From 18c6faa9624bf5d9d7da3906a8a1b909dba5899b Mon Sep 17 00:00:00 2001
From: Paolo Ciarrocchi <paolo.ciarrocchi@gmail.com>
Date: Mon, 28 Jul 2008 13:00:49 +0200
Subject: [PATCH 01/14] [CPUFREQ] Coding style fixes to
 arch/x86/kernel/cpu/cpufreq/elanfreq.c

Before:
total: 15 errors, 10 warnings, 308 lines checked

After:
total: 0 errors, 4 warnings, 308 lines checked

paolo@paolo-desktop:~/linux.trees.git$ md5sum /tmp/elafreq.o.*
add1d36c2f077c5aab7682e8642a9f34  /tmp/elafreq.o.after
add1d36c2f077c5aab7682e8642a9f34  /tmp/elafreq.o.before

paolo@paolo-desktop:~/linux.trees.git$ size /tmp/elafreq.o.*
   text    data     bss     dec     hex filename
    934     270       4    1208     4b8 /tmp/elafreq.o.after
    934     270       4    1208     4b8 /tmp/elafreq.o.before

Signed-off-by: Paolo Ciarrocchi <paolo.ciarrocchi@gmail.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 arch/x86/kernel/cpu/cpufreq/elanfreq.c | 42 +++++++++++++-------------
 1 file changed, 21 insertions(+), 21 deletions(-)

diff --git a/arch/x86/kernel/cpu/cpufreq/elanfreq.c b/arch/x86/kernel/cpu/cpufreq/elanfreq.c
index e4a4bf870e94..fe613c93b366 100644
--- a/arch/x86/kernel/cpu/cpufreq/elanfreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/elanfreq.c
@@ -25,8 +25,8 @@
 #include <linux/cpufreq.h>
 
 #include <asm/msr.h>
-#include <asm/timex.h>
-#include <asm/io.h>
+#include <linux/timex.h>
+#include <linux/io.h>
 
 #define REG_CSCIR 0x22		/* Chip Setup and Control Index Register    */
 #define REG_CSCDR 0x23		/* Chip Setup and Control Data  Register    */
@@ -82,7 +82,7 @@ static unsigned int elanfreq_get_cpu_frequency(unsigned int cpu)
 	u8 clockspeed_reg;    /* Clock Speed Register */
 
 	local_irq_disable();
-	outb_p(0x80,REG_CSCIR);
+	outb_p(0x80, REG_CSCIR);
 	clockspeed_reg = inb_p(REG_CSCDR);
 	local_irq_enable();
 
@@ -98,10 +98,10 @@ static unsigned int elanfreq_get_cpu_frequency(unsigned int cpu)
 	}
 
 	/* 33 MHz is not 32 MHz... */
-	if ((clockspeed_reg & 0xE0)==0xA0)
+	if ((clockspeed_reg & 0xE0) == 0xA0)
 		return 33000;
 
-	return ((1<<((clockspeed_reg & 0xE0) >> 5)) * 1000);
+	return (1<<((clockspeed_reg & 0xE0) >> 5)) * 1000;
 }
 
 
@@ -117,7 +117,7 @@ static unsigned int elanfreq_get_cpu_frequency(unsigned int cpu)
  *	There is no return value.
  */
 
-static void elanfreq_set_cpu_state (unsigned int state)
+static void elanfreq_set_cpu_state(unsigned int state)
 {
 	struct cpufreq_freqs    freqs;
 
@@ -144,20 +144,20 @@ static void elanfreq_set_cpu_state (unsigned int state)
 	 */
 
 	local_irq_disable();
-	outb_p(0x40,REG_CSCIR);		/* Disable hyperspeed mode */
-	outb_p(0x00,REG_CSCDR);
+	outb_p(0x40, REG_CSCIR);		/* Disable hyperspeed mode */
+	outb_p(0x00, REG_CSCDR);
 	local_irq_enable();		/* wait till internal pipelines and */
 	udelay(1000);			/* buffers have cleaned up          */
 
 	local_irq_disable();
 
 	/* now, set the CPU clock speed register (0x80) */
-	outb_p(0x80,REG_CSCIR);
-	outb_p(elan_multiplier[state].val80h,REG_CSCDR);
+	outb_p(0x80, REG_CSCIR);
+	outb_p(elan_multiplier[state].val80h, REG_CSCDR);
 
 	/* now, the hyperspeed bit in PMU Force Mode Register (0x40) */
-	outb_p(0x40,REG_CSCIR);
-	outb_p(elan_multiplier[state].val40h,REG_CSCDR);
+	outb_p(0x40, REG_CSCIR);
+	outb_p(elan_multiplier[state].val40h, REG_CSCDR);
 	udelay(10000);
 	local_irq_enable();
 
@@ -173,12 +173,12 @@ static void elanfreq_set_cpu_state (unsigned int state)
  *	for the hardware supported by the driver.
  */
 
-static int elanfreq_verify (struct cpufreq_policy *policy)
+static int elanfreq_verify(struct cpufreq_policy *policy)
 {
 	return cpufreq_frequency_table_verify(policy, &elanfreq_table[0]);
 }
 
-static int elanfreq_target (struct cpufreq_policy *policy,
+static int elanfreq_target(struct cpufreq_policy *policy,
 			    unsigned int target_freq,
 			    unsigned int relation)
 {
@@ -205,7 +205,7 @@ static int elanfreq_cpu_init(struct cpufreq_policy *policy)
 
 	/* capability check */
 	if ((c->x86_vendor != X86_VENDOR_AMD) ||
-	    (c->x86 != 4) || (c->x86_model!=10))
+	    (c->x86 != 4) || (c->x86_model != 10))
 		return -ENODEV;
 
 	/* max freq */
@@ -213,7 +213,7 @@ static int elanfreq_cpu_init(struct cpufreq_policy *policy)
 		max_freq = elanfreq_get_cpu_frequency(0);
 
 	/* table init */
-	for (i=0; (elanfreq_table[i].frequency != CPUFREQ_TABLE_END); i++) {
+	for (i = 0; (elanfreq_table[i].frequency != CPUFREQ_TABLE_END); i++) {
 		if (elanfreq_table[i].frequency > max_freq)
 			elanfreq_table[i].frequency = CPUFREQ_ENTRY_INVALID;
 	}
@@ -224,7 +224,7 @@ static int elanfreq_cpu_init(struct cpufreq_policy *policy)
 
 	result = cpufreq_frequency_table_cpuinfo(policy, elanfreq_table);
 	if (result)
-		return (result);
+		return result;
 
 	cpufreq_frequency_table_get_attr(elanfreq_table, policy->cpu);
 	return 0;
@@ -260,7 +260,7 @@ __setup("elanfreq=", elanfreq_setup);
 #endif
 
 
-static struct freq_attr* elanfreq_attr[] = {
+static struct freq_attr *elanfreq_attr[] = {
 	&cpufreq_freq_attr_scaling_available_freqs,
 	NULL,
 };
@@ -284,9 +284,9 @@ static int __init elanfreq_init(void)
 
 	/* Test if we have the right hardware */
 	if ((c->x86_vendor != X86_VENDOR_AMD) ||
-		(c->x86 != 4) || (c->x86_model!=10)) {
+		(c->x86 != 4) || (c->x86_model != 10)) {
 		printk(KERN_INFO "elanfreq: error: no Elan processor found!\n");
-                return -ENODEV;
+		return -ENODEV;
 	}
 	return cpufreq_register_driver(&elanfreq_driver);
 }
@@ -298,7 +298,7 @@ static void __exit elanfreq_exit(void)
 }
 
 
-module_param (max_freq, int, 0444);
+module_param(max_freq, int, 0444);
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Robert Schwebel <r.schwebel@pengutronix.de>, Sven Geggus <sven@geggus.net>");

From 8d2d2051e50c4ca0207a42b243369e22f7d90a5c Mon Sep 17 00:00:00 2001
From: Paolo Ciarrocchi <paolo.ciarrocchi@gmail.com>
Date: Mon, 28 Jul 2008 21:08:16 +0200
Subject: [PATCH 02/14] [CPUFREQ] Coding style fixes to
 arch/x86/kernel/cpu/cpufreq/powernow-k6.c

Before:
total: 11 errors, 15 warnings, 255 lines checked

After:
total: 0 errors, 6 warnings, 254 lines checked

paolo@paolo-desktop:~/linux.trees.git$ md5sum /tmp/powernow-k6.o.*
476932f5e1ffe365db9d1dfb3f860369  /tmp/powernow-k6.o.after
476932f5e1ffe365db9d1dfb3f860369  /tmp/powernow-k6.o.before

Signed-off-by: Paolo Ciarrocchi <paolo.ciarrocchi@gmail.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 arch/x86/kernel/cpu/cpufreq/powernow-k6.c | 41 +++++++++++------------
 1 file changed, 20 insertions(+), 21 deletions(-)

diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c
index eb9b62b0830c..b5ced806a316 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c
@@ -15,12 +15,11 @@
 #include <linux/slab.h>
 
 #include <asm/msr.h>
-#include <asm/timex.h>
-#include <asm/io.h>
+#include <linux/timex.h>
+#include <linux/io.h>
 
-
-#define POWERNOW_IOPORT 0xfff0         /* it doesn't matter where, as long
-					  as it is unused */
+#define POWERNOW_IOPORT 0xfff0          /* it doesn't matter where, as long
+					   as it is unused */
 
 static unsigned int                     busfreq;   /* FSB, in 10 kHz */
 static unsigned int                     max_multiplier;
@@ -53,7 +52,7 @@ static int powernow_k6_get_cpu_multiplier(void)
 
 	msrval = POWERNOW_IOPORT + 0x1;
 	wrmsr(MSR_K6_EPMR, msrval, 0); /* enable the PowerNow port */
-	invalue=inl(POWERNOW_IOPORT + 0x8);
+	invalue = inl(POWERNOW_IOPORT + 0x8);
 	msrval = POWERNOW_IOPORT + 0x0;
 	wrmsr(MSR_K6_EPMR, msrval, 0); /* disable it again */
 
@@ -67,9 +66,9 @@ static int powernow_k6_get_cpu_multiplier(void)
  *
  *   Tries to change the PowerNow! multiplier
  */
-static void powernow_k6_set_state (unsigned int best_i)
+static void powernow_k6_set_state(unsigned int best_i)
 {
-	unsigned long           outvalue=0, invalue=0;
+	unsigned long           outvalue = 0, invalue = 0;
 	unsigned long           msrval;
 	struct cpufreq_freqs    freqs;
 
@@ -90,10 +89,10 @@ static void powernow_k6_set_state (unsigned int best_i)
 
 	msrval = POWERNOW_IOPORT + 0x1;
 	wrmsr(MSR_K6_EPMR, msrval, 0); /* enable the PowerNow port */
-	invalue=inl(POWERNOW_IOPORT + 0x8);
+	invalue = inl(POWERNOW_IOPORT + 0x8);
 	invalue = invalue & 0xf;
 	outvalue = outvalue | invalue;
-	outl(outvalue ,(POWERNOW_IOPORT + 0x8));
+	outl(outvalue , (POWERNOW_IOPORT + 0x8));
 	msrval = POWERNOW_IOPORT + 0x0;
 	wrmsr(MSR_K6_EPMR, msrval, 0); /* disable it again */
 
@@ -124,7 +123,7 @@ static int powernow_k6_verify(struct cpufreq_policy *policy)
  *
  * sets a new CPUFreq policy
  */
-static int powernow_k6_target (struct cpufreq_policy *policy,
+static int powernow_k6_target(struct cpufreq_policy *policy,
 			       unsigned int target_freq,
 			       unsigned int relation)
 {
@@ -152,7 +151,7 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy)
 	busfreq = cpu_khz / max_multiplier;
 
 	/* table init */
-	for (i=0; (clock_ratio[i].frequency != CPUFREQ_TABLE_END); i++) {
+	for (i = 0; (clock_ratio[i].frequency != CPUFREQ_TABLE_END); i++) {
 		if (clock_ratio[i].index > max_multiplier)
 			clock_ratio[i].frequency = CPUFREQ_ENTRY_INVALID;
 		else
@@ -165,7 +164,7 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy)
 
 	result = cpufreq_frequency_table_cpuinfo(policy, clock_ratio);
 	if (result)
-		return (result);
+		return result;
 
 	cpufreq_frequency_table_get_attr(clock_ratio, policy->cpu);
 
@@ -176,8 +175,8 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy)
 static int powernow_k6_cpu_exit(struct cpufreq_policy *policy)
 {
 	unsigned int i;
-	for (i=0; i<8; i++) {
-		if (i==max_multiplier)
+	for (i = 0; i < 8; i++) {
+		if (i == max_multiplier)
 			powernow_k6_set_state(i);
 	}
 	cpufreq_frequency_table_put_attr(policy->cpu);
@@ -189,7 +188,7 @@ static unsigned int powernow_k6_get(unsigned int cpu)
 	return busfreq * powernow_k6_get_cpu_multiplier();
 }
 
-static struct freq_attr* powernow_k6_attr[] = {
+static struct freq_attr *powernow_k6_attr[] = {
 	&cpufreq_freq_attr_scaling_available_freqs,
 	NULL,
 };
@@ -227,7 +226,7 @@ static int __init powernow_k6_init(void)
 	}
 
 	if (cpufreq_register_driver(&powernow_k6_driver)) {
-		release_region (POWERNOW_IOPORT, 16);
+		release_region(POWERNOW_IOPORT, 16);
 		return -EINVAL;
 	}
 
@@ -243,13 +242,13 @@ static int __init powernow_k6_init(void)
 static void __exit powernow_k6_exit(void)
 {
 	cpufreq_unregister_driver(&powernow_k6_driver);
-	release_region (POWERNOW_IOPORT, 16);
+	release_region(POWERNOW_IOPORT, 16);
 }
 
 
-MODULE_AUTHOR ("Arjan van de Ven <arjanv@redhat.com>, Dave Jones <davej@codemonkey.org.uk>, Dominik Brodowski <linux@brodo.de>");
-MODULE_DESCRIPTION ("PowerNow! driver for AMD K6-2+ / K6-3+ processors.");
-MODULE_LICENSE ("GPL");
+MODULE_AUTHOR("Arjan van de Ven <arjanv@redhat.com>, Dave Jones <davej@codemonkey.org.uk>, Dominik Brodowski <linux@brodo.de>");
+MODULE_DESCRIPTION("PowerNow! driver for AMD K6-2+ / K6-3+ processors.");
+MODULE_LICENSE("GPL");
 
 module_init(powernow_k6_init);
 module_exit(powernow_k6_exit);

From 847aef6ffd85787b62395b64719f8f7c5975bf1b Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Mon, 14 Jul 2008 11:59:44 +0900
Subject: [PATCH 03/14] [CPUFREQ] acpi-cpufreq: add error handling for
 cpufreq_register_driver() error

add error handling for cpufreq_register_driver() error

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: cpufreq@lists.linux.org.uk
Signed-off-by: Dave Jones <davej@redhat.com>
---
 arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index dd097b835839..1def5b06fa4a 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -785,7 +785,11 @@ static int __init acpi_cpufreq_init(void)
 	if (ret)
 		return ret;
 
-	return cpufreq_register_driver(&acpi_cpufreq_driver);
+	ret = cpufreq_register_driver(&acpi_cpufreq_driver);
+	if (ret)
+		free_percpu(acpi_perf_data);
+
+	return ret;
 }
 
 static void __exit acpi_cpufreq_exit(void)
@@ -795,8 +799,6 @@ static void __exit acpi_cpufreq_exit(void)
 	cpufreq_unregister_driver(&acpi_cpufreq_driver);
 
 	free_percpu(acpi_perf_data);
-
-	return;
 }
 
 module_param(acpi_pstate_strict, uint, 0644);

From 888a794cacd8950ac6be701db20b62a4ab2ce90c Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Mon, 14 Jul 2008 12:00:45 +0900
Subject: [PATCH 04/14] [CPUFREQ] add error handling for
 cpufreq_register_governor() error

Add error handling for cpufreq_register_governor() error

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: cpufreq@lists.linux.org.uk
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq_ondemand.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index 33855cb3cf16..67c9d4f9edc0 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -589,12 +589,18 @@ EXPORT_SYMBOL(cpufreq_gov_ondemand);
 
 static int __init cpufreq_gov_dbs_init(void)
 {
+	int err;
+
 	kondemand_wq = create_workqueue("kondemand");
 	if (!kondemand_wq) {
 		printk(KERN_ERR "Creation of kondemand failed\n");
 		return -EFAULT;
 	}
-	return cpufreq_register_governor(&cpufreq_gov_ondemand);
+	err = cpufreq_register_governor(&cpufreq_gov_ondemand);
+	if (err)
+		destroy_workqueue(kondemand_wq);
+
+	return err;
 }
 
 static void __exit cpufreq_gov_dbs_exit(void)

From f1829e4a371f26430185a9d5b97b8d9d19824e08 Mon Sep 17 00:00:00 2001
From: Julia Lawall <julia@diku.dk>
Date: Fri, 25 Jul 2008 22:44:53 +0200
Subject: [PATCH 05/14] [CPUFREQ] drivers/cpufreq/cpufreq.c: Adjust error
 handling code involving cpufreq_cpu_put

After calling cpufreq_cpu_get, error handling code should call
cpufreq_cpu_put.

The semantic match that finds this problem is as follows:
(http://www.emn.fr/x-info/coccinelle/)

// <smpl>
@r@
expression x,E;
statement S;
position p1,p2,p3;
@@

(
if ((x = cpufreq_cpu_get@p1(...)) == NULL || ...) S
|
x = cpufreq_cpu_get@p1(...)
... when != x
if (x == NULL || ...) S
)
<...
if@p3 (...) { ... when != cpufreq_cpu_put(x)
                  when != if (x) { ... cpufreq_cpu_put(x); ...}
    return@p2 ...;
}
...>
(
return x;
|
return 0;
|
x = E
|
E = x
|
cpufreq_cpu_put(x)
)

@exists@
position r.p1,r.p2,r.p3;
expression x;
int ret != 0;
statement S;
@@

* x = cpufreq_cpu_get@p1(...)
  <...
* if@p3 (...)
  S
  ...>
* return@p2 \(NULL\|ret\);
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq.c | 24 ++++++++++++++++--------
 1 file changed, 16 insertions(+), 8 deletions(-)

diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 8a67f16987db..9bbdc258624c 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -1467,20 +1467,22 @@ int cpufreq_driver_target(struct cpufreq_policy *policy,
 			  unsigned int target_freq,
 			  unsigned int relation)
 {
-	int ret;
+	int ret = -EINVAL;
 
 	policy = cpufreq_cpu_get(policy->cpu);
 	if (!policy)
-		return -EINVAL;
+		goto no_policy;
 
 	if (unlikely(lock_policy_rwsem_write(policy->cpu)))
-		return -EINVAL;
+		goto fail;
 
 	ret = __cpufreq_driver_target(policy, target_freq, relation);
 
 	unlock_policy_rwsem_write(policy->cpu);
 
+fail:
 	cpufreq_cpu_put(policy);
+no_policy:
 	return ret;
 }
 EXPORT_SYMBOL_GPL(cpufreq_driver_target);
@@ -1717,13 +1719,17 @@ int cpufreq_update_policy(unsigned int cpu)
 {
 	struct cpufreq_policy *data = cpufreq_cpu_get(cpu);
 	struct cpufreq_policy policy;
-	int ret = 0;
+	int ret;
 
-	if (!data)
-		return -ENODEV;
+	if (!data) {
+		ret = -ENODEV;
+		goto no_policy;
+	}
 
-	if (unlikely(lock_policy_rwsem_write(cpu)))
-		return -EINVAL;
+	if (unlikely(lock_policy_rwsem_write(cpu))) {
+		ret = -EINVAL;
+		goto fail;
+	}
 
 	dprintk("updating policy for CPU %u\n", cpu);
 	memcpy(&policy, data, sizeof(struct cpufreq_policy));
@@ -1750,7 +1756,9 @@ int cpufreq_update_policy(unsigned int cpu)
 
 	unlock_policy_rwsem_write(cpu);
 
+fail:
 	cpufreq_cpu_put(data);
+no_policy:
 	return ret;
 }
 EXPORT_SYMBOL(cpufreq_update_policy);

From 8217e4f4c93e5fb59bb3cd1e6135213889349f86 Mon Sep 17 00:00:00 2001
From: Ben Slusky <sluskyb@paranoiacs.org>
Date: Mon, 7 Jul 2008 13:16:20 -0400
Subject: [PATCH 06/14] [CPUFREQ] use deferrable delayed work init in
 conservative governor

Venki Pallipadi made a similar change to the ondemand governor a while
back (in commit 28287033e12463c8ff89f1ea8038783d0360391c). It seems to
work just as well in the conservative governor, leading to fewer wakeups
as reported by powertop.

Signed-off-by: Ben Slusky <sluskyb@paranoiacs.org>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq_conservative.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/cpufreq/cpufreq_conservative.c b/drivers/cpufreq/cpufreq_conservative.c
index ac0bbf2d234f..4ee53a4c6d2c 100644
--- a/drivers/cpufreq/cpufreq_conservative.c
+++ b/drivers/cpufreq/cpufreq_conservative.c
@@ -460,6 +460,7 @@ static void do_dbs_timer(struct work_struct *work)
 
 static inline void dbs_timer_init(void)
 {
+	init_timer_deferrable(&dbs_work.timer);
 	schedule_delayed_work(&dbs_work,
 			usecs_to_jiffies(dbs_tuners_ins.sampling_rate));
 	return;

From bf0b90e357c883e8efd72954432efe652de74c76 Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Mon, 4 Aug 2008 11:59:07 -0700
Subject: [PATCH 07/14] [CPUFREQ][1/6] cpufreq: Add cpu number parameter to
 __cpufreq_driver_getavg()

Add a cpu parameter to __cpufreq_driver_getavg(). This is needed for software
cpufreq coordination where policy->cpu may not be same as the CPU on which we
want to getavg frequency.

A follow-on patch will use this parameter to getavg freq from all cpus
in policy->cpus.

Change since last patch. Fix the offline/online and suspend/resume
oops reported by Youquan Song <youquan.song@intel.com>

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c | 5 +++--
 drivers/cpufreq/cpufreq.c                  | 6 +++---
 drivers/cpufreq/cpufreq_ondemand.c         | 2 +-
 include/linux/cpufreq.h                    | 7 +++++--
 4 files changed, 12 insertions(+), 8 deletions(-)

diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index 1def5b06fa4a..c24c4a487b7c 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -256,7 +256,8 @@ static u32 get_cur_val(const cpumask_t *mask)
  * Only IA32_APERF/IA32_MPERF ratio is architecturally defined and
  * no meaning should be associated with absolute values of these MSRs.
  */
-static unsigned int get_measured_perf(unsigned int cpu)
+static unsigned int get_measured_perf(struct cpufreq_policy *policy,
+				      unsigned int cpu)
 {
 	union {
 		struct {
@@ -326,7 +327,7 @@ static unsigned int get_measured_perf(unsigned int cpu)
 
 #endif
 
-	retval = per_cpu(drv_data, cpu)->max_freq * perf_percent / 100;
+	retval = per_cpu(drv_data, policy->cpu)->max_freq * perf_percent / 100;
 
 	put_cpu();
 	set_cpus_allowed_ptr(current, &saved_mask);
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 9bbdc258624c..31d6f535a79d 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -1487,7 +1487,7 @@ no_policy:
 }
 EXPORT_SYMBOL_GPL(cpufreq_driver_target);
 
-int __cpufreq_driver_getavg(struct cpufreq_policy *policy)
+int __cpufreq_driver_getavg(struct cpufreq_policy *policy, unsigned int cpu)
 {
 	int ret = 0;
 
@@ -1495,8 +1495,8 @@ int __cpufreq_driver_getavg(struct cpufreq_policy *policy)
 	if (!policy)
 		return -EINVAL;
 
-	if (cpu_online(policy->cpu) && cpufreq_driver->getavg)
-		ret = cpufreq_driver->getavg(policy->cpu);
+	if (cpu_online(cpu) && cpufreq_driver->getavg)
+		ret = cpufreq_driver->getavg(policy, cpu);
 
 	cpufreq_cpu_put(policy);
 	return ret;
diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index 67c9d4f9edc0..f56debd9a8d7 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -415,7 +415,7 @@ static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info)
 	if (load < (dbs_tuners_ins.up_threshold - 10)) {
 		unsigned int freq_next, freq_cur;
 
-		freq_cur = __cpufreq_driver_getavg(policy);
+		freq_cur = __cpufreq_driver_getavg(policy, policy->cpu);
 		if (!freq_cur)
 			freq_cur = policy->cur;
 
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 6fd5668aa572..1ee608fd7b77 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -187,7 +187,8 @@ extern int __cpufreq_driver_target(struct cpufreq_policy *policy,
 				   unsigned int relation);
 
 
-extern int __cpufreq_driver_getavg(struct cpufreq_policy *policy);
+extern int __cpufreq_driver_getavg(struct cpufreq_policy *policy,
+				   unsigned int cpu);
 
 int cpufreq_register_governor(struct cpufreq_governor *governor);
 void cpufreq_unregister_governor(struct cpufreq_governor *governor);
@@ -226,7 +227,9 @@ struct cpufreq_driver {
 	unsigned int	(*get)	(unsigned int cpu);
 
 	/* optional */
-	unsigned int (*getavg)	(unsigned int cpu);
+	unsigned int (*getavg)	(struct cpufreq_policy *policy,
+				 unsigned int cpu);
+
 	int	(*exit)		(struct cpufreq_policy *policy);
 	int	(*suspend)	(struct cpufreq_policy *policy, pm_message_t pmsg);
 	int	(*resume)	(struct cpufreq_policy *policy);

From c43aa3bd99a67009a167430e80c5fde6f37288d8 Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Mon, 4 Aug 2008 11:59:08 -0700
Subject: [PATCH 08/14] [CPUFREQ][2/6] cpufreq: Change load calculation in
 ondemand for software coordination

Change the load calculation algorithm in ondemand to work well with software
coordination of frequency across the dependent cpus.

Multiply individual CPU utilization with the average freq of that logical CPU
during the measurement interval (using getavg call). And find the max CPU
utilization number in terms of CPU freq. That number is then used to
get to the target freq for next sampling interval.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq_ondemand.c | 67 ++++++++++++++++--------------
 1 file changed, 36 insertions(+), 31 deletions(-)

diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index f56debd9a8d7..42fcb146ba22 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -334,9 +334,7 @@ static struct attribute_group dbs_attr_group = {
 
 static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info)
 {
-	unsigned int idle_ticks, total_ticks;
-	unsigned int load = 0;
-	cputime64_t cur_jiffies;
+	unsigned int max_load_freq;
 
 	struct cpufreq_policy *policy;
 	unsigned int j;
@@ -346,13 +344,7 @@ static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info)
 
 	this_dbs_info->freq_lo = 0;
 	policy = this_dbs_info->cur_policy;
-	cur_jiffies = jiffies64_to_cputime64(get_jiffies_64());
-	total_ticks = (unsigned int) cputime64_sub(cur_jiffies,
-			this_dbs_info->prev_cpu_wall);
-	this_dbs_info->prev_cpu_wall = get_jiffies_64();
 
-	if (!total_ticks)
-		return;
 	/*
 	 * Every sampling_rate, we check, if current idle time is less
 	 * than 20% (default), then we try to increase frequency
@@ -365,27 +357,46 @@ static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info)
 	 * 5% (default) of current frequency
 	 */
 
-	/* Get Idle Time */
-	idle_ticks = UINT_MAX;
+	/* Get Absolute Load - in terms of freq */
+	max_load_freq = 0;
+
 	for_each_cpu_mask_nr(j, policy->cpus) {
-		cputime64_t total_idle_ticks;
-		unsigned int tmp_idle_ticks;
 		struct cpu_dbs_info_s *j_dbs_info;
+		cputime64_t cur_wall_time, cur_idle_time;
+		unsigned int idle_time, wall_time;
+		unsigned int load, load_freq;
+		int freq_avg;
 
 		j_dbs_info = &per_cpu(cpu_dbs_info, j);
-		total_idle_ticks = get_cpu_idle_time(j);
-		tmp_idle_ticks = (unsigned int) cputime64_sub(total_idle_ticks,
-				j_dbs_info->prev_cpu_idle);
-		j_dbs_info->prev_cpu_idle = total_idle_ticks;
+		cur_wall_time = jiffies64_to_cputime64(get_jiffies_64());
+		wall_time = (unsigned int) cputime64_sub(cur_wall_time,
+				j_dbs_info->prev_cpu_wall);
+		j_dbs_info->prev_cpu_wall = cur_wall_time;
 
-		if (tmp_idle_ticks < idle_ticks)
-			idle_ticks = tmp_idle_ticks;
+		cur_idle_time = get_cpu_idle_time(j);
+		idle_time = (unsigned int) cputime64_sub(cur_idle_time,
+				j_dbs_info->prev_cpu_idle);
+		j_dbs_info->prev_cpu_idle = cur_idle_time;
+
+		if (unlikely(wall_time <= idle_time ||
+			     (cputime_to_msecs(wall_time) <
+			      dbs_tuners_ins.sampling_rate / (2 * 1000)))) {
+			continue;
+		}
+
+		load = 100 * (wall_time - idle_time) / wall_time;
+
+		freq_avg = __cpufreq_driver_getavg(policy, j);
+		if (freq_avg <= 0)
+			freq_avg = policy->cur;
+
+		load_freq = load * freq_avg;
+		if (load_freq > max_load_freq)
+			max_load_freq = load_freq;
 	}
-	if (likely(total_ticks > idle_ticks))
-		load = (100 * (total_ticks - idle_ticks)) / total_ticks;
 
 	/* Check for frequency increase */
-	if (load > dbs_tuners_ins.up_threshold) {
+	if (max_load_freq > dbs_tuners_ins.up_threshold * policy->cur) {
 		/* if we are already at full speed then break out early */
 		if (!dbs_tuners_ins.powersave_bias) {
 			if (policy->cur == policy->max)
@@ -412,15 +423,9 @@ static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info)
 	 * can support the current CPU usage without triggering the up
 	 * policy. To be safe, we focus 10 points under the threshold.
 	 */
-	if (load < (dbs_tuners_ins.up_threshold - 10)) {
-		unsigned int freq_next, freq_cur;
-
-		freq_cur = __cpufreq_driver_getavg(policy, policy->cpu);
-		if (!freq_cur)
-			freq_cur = policy->cur;
-
-		freq_next = (freq_cur * load) /
-			(dbs_tuners_ins.up_threshold - 10);
+	if (max_load_freq < (dbs_tuners_ins.up_threshold - 10) * policy->cur) {
+		unsigned int freq_next;
+		freq_next = max_load_freq / (dbs_tuners_ins.up_threshold - 10);
 
 		if (!dbs_tuners_ins.powersave_bias) {
 			__cpufreq_driver_target(policy, freq_next,

From 3430502d356284ff4f7782d75bb01a402fd3d45e Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Mon, 4 Aug 2008 11:59:09 -0700
Subject: [PATCH 09/14] [CPUFREQ][3/6] cpufreq: get_cpu_idle_time() changes in
 ondemand for idle-microaccounting

Preparatory changes for doing idle micro-accounting in ondemand governor.
get_cpu_idle_time() gets extra parameter and returns idle time and also the
wall time that corresponds to the idle time measurement.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq_ondemand.c | 29 +++++++++++++++--------------
 1 file changed, 15 insertions(+), 14 deletions(-)

diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index 42fcb146ba22..b935092aab21 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -94,13 +94,13 @@ static struct dbs_tuners {
 	.powersave_bias = 0,
 };
 
-static inline cputime64_t get_cpu_idle_time(unsigned int cpu)
+static inline cputime64_t get_cpu_idle_time(unsigned int cpu, cputime64_t *wall)
 {
 	cputime64_t idle_time;
-	cputime64_t cur_jiffies;
+	cputime64_t cur_wall_time;
 	cputime64_t busy_time;
 
-	cur_jiffies = jiffies64_to_cputime64(get_jiffies_64());
+	cur_wall_time = jiffies64_to_cputime64(get_jiffies_64());
 	busy_time = cputime64_add(kstat_cpu(cpu).cpustat.user,
 			kstat_cpu(cpu).cpustat.system);
 
@@ -113,7 +113,10 @@ static inline cputime64_t get_cpu_idle_time(unsigned int cpu)
 				kstat_cpu(cpu).cpustat.nice);
 	}
 
-	idle_time = cputime64_sub(cur_jiffies, busy_time);
+	idle_time = cputime64_sub(cur_wall_time, busy_time);
+	if (wall)
+		*wall = cur_wall_time;
+
 	return idle_time;
 }
 
@@ -277,8 +280,8 @@ static ssize_t store_ignore_nice_load(struct cpufreq_policy *policy,
 	for_each_online_cpu(j) {
 		struct cpu_dbs_info_s *dbs_info;
 		dbs_info = &per_cpu(cpu_dbs_info, j);
-		dbs_info->prev_cpu_idle = get_cpu_idle_time(j);
-		dbs_info->prev_cpu_wall = get_jiffies_64();
+		dbs_info->prev_cpu_idle = get_cpu_idle_time(j,
+						&dbs_info->prev_cpu_wall);
 	}
 	mutex_unlock(&dbs_mutex);
 
@@ -368,21 +371,19 @@ static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info)
 		int freq_avg;
 
 		j_dbs_info = &per_cpu(cpu_dbs_info, j);
-		cur_wall_time = jiffies64_to_cputime64(get_jiffies_64());
+
+		cur_idle_time = get_cpu_idle_time(j, &cur_wall_time);
+
 		wall_time = (unsigned int) cputime64_sub(cur_wall_time,
 				j_dbs_info->prev_cpu_wall);
 		j_dbs_info->prev_cpu_wall = cur_wall_time;
 
-		cur_idle_time = get_cpu_idle_time(j);
 		idle_time = (unsigned int) cputime64_sub(cur_idle_time,
 				j_dbs_info->prev_cpu_idle);
 		j_dbs_info->prev_cpu_idle = cur_idle_time;
 
-		if (unlikely(wall_time <= idle_time ||
-			     (cputime_to_msecs(wall_time) <
-			      dbs_tuners_ins.sampling_rate / (2 * 1000)))) {
+		if (unlikely(!wall_time || wall_time < idle_time))
 			continue;
-		}
 
 		load = 100 * (wall_time - idle_time) / wall_time;
 
@@ -531,8 +532,8 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy,
 			j_dbs_info = &per_cpu(cpu_dbs_info, j);
 			j_dbs_info->cur_policy = policy;
 
-			j_dbs_info->prev_cpu_idle = get_cpu_idle_time(j);
-			j_dbs_info->prev_cpu_wall = get_jiffies_64();
+			j_dbs_info->prev_cpu_idle = get_cpu_idle_time(j,
+						&j_dbs_info->prev_cpu_wall);
 		}
 		this_dbs_info->cpu = cpu;
 		/*

From e9d95bf7eb929b9ddc9af9f4327b76c77ed4c7d6 Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Mon, 4 Aug 2008 11:59:10 -0700
Subject: [PATCH 10/14] [CPUFREQ][4/6] cpufreq_ondemand: Parameterize down
 differential

Use a parameter for down differential, instead of hardcoded 10%. Follow-on
patch changes the down-differential dynamically, based on whether
we are using idle micro-accounting or not.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq_ondemand.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index b935092aab21..3f898606e68c 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -24,6 +24,7 @@
  * It helps to keep variable names smaller, simpler
  */
 
+#define DEF_FREQUENCY_DOWN_DIFFERENTIAL		(10)
 #define DEF_FREQUENCY_UP_THRESHOLD		(80)
 #define MIN_FREQUENCY_UP_THRESHOLD		(11)
 #define MAX_FREQUENCY_UP_THRESHOLD		(100)
@@ -86,10 +87,12 @@ static struct workqueue_struct	*kondemand_wq;
 static struct dbs_tuners {
 	unsigned int sampling_rate;
 	unsigned int up_threshold;
+	unsigned int down_differential;
 	unsigned int ignore_nice;
 	unsigned int powersave_bias;
 } dbs_tuners_ins = {
 	.up_threshold = DEF_FREQUENCY_UP_THRESHOLD,
+	.down_differential = DEF_FREQUENCY_DOWN_DIFFERENTIAL,
 	.ignore_nice = 0,
 	.powersave_bias = 0,
 };
@@ -424,9 +427,13 @@ static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info)
 	 * can support the current CPU usage without triggering the up
 	 * policy. To be safe, we focus 10 points under the threshold.
 	 */
-	if (max_load_freq < (dbs_tuners_ins.up_threshold - 10) * policy->cur) {
+	if (max_load_freq <
+	    (dbs_tuners_ins.up_threshold - dbs_tuners_ins.down_differential) *
+	     policy->cur) {
 		unsigned int freq_next;
-		freq_next = max_load_freq / (dbs_tuners_ins.up_threshold - 10);
+		freq_next = max_load_freq /
+				(dbs_tuners_ins.up_threshold -
+				 dbs_tuners_ins.down_differential);
 
 		if (!dbs_tuners_ins.powersave_bias) {
 			__cpufreq_driver_target(policy, freq_next,

From 8083e4ad970e4eb567e31037060cdd4ba346f0c0 Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Mon, 4 Aug 2008 11:59:11 -0700
Subject: [PATCH 11/14] [CPUFREQ][5/6] cpufreq: Changes to
 get_cpu_idle_time_us(), used by ondemand governor

export get_cpu_idle_time_us() for it to be used in ondemand governor.
Last update time can be current time when the CPU is currently non-idle,
accounting for the busy time since last idle.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 include/linux/tick.h     |  2 +-
 kernel/time/tick-sched.c | 11 ++++++++++-
 2 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/include/linux/tick.h b/include/linux/tick.h
index 8cf8cfe2cc97..98921a3e1aa8 100644
--- a/include/linux/tick.h
+++ b/include/linux/tick.h
@@ -126,7 +126,7 @@ static inline ktime_t tick_nohz_get_sleep_length(void)
 	return len;
 }
 static inline void tick_nohz_stop_idle(int cpu) { }
-static inline u64 get_cpu_idle_time_us(int cpu, u64 *unused) { return 0; }
+static inline u64 get_cpu_idle_time_us(int cpu, u64 *unused) { return -1; }
 # endif /* !NO_HZ */
 
 #endif
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index cb02324bdb88..a4d219398167 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -20,6 +20,7 @@
 #include <linux/profile.h>
 #include <linux/sched.h>
 #include <linux/tick.h>
+#include <linux/module.h>
 
 #include <asm/irq_regs.h>
 
@@ -190,9 +191,17 @@ u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time)
 {
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
 
-	*last_update_time = ktime_to_us(ts->idle_lastupdate);
+	if (!tick_nohz_enabled)
+		return -1;
+
+	if (ts->idle_active)
+		*last_update_time = ktime_to_us(ts->idle_lastupdate);
+	else
+		*last_update_time = ktime_to_us(ktime_get());
+
 	return ktime_to_us(ts->idle_sleeptime);
 }
+EXPORT_SYMBOL_GPL(get_cpu_idle_time_us);
 
 /**
  * tick_nohz_stop_sched_tick - stop the idle tick from the idle task

From 808009131046b62ac434dbc796c0fe8accaab415 Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Mon, 4 Aug 2008 11:59:12 -0700
Subject: [PATCH 12/14] [CPUFREQ][6/6] cpufreq: Add idle microaccounting in
 ondemand governor

Use get_cpu_idle_time_us() to get micro-accounted idle information.
This enables ondemand to get more accurate idle and busy timings
than the jiffy based calculation. As a result, we can decrease
the ondemand safety gaurd band from 80-10 to 95-3.

Results in more aggressive power savings.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq_ondemand.c | 45 +++++++++++++++++++++++++++++-
 1 file changed, 44 insertions(+), 1 deletion(-)

diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index 3f898606e68c..39253d6d51ef 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -18,6 +18,9 @@
 #include <linux/jiffies.h>
 #include <linux/kernel_stat.h>
 #include <linux/mutex.h>
+#include <linux/hrtimer.h>
+#include <linux/tick.h>
+#include <linux/ktime.h>
 
 /*
  * dbs is used in this file as a shortform for demandbased switching
@@ -26,6 +29,8 @@
 
 #define DEF_FREQUENCY_DOWN_DIFFERENTIAL		(10)
 #define DEF_FREQUENCY_UP_THRESHOLD		(80)
+#define MICRO_FREQUENCY_DOWN_DIFFERENTIAL	(3)
+#define MICRO_FREQUENCY_UP_THRESHOLD		(95)
 #define MIN_FREQUENCY_UP_THRESHOLD		(11)
 #define MAX_FREQUENCY_UP_THRESHOLD		(100)
 
@@ -58,6 +63,7 @@ enum {DBS_NORMAL_SAMPLE, DBS_SUB_SAMPLE};
 struct cpu_dbs_info_s {
 	cputime64_t prev_cpu_idle;
 	cputime64_t prev_cpu_wall;
+	cputime64_t prev_cpu_nice;
 	struct cpufreq_policy *cur_policy;
  	struct delayed_work work;
 	struct cpufreq_frequency_table *freq_table;
@@ -97,7 +103,8 @@ static struct dbs_tuners {
 	.powersave_bias = 0,
 };
 
-static inline cputime64_t get_cpu_idle_time(unsigned int cpu, cputime64_t *wall)
+static inline cputime64_t get_cpu_idle_time_jiffy(unsigned int cpu,
+							cputime64_t *wall)
 {
 	cputime64_t idle_time;
 	cputime64_t cur_wall_time;
@@ -123,6 +130,33 @@ static inline cputime64_t get_cpu_idle_time(unsigned int cpu, cputime64_t *wall)
 	return idle_time;
 }
 
+static inline cputime64_t get_cpu_idle_time(unsigned int cpu, cputime64_t *wall)
+{
+	u64 idle_time = get_cpu_idle_time_us(cpu, wall);
+
+	if (idle_time == -1ULL)
+		return get_cpu_idle_time_jiffy(cpu, wall);
+
+	if (dbs_tuners_ins.ignore_nice) {
+		cputime64_t cur_nice;
+		unsigned long cur_nice_jiffies;
+		struct cpu_dbs_info_s *dbs_info;
+
+		dbs_info = &per_cpu(cpu_dbs_info, cpu);
+		cur_nice = cputime64_sub(kstat_cpu(cpu).cpustat.nice,
+					 dbs_info->prev_cpu_nice);
+		/*
+		 * Assumption: nice time between sampling periods will be
+		 * less than 2^32 jiffies for 32 bit sys
+		 */
+		cur_nice_jiffies = (unsigned long)
+					cputime64_to_jiffies64(cur_nice);
+		dbs_info->prev_cpu_nice = kstat_cpu(cpu).cpustat.nice;
+		return idle_time + jiffies_to_usecs(cur_nice_jiffies);
+	}
+	return idle_time;
+}
+
 /*
  * Find right freq to be set now with powersave_bias on.
  * Returns the freq_hi to be used right now and will set freq_hi_jiffies,
@@ -603,6 +637,15 @@ EXPORT_SYMBOL(cpufreq_gov_ondemand);
 static int __init cpufreq_gov_dbs_init(void)
 {
 	int err;
+	cputime64_t wall;
+	u64 idle_time = get_cpu_idle_time_us(smp_processor_id(), &wall);
+
+	if (idle_time != -1ULL) {
+		/* Idle micro accounting is supported. Use finer thresholds */
+		dbs_tuners_ins.up_threshold = MICRO_FREQUENCY_UP_THRESHOLD;
+		dbs_tuners_ins.down_differential =
+					MICRO_FREQUENCY_DOWN_DIFFERENTIAL;
+	}
 
 	kondemand_wq = create_workqueue("kondemand");
 	if (!kondemand_wq) {

From c4d14bc0bb5d13e316890651ae4518b764c3344c Mon Sep 17 00:00:00 2001
From: Sven Wegener <sven.wegener@stealer.net>
Date: Sat, 20 Sep 2008 16:50:08 +0200
Subject: [PATCH 13/14] [CPUFREQ] Don't export governors for default governor

We don't need to export the governors for use as the default governor,
because the default governor will be built-in anyway and we can access
the symbol directly.

This also fixes the following sparse warnings:

drivers/cpufreq/cpufreq_conservative.c:578:25: warning: symbol 'cpufreq_gov_conservative' was not declared. Should it be static?
drivers/cpufreq/cpufreq_ondemand.c:582:25: warning: symbol 'cpufreq_gov_ondemand' was not declared. Should it be static?
drivers/cpufreq/cpufreq_performance.c:39:25: warning: symbol 'cpufreq_gov_performance' was not declared. Should it be static?
drivers/cpufreq/cpufreq_powersave.c:38:25: warning: symbol 'cpufreq_gov_powersave' was not declared. Should it be static?
drivers/cpufreq/cpufreq_userspace.c:190:25: warning: symbol 'cpufreq_gov_userspace' was not declared. Should it be static?

Signed-off-by: Sven Wegener <sven.wegener@stealer.net>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq_conservative.c | 4 +++-
 drivers/cpufreq/cpufreq_ondemand.c     | 4 +++-
 drivers/cpufreq/cpufreq_performance.c  | 4 +++-
 drivers/cpufreq/cpufreq_powersave.c    | 4 +++-
 drivers/cpufreq/cpufreq_userspace.c    | 4 +++-
 5 files changed, 15 insertions(+), 5 deletions(-)

diff --git a/drivers/cpufreq/cpufreq_conservative.c b/drivers/cpufreq/cpufreq_conservative.c
index 4ee53a4c6d2c..e2657837d954 100644
--- a/drivers/cpufreq/cpufreq_conservative.c
+++ b/drivers/cpufreq/cpufreq_conservative.c
@@ -576,13 +576,15 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy,
 	return 0;
 }
 
+#ifndef CONFIG_CPU_FREQ_DEFAULT_GOV_CONSERVATIVE
+static
+#endif
 struct cpufreq_governor cpufreq_gov_conservative = {
 	.name			= "conservative",
 	.governor		= cpufreq_governor_dbs,
 	.max_transition_latency	= TRANSITION_LATENCY_LIMIT,
 	.owner			= THIS_MODULE,
 };
-EXPORT_SYMBOL(cpufreq_gov_conservative);
 
 static int __init cpufreq_gov_dbs_init(void)
 {
diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index 39253d6d51ef..bd069349bcae 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -626,13 +626,15 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy,
 	return 0;
 }
 
+#ifndef CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND
+static
+#endif
 struct cpufreq_governor cpufreq_gov_ondemand = {
 	.name			= "ondemand",
 	.governor		= cpufreq_governor_dbs,
 	.max_transition_latency = TRANSITION_LATENCY_LIMIT,
 	.owner			= THIS_MODULE,
 };
-EXPORT_SYMBOL(cpufreq_gov_ondemand);
 
 static int __init cpufreq_gov_dbs_init(void)
 {
diff --git a/drivers/cpufreq/cpufreq_performance.c b/drivers/cpufreq/cpufreq_performance.c
index e8e1451ef1c1..7e2e515087f8 100644
--- a/drivers/cpufreq/cpufreq_performance.c
+++ b/drivers/cpufreq/cpufreq_performance.c
@@ -36,12 +36,14 @@ static int cpufreq_governor_performance(struct cpufreq_policy *policy,
 	return 0;
 }
 
+#ifdef CONFIG_CPU_FREQ_GOV_PERFORMANCE_MODULE
+static
+#endif
 struct cpufreq_governor cpufreq_gov_performance = {
 	.name		= "performance",
 	.governor	= cpufreq_governor_performance,
 	.owner		= THIS_MODULE,
 };
-EXPORT_SYMBOL(cpufreq_gov_performance);
 
 
 static int __init cpufreq_gov_performance_init(void)
diff --git a/drivers/cpufreq/cpufreq_powersave.c b/drivers/cpufreq/cpufreq_powersave.c
index 88d2f44fba48..e6db5faf3eb1 100644
--- a/drivers/cpufreq/cpufreq_powersave.c
+++ b/drivers/cpufreq/cpufreq_powersave.c
@@ -35,12 +35,14 @@ static int cpufreq_governor_powersave(struct cpufreq_policy *policy,
 	return 0;
 }
 
+#ifndef CONFIG_CPU_FREQ_DEFAULT_GOV_POWERSAVE
+static
+#endif
 struct cpufreq_governor cpufreq_gov_powersave = {
 	.name		= "powersave",
 	.governor	= cpufreq_governor_powersave,
 	.owner		= THIS_MODULE,
 };
-EXPORT_SYMBOL(cpufreq_gov_powersave);
 
 static int __init cpufreq_gov_powersave_init(void)
 {
diff --git a/drivers/cpufreq/cpufreq_userspace.c b/drivers/cpufreq/cpufreq_userspace.c
index 32244aa7cc0c..1442bbada053 100644
--- a/drivers/cpufreq/cpufreq_userspace.c
+++ b/drivers/cpufreq/cpufreq_userspace.c
@@ -187,6 +187,9 @@ static int cpufreq_governor_userspace(struct cpufreq_policy *policy,
 }
 
 
+#ifndef CONFIG_CPU_FREQ_DEFAULT_GOV_USERSPACE
+static
+#endif
 struct cpufreq_governor cpufreq_gov_userspace = {
 	.name		= "userspace",
 	.governor	= cpufreq_governor_userspace,
@@ -194,7 +197,6 @@ struct cpufreq_governor cpufreq_gov_userspace = {
 	.show_setspeed	= show_speed,
 	.owner		= THIS_MODULE,
 };
-EXPORT_SYMBOL(cpufreq_gov_userspace);
 
 static int __init cpufreq_gov_userspace_init(void)
 {

From 4f6e6b9f97b0ce98a8d1da65adbaf743bd0486a9 Mon Sep 17 00:00:00 2001
From: Andrea Righi <righi.andrea@gmail.com>
Date: Thu, 18 Sep 2008 10:43:40 +0000
Subject: [PATCH 14/14] [CPUFREQ] Fix BUG: using smp_processor_id() in
 preemptible code

Use get_cpu()/put_cpu() in cpufreq_ondemand init routine, instead of
smp_processor_id() to avoid the following BUG:

[   35.313118] BUG: using smp_processor_id() in preemptible [00000000] code=: modprobe/4952
[   35.313132] caller is cpufreq_gov_dbs_init+0xa/0x8f [cpufreq_ondemand]
[   35.313140] Pid: 4952, comm: modprobe Not tainted 2.6.27-rc5-mm1 #23
[   35.313145] Call Trace:
[   35.313158]  [<ffffffff80361ff7>] debug_smp_processor_id+0xd7/0xe0
[   35.313167]  [<ffffffffa010800a>] cpufreq_gov_dbs_init+0xa/0x8f [cpufreq_ondemand]
[   35.313176]  [<ffffffff8020903b>] _stext+0x3b/0x160
[   35.313185]  [<ffffffff804768c5>] __mutex_unlock_slowpath+0xe5/0x190
[   35.313195]  [<ffffffff8026236a>] trace_hardirqs_on_caller+0xca/0x140
[   35.313205]  [<ffffffff8026ef4c>] sys_init_module+0xdc/0x210
[   35.313212]  [<ffffffff8020b7cb>] system_call_fastpath+0x16/0x1b

Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq_ondemand.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index bd069349bcae..2ab3c12b88af 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -640,8 +640,11 @@ static int __init cpufreq_gov_dbs_init(void)
 {
 	int err;
 	cputime64_t wall;
-	u64 idle_time = get_cpu_idle_time_us(smp_processor_id(), &wall);
+	u64 idle_time;
+	int cpu = get_cpu();
 
+	idle_time = get_cpu_idle_time_us(cpu, &wall);
+	put_cpu();
 	if (idle_time != -1ULL) {
 		/* Idle micro accounting is supported. Use finer thresholds */
 		dbs_tuners_ins.up_threshold = MICRO_FREQUENCY_UP_THRESHOLD;