_substates >> ((cstate) * 4))
-				& MWAIT_SUBSTATE_MASK;
-
-	if (num_substates <= 1)
-		return 0;
-
-	substate_choice = ((power_policy + (power_policy + 1) *
-				(num_substates - 1)) / 16);
-
-	return substate_choice;
-}
-
-/*
- * choose_zero_substate()
- */
-static int choose_zero_substate(int cstate)
-{
-	return 0;
-}
-
 /**
  * intel_idle
  * @dev: cpuidle_device
@@ -221,8 +183,6 @@ static int intel_idle(struct cpuidle_device *dev, struct cpuidle_state *state)
 
 	cstate = (((eax) >> MWAIT_SUBSTATE_SIZE) & MWAIT_CSTATE_MASK) + 1;
 
-	eax = eax + (choose_substate)(cstate);
-
 	local_irq_disable();
 
 	if (!(lapic_timer_reliable_states & (1 << (cstate))))
@@ -301,13 +261,11 @@ static int intel_idle_probe(void)
 	case 0x25:	/* Westmere */
 	case 0x2C:	/* Westmere */
 		cpuidle_state_table = nehalem_cstates;
-		choose_substate = choose_tunable_substate;
 		break;
 
 	case 0x1C:	/* 28 - Atom Processor */
 		lapic_timer_reliable_states = (1 << 2) | (1 << 1); /* C2, C1 */
 		cpuidle_state_table = atom_cstates;
-		choose_substate = choose_zero_substate;
 		break;
 #ifdef FUTURE_USE
 	case 0x17:	/* 23 - Core 2 Duo */
@@ -447,7 +405,6 @@ static void __exit intel_idle_exit(void)
 module_init(intel_idle_init);
 module_exit(intel_idle_exit);
 
-module_param(power_policy, int, 0644);
 module_param(max_cstate, int, 0444);
 
 MODULE_AUTHOR("Len Brown <len.brown@intel.com>");
-- 
1.7.2.1.95.g3d045

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 2/5] intel_idle: add support for Westmere-EX
  2010-08-15  5:22   ` Len Brown
@ 2010-08-15  5:22     ` Len Brown
  -1 siblings, 0 replies; 11+ messages in thread
From: Len Brown @ 2010-08-15  5:22 UTC (permalink / raw)
  To: linux-pm; +Cc: linux-kernel, Len Brown

From: Len Brown <len.brown@intel.com>

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/intel_idle.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 6d43bc3..97655ed 100755
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -256,6 +256,7 @@ static int intel_idle_probe(void)
 	case 0x1E:	/* Core i7 and i5 Processor - Lynnfield Jasper Forest */
 	case 0x1F:	/* Core i7 and i5 Processor - Nehalem */
 	case 0x2E:	/* Nehalem-EX Xeon */
+	case 0x2F:	/* Westmere-EX Xeon */
 		lapic_timer_reliable_states = (1 << 1);	 /* C1 */
 
 	case 0x25:	/* Westmere */
-- 
1.7.2.1.95.g3d045


^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 2/5] intel_idle: add support for Westmere-EX
@ 2010-08-15  5:22     ` Len Brown
  0 siblings, 0 replies; 11+ messages in thread
From: Len Brown @ 2010-08-15  5:22 UTC (permalink / raw)
  To: linux-pm; +Cc: Len Brown, linux-kernel

From: Len Brown <len.brown@intel.com>

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/intel_idle.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 6d43bc3..97655ed 100755
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -256,6 +256,7 @@ static int intel_idle_probe(void)
 	case 0x1E:	/* Core i7 and i5 Processor - Lynnfield Jasper Forest */
 	case 0x1F:	/* Core i7 and i5 Processor - Nehalem */
 	case 0x2E:	/* Nehalem-EX Xeon */
+	case 0x2F:	/* Westmere-EX Xeon */
 		lapic_timer_reliable_states = (1 << 1);	 /* C1 */
 
 	case 0x25:	/* Westmere */
-- 
1.7.2.1.95.g3d045

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 3/5] intel_idle: disable module support
  2010-08-15  5:22   ` Len Brown
                     ` (2 preceding siblings ...)
  (?)
@ 2010-08-15  5:22   ` Len Brown
  -1 siblings, 0 replies; 11+ messages in thread
From: Len Brown @ 2010-08-15  5:22 UTC (permalink / raw)
  To: linux-pm; +Cc: linux-kernel, Len Brown

From: Len Brown <len.brown@intel.com>

Right now the module capability is cauing more trouble
than it is worth.  At least one distro built intel_idle as a module
where it lost the init race with ACPI, making it useless.

Make intel_idle bool so that if you select it, you will use it.

We can restore module capability after cpuidle is enhanced
to handle run-time changing of idle drivers.

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/Kconfig |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/idle/Kconfig b/drivers/idle/Kconfig
index fb5c518..82e4e86 100644
--- a/drivers/idle/Kconfig
+++ b/drivers/idle/Kconfig
@@ -1,5 +1,5 @@
 config INTEL_IDLE
-	tristate "Cpuidle Driver for Intel Processors"
+	bool "Cpuidle Driver for Intel Processors"
 	depends on CPU_IDLE
 	depends on X86
 	depends on CPU_SUP_INTEL
-- 
1.7.2.1.95.g3d045


^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 3/5] intel_idle: disable module support
  2010-08-15  5:22   ` Len Brown
  (?)
  (?)
@ 2010-08-15  5:22   ` Len Brown
  -1 siblings, 0 replies; 11+ messages in thread
From: Len Brown @ 2010-08-15  5:22 UTC (permalink / raw)
  To: linux-pm; +Cc: Len Brown, linux-kernel

From: Len Brown <len.brown@intel.com>

Right now the module capability is cauing more trouble
than it is worth.  At least one distro built intel_idle as a module
where it lost the init race with ACPI, making it useless.

Make intel_idle bool so that if you select it, you will use it.

We can restore module capability after cpuidle is enhanced
to handle run-time changing of idle drivers.

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/Kconfig |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/idle/Kconfig b/drivers/idle/Kconfig
index fb5c518..82e4e86 100644
--- a/drivers/idle/Kconfig
+++ b/drivers/idle/Kconfig
@@ -1,5 +1,5 @@
 config INTEL_IDLE
-	tristate "Cpuidle Driver for Intel Processors"
+	bool "Cpuidle Driver for Intel Processors"
 	depends on CPU_IDLE
 	depends on X86
 	depends on CPU_SUP_INTEL
-- 
1.7.2.1.95.g3d045

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 4/5] intel_idle: no longer EXPERIMENTAL
  2010-08-15  5:22   ` Len Brown
                     ` (4 preceding siblings ...)
  (?)
@ 2010-08-15  5:22   ` Len Brown
  -1 siblings, 0 replies; 11+ messages in thread
From: Len Brown @ 2010-08-15  5:22 UTC (permalink / raw)
  To: linux-pm; +Cc: linux-kernel, Len Brown

From: Len Brown <len.brown@intel.com>

This is a fully supported driver.

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/Kconfig |    1 -
 1 files changed, 0 insertions(+), 1 deletions(-)

diff --git a/drivers/idle/Kconfig b/drivers/idle/Kconfig
index 82e4e86..8489eb5 100644
--- a/drivers/idle/Kconfig
+++ b/drivers/idle/Kconfig
@@ -3,7 +3,6 @@ config INTEL_IDLE
 	depends on CPU_IDLE
 	depends on X86
 	depends on CPU_SUP_INTEL
-	depends on EXPERIMENTAL
 	help
 	  Enable intel_idle, a cpuidle driver that includes knowledge of
 	  native Intel hardware idle features.  The acpi_idle driver
-- 
1.7.2.1.95.g3d045


^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 4/5] intel_idle: no longer EXPERIMENTAL
  2010-08-15  5:22   ` Len Brown
                     ` (3 preceding siblings ...)
  (?)
@ 2010-08-15  5:22   ` Len Brown
  -1 siblings, 0 replies; 11+ messages in thread
From: Len Brown @ 2010-08-15  5:22 UTC (permalink / raw)
  To: linux-pm; +Cc: Len Brown, linux-kernel

From: Len Brown <len.brown@intel.com>

This is a fully supported driver.

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/Kconfig |    1 -
 1 files changed, 0 insertions(+), 1 deletions(-)

diff --git a/drivers/idle/Kconfig b/drivers/idle/Kconfig
index 82e4e86..8489eb5 100644
--- a/drivers/idle/Kconfig
+++ b/drivers/idle/Kconfig
@@ -3,7 +3,6 @@ config INTEL_IDLE
 	depends on CPU_IDLE
 	depends on X86
 	depends on CPU_SUP_INTEL
-	depends on EXPERIMENTAL
 	help
 	  Enable intel_idle, a cpuidle driver that includes knowledge of
 	  native Intel hardware idle features.  The acpi_idle driver
-- 
1.7.2.1.95.g3d045

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 5/5] intel_idle: recognize Lincroft Atom Processor
  2010-08-15  5:22   ` Len Brown
                     ` (6 preceding siblings ...)
  (?)
@ 2010-08-15  5:22   ` Len Brown
  -1 siblings, 0 replies; 11+ messages in thread
From: Len Brown @ 2010-08-15  5:22 UTC (permalink / raw)
  To: linux-pm; +Cc: linux-kernel, Arjan van de Ven, Len Brown

From: Arjan van de Ven <arjan@linux.intel.com>

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/intel_idle.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 97655ed..b1909fb 100755
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -265,6 +265,7 @@ static int intel_idle_probe(void)
 		break;
 
 	case 0x1C:	/* 28 - Atom Processor */
+	case 0x26:	/* 38 - Lincroft Atom Processor */
 		lapic_timer_reliable_states = (1 << 2) | (1 << 1); /* C2, C1 */
 		cpuidle_state_table = atom_cstates;
 		break;
-- 
1.7.2.1.95.g3d045


^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 5/5] intel_idle: recognize Lincroft Atom Processor
  2010-08-15  5:22   ` Len Brown
                     ` (5 preceding siblings ...)
  (?)
@ 2010-08-15  5:22   ` Len Brown
  -1 siblings, 0 replies; 11+ messages in thread
From: Len Brown @ 2010-08-15  5:22 UTC (permalink / raw)
  To: linux-pm; +Cc: Len Brown, Arjan van de Ven, linux-kernel

From: Arjan van de Ven <arjan@linux.intel.com>

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/intel_idle.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 97655ed..b1909fb 100755
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -265,6 +265,7 @@ static int intel_idle_probe(void)
 		break;
 
 	case 0x1C:	/* 28 - Atom Processor */
+	case 0x26:	/* 38 - Lincroft Atom Processor */
 		lapic_timer_reliable_states = (1 << 2) | (1 << 1); /* C2, C1 */
 		cpuidle_state_table = atom_cstates;
 		break;
-- 
1.7.2.1.95.g3d045

^ permalink raw reply related	[flat|nested] 11+ messages in thread

end of thread, other threads:[~2010-08-15  5:22 UTC | newest]

Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-08-15  5:22 idle patches for 2.6.36.merge Len Brown
2010-08-15  5:22 ` [PATCH 1/5] intel_idle: delete power_policy modparam, and choose substate functions Len Brown
2010-08-15  5:22   ` Len Brown
2010-08-15  5:22   ` [PATCH 2/5] intel_idle: add support for Westmere-EX Len Brown
2010-08-15  5:22     ` Len Brown
2010-08-15  5:22   ` [PATCH 3/5] intel_idle: disable module support Len Brown
2010-08-15  5:22   ` Len Brown
2010-08-15  5:22   ` [PATCH 4/5] intel_idle: no longer EXPERIMENTAL Len Brown
2010-08-15  5:22   ` Len Brown
2010-08-15  5:22   ` [PATCH 5/5] intel_idle: recognize Lincroft Atom Processor Len Brown
2010-08-15  5:22   ` Len Brown

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.