[PATCH 0/3 V3] Introduce strtobool (previously usr_strtobool).

[PATCH 0/3 V3] Introduce strtobool (previously usr_strtobool).

[Jonathan Cameron]

Hi All,

Third pass at introducing a new function to unify code that
is attempting to get a Boolean value from user input strings.

Rusty made the point that the usr_ in V2 implied user pointers, which
these aren't.  Hence the usr_ prefix has been dropped.  The naming
is still distinct enough from kstrtox to avoid any implication that
this function has the same tight parameter passing that those
functions have.

Use cases are the two converted here and numerous Boolean attributes in
sysfs. It is for the sysfs use cases that I'm interested, but Greg KH
pointed out a good start would be to unify these two, so that is
what we have here.

Jonathan Cameron (3):
  Add a strtobool function matching semantics of existing in kernel
    equivalents
  debugfs: move to new strtobool
  params.c: Use new strtobool function to process boolean inputs

 fs/debugfs/file.c      |   17 ++++-------------
 include/linux/string.h |    1 +
 kernel/params.c        |   14 ++++----------
 lib/string.c           |   29 +++++++++++++++++++++++++++++
 4 files changed, 38 insertions(+), 23 deletions(-)

-- 
1.7.3.4

[PATCH 1/3] Add a strtobool function matching semantics of existing in kernel equivalents

[Jonathan Cameron]

This is a rename of the usr_strtobool proposal, which was a renamed,
relocated and fixed version of previous kstrtobool RFC

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
---
 include/linux/string.h |    1 +
 lib/string.c           |   29 +++++++++++++++++++++++++++++
 2 files changed, 30 insertions(+), 0 deletions(-)

diff --git a/include/linux/string.h b/include/linux/string.h
index a716ee2..a176db2 100644
--- a/include/linux/string.h
+++ b/include/linux/string.h
@@ -123,6 +123,7 @@ extern char **argv_split(gfp_t gfp, const char *str, int *argcp);
 extern void argv_free(char **argv);
 
 extern bool sysfs_streq(const char *s1, const char *s2);
+extern int strtobool(const char *s, bool *res);
 
 #ifdef CONFIG_BINARY_PRINTF
 int vbin_printf(u32 *bin_buf, size_t size, const char *fmt, va_list args);
diff --git a/lib/string.c b/lib/string.c
index f71bead..01fad9b 100644
--- a/lib/string.c
+++ b/lib/string.c
@@ -535,6 +535,35 @@ bool sysfs_streq(const char *s1, const char *s2)
 }
 EXPORT_SYMBOL(sysfs_streq);
 
+/**
+ * strtobool - convert common user inputs into boolean values
+ * @s: input string
+ * @res: result
+ *
+ * This routine returns 0 iff the first character is one of 'Yy1Nn0'.
+ * Otherwise it will return -EINVAL.  Value pointed to by res is
+ * updated upon finding a match.
+ */
+int strtobool(const char *s, bool *res)
+{
+	switch (s[0]) {
+	case 'y':
+	case 'Y':
+	case '1':
+		*res = true;
+		break;
+	case 'n':
+	case 'N':
+	case '0':
+		*res = false;
+		break;
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(strtobool);
+
 #ifndef __HAVE_ARCH_MEMSET
 /**
  * memset - Fill a region of memory with the given value
-- 
1.7.3.4

[PATCH 2/3] debugfs: move to new strtobool

[Jonathan Cameron]

No functional changes requires that we eat errors from strtobool.
If people want to not do this, then it should be fixed at a later date.

V2: Simplification suggested by Rusty Russell removes the need for
additional variable ret.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
---
 fs/debugfs/file.c |   17 ++++-------------
 1 files changed, 4 insertions(+), 13 deletions(-)

diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c
index 89d394d..568304d 100644
--- a/fs/debugfs/file.c
+++ b/fs/debugfs/file.c
@@ -429,25 +429,16 @@ static ssize_t write_file_bool(struct file *file, const char __user *user_buf,
 {
 	char buf[32];
 	int buf_size;
+	bool bv;
 	u32 *val = file->private_data;
 
 	buf_size = min(count, (sizeof(buf)-1));
 	if (copy_from_user(buf, user_buf, buf_size))
 		return -EFAULT;
 
-	switch (buf[0]) {
-	case 'y':
-	case 'Y':
-	case '1':
-		*val = 1;
-		break;
-	case 'n':
-	case 'N':
-	case '0':
-		*val = 0;
-		break;
-	}
-	
+	if (strtobool(buf, &bv) == 0)
+		*val = bv;
+
 	return count;
 }
 
-- 
1.7.3.4

[PATCH 3/3] params.c: Use new strtobool function to process boolean inputs

[Jonathan Cameron]

No functional changes.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
---
 kernel/params.c |   14 ++++----------
 1 files changed, 4 insertions(+), 10 deletions(-)

diff --git a/kernel/params.c b/kernel/params.c
index 7ab388a..6888761 100644
--- a/kernel/params.c
+++ b/kernel/params.c
@@ -297,21 +297,15 @@ EXPORT_SYMBOL(param_ops_charp);
 int param_set_bool(const char *val, const struct kernel_param *kp)
 {
 	bool v;
+	int ret;
 
 	/* No equals means "set"... */
 	if (!val) val = "1";
 
 	/* One of =[yYnN01] */
-	switch (val[0]) {
-	case 'y': case 'Y': case '1':
-		v = true;
-		break;
-	case 'n': case 'N': case '0':
-		v = false;
-		break;
-	default:
-		return -EINVAL;
-	}
+	ret = strtobool(val, &v);
+	if (ret)
+		return ret;
 
 	if (kp->flags & KPARAM_ISBOOL)
 		*(bool *)kp->arg = v;
-- 
1.7.3.4

Re: [PATCH 1/3] Add a strtobool function matching semantics of existing in kernel equivalents

[Ryan Mallon]

On 04/19/2011 11:43 PM, Jonathan Cameron wrote:
> This is a rename of the usr_strtobool proposal, which was a renamed,
> relocated and fixed version of previous kstrtobool RFC
> 
> Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
> ---
>  include/linux/string.h |    1 +
>  lib/string.c           |   29 +++++++++++++++++++++++++++++
>  2 files changed, 30 insertions(+), 0 deletions(-)
> 
> diff --git a/include/linux/string.h b/include/linux/string.h
> index a716ee2..a176db2 100644
> --- a/include/linux/string.h
> +++ b/include/linux/string.h
> @@ -123,6 +123,7 @@ extern char **argv_split(gfp_t gfp, const char *str, int *argcp);
>  extern void argv_free(char **argv);
>  
>  extern bool sysfs_streq(const char *s1, const char *s2);
> +extern int strtobool(const char *s, bool *res);
>  
>  #ifdef CONFIG_BINARY_PRINTF
>  int vbin_printf(u32 *bin_buf, size_t size, const char *fmt, va_list args);
> diff --git a/lib/string.c b/lib/string.c
> index f71bead..01fad9b 100644
> --- a/lib/string.c
> +++ b/lib/string.c
> @@ -535,6 +535,35 @@ bool sysfs_streq(const char *s1, const char *s2)
>  }
>  EXPORT_SYMBOL(sysfs_streq);
>  
> +/**
> + * strtobool - convert common user inputs into boolean values
> + * @s: input string
> + * @res: result
> + *
> + * This routine returns 0 iff the first character is one of 'Yy1Nn0'.
> + * Otherwise it will return -EINVAL.  Value pointed to by res is
> + * updated upon finding a match.
> + */
> +int strtobool(const char *s, bool *res)
> +{

Might be worth doing:

	if (!s)
		return -EINVAL;

here to avoid blowing up if we get passed a NULL string.

~Ryan

> +	switch (s[0]) {
> +	case 'y':
> +	case 'Y':
> +	case '1':
> +		*res = true;
> +		break;
> +	case 'n':
> +	case 'N':
> +	case '0':
> +		*res = false;
> +		break;
> +	default:
> +		return -EINVAL;
> +	}
> +	return 0;
> +}
> +EXPORT_SYMBOL(strtobool);
> +
>  #ifndef __HAVE_ARCH_MEMSET
>  /**
>   * memset - Fill a region of memory with the given value


-- 
Bluewater Systems Ltd - ARM Technology Solution Centre

Ryan Mallon         		5 Amuri Park, 404 Barbadoes St
ryan@bluewatersys.com         	PO Box 13 889, Christchurch 8013
http://www.bluewatersys.com	New Zealand
Phone: +64 3 3779127		Freecall: Australia 1800 148 751
Fax:   +64 3 3779135			  USA 1800 261 2934

Re: [PATCH 2/3] debugfs: move to new strtobool

[Ryan Mallon]

On 04/19/2011 11:43 PM, Jonathan Cameron wrote:
> No functional changes requires that we eat errors from strtobool.
> If people want to not do this, then it should be fixed at a later date.

May as well fix it now or it will get forgotten about. A second patch on
top of this can fix the bug.

~Ryan

> V2: Simplification suggested by Rusty Russell removes the need for
> additional variable ret.
> 
> Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
> ---
>  fs/debugfs/file.c |   17 ++++-------------
>  1 files changed, 4 insertions(+), 13 deletions(-)
> 
> diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c
> index 89d394d..568304d 100644
> --- a/fs/debugfs/file.c
> +++ b/fs/debugfs/file.c
> @@ -429,25 +429,16 @@ static ssize_t write_file_bool(struct file *file, const char __user *user_buf,
>  {
>  	char buf[32];
>  	int buf_size;
> +	bool bv;
>  	u32 *val = file->private_data;
>  
>  	buf_size = min(count, (sizeof(buf)-1));
>  	if (copy_from_user(buf, user_buf, buf_size))
>  		return -EFAULT;
>  
> -	switch (buf[0]) {
> -	case 'y':
> -	case 'Y':
> -	case '1':
> -		*val = 1;
> -		break;
> -	case 'n':
> -	case 'N':
> -	case '0':
> -		*val = 0;
> -		break;
> -	}
> -	
> +	if (strtobool(buf, &bv) == 0)
> +		*val = bv;
> +
>  	return count;
>  }
>  


-- 
Bluewater Systems Ltd - ARM Technology Solution Centre

Ryan Mallon         		5 Amuri Park, 404 Barbadoes St
ryan@bluewatersys.com         	PO Box 13 889, Christchurch 8013
http://www.bluewatersys.com	New Zealand
Phone: +64 3 3779127		Freecall: Australia 1800 148 751
Fax:   +64 3 3779135			  USA 1800 261 2934

Re: [PATCH 2/3] debugfs: move to new strtobool

[Jonathan Cameron]

On 04/19/11 21:30, Ryan Mallon wrote:
> On 04/19/2011 11:43 PM, Jonathan Cameron wrote:
>> No functional changes requires that we eat errors from strtobool.
>> If people want to not do this, then it should be fixed at a later date.
> 
> May as well fix it now or it will get forgotten about. A second patch on
> top of this can fix the bug.
What worries me about this 'fix' is that it may well break some 'interesting'
bit of userspace code.  It would count as a userspace api change, be it
a fairly minor one.

Greg, do you thing it's worth returning an error on a non bool value?
> 
> ~Ryan
> 
>> V2: Simplification suggested by Rusty Russell removes the need for
>> additional variable ret.
>>
>> Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
>> ---
>>  fs/debugfs/file.c |   17 ++++-------------
>>  1 files changed, 4 insertions(+), 13 deletions(-)
>>
>> diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c
>> index 89d394d..568304d 100644
>> --- a/fs/debugfs/file.c
>> +++ b/fs/debugfs/file.c
>> @@ -429,25 +429,16 @@ static ssize_t write_file_bool(struct file *file, const char __user *user_buf,
>>  {
>>  	char buf[32];
>>  	int buf_size;
>> +	bool bv;
>>  	u32 *val = file->private_data;
>>  
>>  	buf_size = min(count, (sizeof(buf)-1));
>>  	if (copy_from_user(buf, user_buf, buf_size))
>>  		return -EFAULT;
>>  
>> -	switch (buf[0]) {
>> -	case 'y':
>> -	case 'Y':
>> -	case '1':
>> -		*val = 1;
>> -		break;
>> -	case 'n':
>> -	case 'N':
>> -	case '0':
>> -		*val = 0;
>> -		break;
>> -	}
>> -	
>> +	if (strtobool(buf, &bv) == 0)
>> +		*val = bv;
>> +
>>  	return count;
>>  }
>>  
> 
> 

Re: [PATCH 2/3] debugfs: move to new strtobool

[Greg KH]

On Wed, Apr 20, 2011 at 10:33:22AM +0100, Jonathan Cameron wrote:
> On 04/19/11 21:30, Ryan Mallon wrote:
> > On 04/19/2011 11:43 PM, Jonathan Cameron wrote:
> >> No functional changes requires that we eat errors from strtobool.
> >> If people want to not do this, then it should be fixed at a later date.
> > 
> > May as well fix it now or it will get forgotten about. A second patch on
> > top of this can fix the bug.
> What worries me about this 'fix' is that it may well break some 'interesting'
> bit of userspace code.  It would count as a userspace api change, be it
> a fairly minor one.
> 
> Greg, do you thing it's worth returning an error on a non bool value?

No one has ever complained about it, so I doubt it's a big issue.

So I'd say to just leave it as-is for now.

thanks,

greg k-h

Re: [PATCH 2/3] debugfs: move to new strtobool

[Ryan Mallon]

On 04/26/2011 10:54 AM, Greg KH wrote:
> On Wed, Apr 20, 2011 at 10:33:22AM +0100, Jonathan Cameron wrote:
>> On 04/19/11 21:30, Ryan Mallon wrote:
>>> On 04/19/2011 11:43 PM, Jonathan Cameron wrote:
>>>> No functional changes requires that we eat errors from strtobool.
>>>> If people want to not do this, then it should be fixed at a later date.
>>>
>>> May as well fix it now or it will get forgotten about. A second patch on
>>> top of this can fix the bug.
>> What worries me about this 'fix' is that it may well break some 'interesting'
>> bit of userspace code.  It would count as a userspace api change, be it
>> a fairly minor one.
>>
>> Greg, do you thing it's worth returning an error on a non bool value?
> 
> No one has ever complained about it, so I doubt it's a big issue.

Probably because nobody checks for errors in user space :-p

~Ryan

-- 
Bluewater Systems Ltd - ARM Technology Solution Centre

Ryan Mallon         		5 Amuri Park, 404 Barbadoes St
ryan@bluewatersys.com         	PO Box 13 889, Christchurch 8013
http://www.bluewatersys.com	New Zealand
Phone: +64 3 3779127		Freecall: Australia 1800 148 751
Fax:   +64 3 3779135			  USA 1800 261 2934

[PATCH] mm/memcg: replace inexistence move_lock_page_cgroup() by move_lock_mem_cgroup() in comment

[Wanpeng Li]

From: Wanpeng Li <liwp@linux.vnet.ibm.com>

Signed-off-by: Wanpeng Li <liwp.linux@gmail.com>
---
 mm/memcontrol.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 3d318f6..63e36e7 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1899,7 +1899,7 @@ again:
 		return;
 	/*
 	 * If this memory cgroup is not under account moving, we don't
-	 * need to take move_lock_page_cgroup(). Because we already hold
+	 * need to take move_lock_mem_cgroup(). Because we already hold
 	 * rcu_read_lock(), any calls to move_account will be delayed until
 	 * rcu_read_unlock() if mem_cgroup_stolen() == true.
 	 */
@@ -1921,7 +1921,7 @@ void __mem_cgroup_end_update_page_stat(struct page *page, unsigned long *flags)
 	/*
 	 * It's guaranteed that pc->mem_cgroup never changes while
 	 * lock is held because a routine modifies pc->mem_cgroup
-	 * should take move_lock_page_cgroup().
+	 * should take move_lock_mem_cgroup().
 	 */
 	move_unlock_mem_cgroup(pc->mem_cgroup, flags);
 }
-- 
1.7.5.4

Re: [PATCH] mm/memcg: replace inexistence move_lock_page_cgroup() by move_lock_mem_cgroup() in comment

[Kamezawa Hiroyuki]

(2012/07/05 15:28), Wanpeng Li wrote:
> From: Wanpeng Li <liwp@linux.vnet.ibm.com>
> 
> Signed-off-by: Wanpeng Li <liwp.linux@gmail.com>

Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>


> ---
>   mm/memcontrol.c |    4 ++--
>   1 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 3d318f6..63e36e7 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -1899,7 +1899,7 @@ again:
>   		return;
>   	/*
>   	 * If this memory cgroup is not under account moving, we don't
> -	 * need to take move_lock_page_cgroup(). Because we already hold
> +	 * need to take move_lock_mem_cgroup(). Because we already hold
>   	 * rcu_read_lock(), any calls to move_account will be delayed until
>   	 * rcu_read_unlock() if mem_cgroup_stolen() == true.
>   	 */
> @@ -1921,7 +1921,7 @@ void __mem_cgroup_end_update_page_stat(struct page *page, unsigned long *flags)
>   	/*
>   	 * It's guaranteed that pc->mem_cgroup never changes while
>   	 * lock is held because a routine modifies pc->mem_cgroup
> -	 * should take move_lock_page_cgroup().
> +	 * should take move_lock_mem_cgroup().
>   	 */
>   	move_unlock_mem_cgroup(pc->mem_cgroup, flags);
>   }
> 

[PATCH RFC] mm/memcg: calculate max hierarchy limit number instead of min

[Wanpeng Li]

From: Wanpeng Li <liwp@linux.vnet.ibm.com>

Since hierachical_memory_limit shows "of bytes of memory limit with
regard to hierarchy under which the memory cgroup is", the count should
calculate max hierarchy limit when use_hierarchy in order to show hierarchy
subtree limit. hierachical_memsw_limit is the same case.

Signed-off-by: Wanpeng Li <liwp.linux@gmail.com>
---
 mm/memcontrol.c |   14 +++++++-------
 1 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 69a7d45..6392c0a 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3929,10 +3929,10 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
 		unsigned long long *mem_limit, unsigned long long *memsw_limit)
 {
 	struct cgroup *cgroup;
-	unsigned long long min_limit, min_memsw_limit, tmp;
+	unsigned long long max_limit, max_memsw_limit, tmp;
 
-	min_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
-	min_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
+	max_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
+	max_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
 	cgroup = memcg->css.cgroup;
 	if (!memcg->use_hierarchy)
 		goto out;
@@ -3943,13 +3943,13 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
 		if (!memcg->use_hierarchy)
 			break;
 		tmp = res_counter_read_u64(&memcg->res, RES_LIMIT);
-		min_limit = min(min_limit, tmp);
+		max_limit = max(max_limit, tmp);
 		tmp = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
-		min_memsw_limit = min(min_memsw_limit, tmp);
+		max_memsw_limit = max(max_memsw_limit, tmp);
 	}
 out:
-	*mem_limit = min_limit;
-	*memsw_limit = min_memsw_limit;
+	*mem_limit = max_limit;
+	*memsw_limit = max_memsw_limit;
 }
 
 static int mem_cgroup_reset(struct cgroup *cont, unsigned int event)
-- 
1.7.5.4

Re: [PATCH RFC] mm/memcg: calculate max hierarchy limit number instead of min

[Michal Hocko]

On Wed 11-07-12 21:24:41, Wanpeng Li wrote:
> From: Wanpeng Li <liwp@linux.vnet.ibm.com>
> 
> Since hierachical_memory_limit shows "of bytes of memory limit with
> regard to hierarchy under which the memory cgroup is", the count should
> calculate max hierarchy limit when use_hierarchy in order to show hierarchy
> subtree limit. hierachical_memsw_limit is the same case.

No the patch is wrong. The hierarchical limit says when we start
reclaiming in the hierarchy and that one is triggered on smallest limit
up the way to the hierarchy root.

What are you trying to accomplish here?

> Signed-off-by: Wanpeng Li <liwp.linux@gmail.com>
> ---
>  mm/memcontrol.c |   14 +++++++-------
>  1 files changed, 7 insertions(+), 7 deletions(-)
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 69a7d45..6392c0a 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -3929,10 +3929,10 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
>  		unsigned long long *mem_limit, unsigned long long *memsw_limit)
>  {
>  	struct cgroup *cgroup;
> -	unsigned long long min_limit, min_memsw_limit, tmp;
> +	unsigned long long max_limit, max_memsw_limit, tmp;
>  
> -	min_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
> -	min_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
> +	max_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
> +	max_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
>  	cgroup = memcg->css.cgroup;
>  	if (!memcg->use_hierarchy)
>  		goto out;
> @@ -3943,13 +3943,13 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
>  		if (!memcg->use_hierarchy)
>  			break;
>  		tmp = res_counter_read_u64(&memcg->res, RES_LIMIT);
> -		min_limit = min(min_limit, tmp);
> +		max_limit = max(max_limit, tmp);
>  		tmp = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
> -		min_memsw_limit = min(min_memsw_limit, tmp);
> +		max_memsw_limit = max(max_memsw_limit, tmp);
>  	}
>  out:
> -	*mem_limit = min_limit;
> -	*memsw_limit = min_memsw_limit;
> +	*mem_limit = max_limit;
> +	*memsw_limit = max_memsw_limit;
>  }
>  
>  static int mem_cgroup_reset(struct cgroup *cont, unsigned int event)
> -- 
> 1.7.5.4
> 

-- 
Michal Hocko
SUSE Labs
SUSE LINUX s.r.o.
Lihovarska 1060/12
190 00 Praha 9    
Czech Republic

Re: [PATCH RFC] mm/memcg: calculate max hierarchy limit number instead of min

[Wanpeng Li]

On Wed, Jul 11, 2012 at 03:47:57PM +0200, Michal Hocko wrote:
>On Wed 11-07-12 21:24:41, Wanpeng Li wrote:
>> From: Wanpeng Li <liwp@linux.vnet.ibm.com>
>> 
>> Since hierachical_memory_limit shows "of bytes of memory limit with
>> regard to hierarchy under which the memory cgroup is", the count should
>> calculate max hierarchy limit when use_hierarchy in order to show hierarchy
>> subtree limit. hierachical_memsw_limit is the same case.
>
>No the patch is wrong. The hierarchical limit says when we start
>reclaiming in the hierarchy and that one is triggered on smallest limit
>up the way to the hierarchy root.

I see function mem_cgroup_hierachy_reclaim is removal, and hierarchical
reclaim is still work? Could you explain me how it works in details, 
thank you for your time Michal.

Thanks & Best Regards,
Wanpeng Li
>
>What are you trying to accomplish here?
>
>> Signed-off-by: Wanpeng Li <liwp.linux@gmail.com>
>> ---
>>  mm/memcontrol.c |   14 +++++++-------
>>  1 files changed, 7 insertions(+), 7 deletions(-)
>> 
>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>> index 69a7d45..6392c0a 100644
>> --- a/mm/memcontrol.c
>> +++ b/mm/memcontrol.c
>> @@ -3929,10 +3929,10 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
>>  		unsigned long long *mem_limit, unsigned long long *memsw_limit)
>>  {
>>  	struct cgroup *cgroup;
>> -	unsigned long long min_limit, min_memsw_limit, tmp;
>> +	unsigned long long max_limit, max_memsw_limit, tmp;
>>  
>> -	min_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
>> -	min_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
>> +	max_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
>> +	max_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
>>  	cgroup = memcg->css.cgroup;
>>  	if (!memcg->use_hierarchy)
>>  		goto out;
>> @@ -3943,13 +3943,13 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
>>  		if (!memcg->use_hierarchy)
>>  			break;
>>  		tmp = res_counter_read_u64(&memcg->res, RES_LIMIT);
>> -		min_limit = min(min_limit, tmp);
>> +		max_limit = max(max_limit, tmp);
>>  		tmp = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
>> -		min_memsw_limit = min(min_memsw_limit, tmp);
>> +		max_memsw_limit = max(max_memsw_limit, tmp);
>>  	}
>>  out:
>> -	*mem_limit = min_limit;
>> -	*memsw_limit = min_memsw_limit;
>> +	*mem_limit = max_limit;
>> +	*memsw_limit = max_memsw_limit;
>>  }
>>  
>>  static int mem_cgroup_reset(struct cgroup *cont, unsigned int event)
>> -- 
>> 1.7.5.4
>> 
>
>-- 
>Michal Hocko
>SUSE Labs
>SUSE LINUX s.r.o.
>Lihovarska 1060/12
>190 00 Praha 9    
>Czech Republic

Re: [PATCH RFC] mm/memcg: calculate max hierarchy limit number instead of min

[Michal Hocko]

On Thu 12-07-12 17:32:11, Wanpeng Li wrote:
> On Wed, Jul 11, 2012 at 03:47:57PM +0200, Michal Hocko wrote:
> >On Wed 11-07-12 21:24:41, Wanpeng Li wrote:
> >> From: Wanpeng Li <liwp@linux.vnet.ibm.com>
> >> 
> >> Since hierachical_memory_limit shows "of bytes of memory limit with
> >> regard to hierarchy under which the memory cgroup is", the count should
> >> calculate max hierarchy limit when use_hierarchy in order to show hierarchy
> >> subtree limit. hierachical_memsw_limit is the same case.
> >
> >No the patch is wrong. The hierarchical limit says when we start
> >reclaiming in the hierarchy and that one is triggered on smallest limit
> >up the way to the hierarchy root.
> 
> I see function mem_cgroup_hierachy_reclaim is removal, and hierarchical
> reclaim is still work? 

I am not aware of that.

> Could you explain me how it works in details, thank you for your time
> Michal.

I am not sure I understand what you are interested in. 

> 
> Thanks & Best Regards,
> Wanpeng Li
> >
> >What are you trying to accomplish here?
> >
> >> Signed-off-by: Wanpeng Li <liwp.linux@gmail.com>
> >> ---
> >>  mm/memcontrol.c |   14 +++++++-------
> >>  1 files changed, 7 insertions(+), 7 deletions(-)
> >> 
> >> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> >> index 69a7d45..6392c0a 100644
> >> --- a/mm/memcontrol.c
> >> +++ b/mm/memcontrol.c
> >> @@ -3929,10 +3929,10 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
> >>  		unsigned long long *mem_limit, unsigned long long *memsw_limit)
> >>  {
> >>  	struct cgroup *cgroup;
> >> -	unsigned long long min_limit, min_memsw_limit, tmp;
> >> +	unsigned long long max_limit, max_memsw_limit, tmp;
> >>  
> >> -	min_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
> >> -	min_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
> >> +	max_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
> >> +	max_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
> >>  	cgroup = memcg->css.cgroup;
> >>  	if (!memcg->use_hierarchy)
> >>  		goto out;
> >> @@ -3943,13 +3943,13 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
> >>  		if (!memcg->use_hierarchy)
> >>  			break;
> >>  		tmp = res_counter_read_u64(&memcg->res, RES_LIMIT);
> >> -		min_limit = min(min_limit, tmp);
> >> +		max_limit = max(max_limit, tmp);
> >>  		tmp = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
> >> -		min_memsw_limit = min(min_memsw_limit, tmp);
> >> +		max_memsw_limit = max(max_memsw_limit, tmp);
> >>  	}
> >>  out:
> >> -	*mem_limit = min_limit;
> >> -	*memsw_limit = min_memsw_limit;
> >> +	*mem_limit = max_limit;
> >> +	*memsw_limit = max_memsw_limit;
> >>  }
> >>  
> >>  static int mem_cgroup_reset(struct cgroup *cont, unsigned int event)
> >> -- 
> >> 1.7.5.4
> >> 
> >
> >-- 
> >Michal Hocko
> >SUSE Labs
> >SUSE LINUX s.r.o.
> >Lihovarska 1060/12
> >190 00 Praha 9    
> >Czech Republic

-- 
Michal Hocko
SUSE Labs
SUSE LINUX s.r.o.
Lihovarska 1060/12
190 00 Praha 9    
Czech Republic

[PATCH] mm/memcg: wrap mem_cgroup_from_css function

[Wanpeng Li]

wrap mem_cgroup_from_css function to clarify get mem cgroup
from cgroup_subsys_state.

Signed-off-by: Wanpeng Li <liwanp@linux.vnet.ibm.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Gavin Shan <shangw@linux.vnet.ibm.com>
Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com>
Cc: linux-kernel@vger.kernel.org
---
 mm/memcontrol.c |   14 ++++++++++----
 1 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 58a08fc..20f6a15 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -396,6 +396,12 @@ static void mem_cgroup_put(struct mem_cgroup *memcg);
 #include <net/sock.h>
 #include <net/ip.h>
 
+static inline
+struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *s)
+{
+	return container_of(s, struct mem_cgroup, css);
+}
+
 static bool mem_cgroup_is_root(struct mem_cgroup *memcg);
 void sock_update_memcg(struct sock *sk)
 {
@@ -820,7 +826,7 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page)
 
 struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont)
 {
-	return container_of(cgroup_subsys_state(cont,
+	return mem_cgroup_from_css(cgroup_subsys_state(cont,
 				mem_cgroup_subsys_id), struct mem_cgroup,
 				css);
 }
@@ -835,7 +841,7 @@ struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p)
 	if (unlikely(!p))
 		return NULL;
 
-	return container_of(task_subsys_state(p, mem_cgroup_subsys_id),
+	return mem_cgroup_from_css(task_subsys_state(p, mem_cgroup_subsys_id),
 				struct mem_cgroup, css);
 }
 
@@ -922,7 +928,7 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root,
 		css = css_get_next(&mem_cgroup_subsys, id + 1, &root->css, &id);
 		if (css) {
 			if (css == &root->css || css_tryget(css))
-				memcg = container_of(css,
+				memcg = mem_cgroup_from_css(css,
 						     struct mem_cgroup, css);
 		} else
 			id = 0;
@@ -2406,7 +2412,7 @@ static struct mem_cgroup *mem_cgroup_lookup(unsigned short id)
 	css = css_lookup(&mem_cgroup_subsys, id);
 	if (!css)
 		return NULL;
-	return container_of(css, struct mem_cgroup, css);
+	return mem_cgroup_from_css(css, struct mem_cgroup, css);
 }
 
 struct mem_cgroup *try_get_mem_cgroup_from_page(struct page *page)
-- 
1.7.5.4

Re: [PATCH] mm/memcg: wrap mem_cgroup_from_css function

[Andrew Morton]

On Wed, 18 Jul 2012 11:05:30 +0800
Wanpeng Li <liwanp@linux.vnet.ibm.com> wrote:

> wrap mem_cgroup_from_css function to clarify get mem cgroup
> from cgroup_subsys_state.

This certainly adds clarity.

But it also adds a little more type-safety - these container_of() calls
can be invoked against *any* struct which has a field called "css". 
With your patch, we add a check that the code is indeed using a
cgroup_subsys_state*.  A small thing, but it's all good.


I changed the patch title to the more idiomatic "memcg: add
mem_cgroup_from_css() helper" and rewrote the changelog to

: Add a mem_cgroup_from_css() helper to replace open-coded invokations of
: container_of().  To clarify the code and to add a little more type safety.

> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -396,6 +396,12 @@ static void mem_cgroup_put(struct mem_cgroup *memcg);
>  #include <net/sock.h>
>  #include <net/ip.h>
>  
> +static inline
> +struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *s)
> +{
> +	return container_of(s, struct mem_cgroup, css);
> +}

And with great self-control, I avoided renaming this to
memcg_from_css().  Sigh.  I guess all that extra typing has cardio
benefits.

Re: [PATCH] mm/memcg: wrap mem_cgroup_from_css function

[Wanpeng Li]

On Wed, Jul 18, 2012 at 02:36:12PM -0700, Andrew Morton wrote:
>On Wed, 18 Jul 2012 11:05:30 +0800
>Wanpeng Li <liwanp@linux.vnet.ibm.com> wrote:
>
>> wrap mem_cgroup_from_css function to clarify get mem cgroup
>> from cgroup_subsys_state.
>
>This certainly adds clarity.
>
>But it also adds a little more type-safety - these container_of() calls
>can be invoked against *any* struct which has a field called "css". 
>With your patch, we add a check that the code is indeed using a
>cgroup_subsys_state*.  A small thing, but it's all good.
>
>
>I changed the patch title to the more idiomatic "memcg: add
>mem_cgroup_from_css() helper" and rewrote the changelog to
>
>: Add a mem_cgroup_from_css() helper to replace open-coded invokations of
>: container_of().  To clarify the code and to add a little more type safety.
>
>> --- a/mm/memcontrol.c
>> +++ b/mm/memcontrol.c
>> @@ -396,6 +396,12 @@ static void mem_cgroup_put(struct mem_cgroup *memcg);
>>  #include <net/sock.h>
>>  #include <net/ip.h>
>>  
>> +static inline
>> +struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *s)
>> +{
>> +	return container_of(s, struct mem_cgroup, css);
>> +}
>
>And with great self-control, I avoided renaming this to
>memcg_from_css().  Sigh.  I guess all that extra typing has cardio
>benefits.

Thank you for your time, Andrew. :-)

Thanks & Best Regards,
Wanpeng Li

Re: [PATCH RFC] mm/memcg: calculate max hierarchy limit number instead of min

[Kamezawa Hiroyuki]

(2012/07/11 22:24), Wanpeng Li wrote:
> From: Wanpeng Li <liwp@linux.vnet.ibm.com>
> 
> Since hierachical_memory_limit shows "of bytes of memory limit with
> regard to hierarchy under which the memory cgroup is", the count should
> calculate max hierarchy limit when use_hierarchy in order to show hierarchy
> subtree limit. hierachical_memsw_limit is the same case.
> 
> Signed-off-by: Wanpeng Li <liwp.linux@gmail.com>

Hm ? What is the hierarchical limit for 'C' in following tree ?

A  ---  limit=1G 
 \
  B --  limit=500M
   \
    C - unlimtied

Thanks,
-Kame


> ---
>   mm/memcontrol.c |   14 +++++++-------
>   1 files changed, 7 insertions(+), 7 deletions(-)
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 69a7d45..6392c0a 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -3929,10 +3929,10 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
>   		unsigned long long *mem_limit, unsigned long long *memsw_limit)
>   {
>   	struct cgroup *cgroup;
> -	unsigned long long min_limit, min_memsw_limit, tmp;
> +	unsigned long long max_limit, max_memsw_limit, tmp;
>   
> -	min_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
> -	min_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
> +	max_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
> +	max_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
>   	cgroup = memcg->css.cgroup;
>   	if (!memcg->use_hierarchy)
>   		goto out;
> @@ -3943,13 +3943,13 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
>   		if (!memcg->use_hierarchy)
>   			break;
>   		tmp = res_counter_read_u64(&memcg->res, RES_LIMIT);
> -		min_limit = min(min_limit, tmp);
> +		max_limit = max(max_limit, tmp);
>   		tmp = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
> -		min_memsw_limit = min(min_memsw_limit, tmp);
> +		max_memsw_limit = max(max_memsw_limit, tmp);
>   	}
>   out:
> -	*mem_limit = min_limit;
> -	*memsw_limit = min_memsw_limit;
> +	*mem_limit = max_limit;
> +	*memsw_limit = max_memsw_limit;
>   }
>   
>   static int mem_cgroup_reset(struct cgroup *cont, unsigned int event)
> 

Re: [PATCH RFC] mm/memcg: calculate max hierarchy limit number instead of min

[Wanpeng Li]

On Thu, Jul 19, 2012 at 03:07:20PM +0900, Kamezawa Hiroyuki wrote:
>(2012/07/11 22:24), Wanpeng Li wrote:
>> From: Wanpeng Li <liwp@linux.vnet.ibm.com>
>> 
>> Since hierachical_memory_limit shows "of bytes of memory limit with
>> regard to hierarchy under which the memory cgroup is", the count should
>> calculate max hierarchy limit when use_hierarchy in order to show hierarchy
>> subtree limit. hierachical_memsw_limit is the same case.
>> 
>> Signed-off-by: Wanpeng Li <liwp.linux@gmail.com>
>
>Hm ? What is the hierarchical limit for 'C' in following tree ?
>
>A  ---  limit=1G 
> \
>  B --  limit=500M
>   \
>    C - unlimtied
>
Hmm, thank you Kame. :-)

Regards,
Wanpeng Li 

>Thanks,
>-Kame
>
>
>> ---
>>   mm/memcontrol.c |   14 +++++++-------
>>   1 files changed, 7 insertions(+), 7 deletions(-)
>> 
>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>> index 69a7d45..6392c0a 100644
>> --- a/mm/memcontrol.c
>> +++ b/mm/memcontrol.c
>> @@ -3929,10 +3929,10 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
>>   		unsigned long long *mem_limit, unsigned long long *memsw_limit)
>>   {
>>   	struct cgroup *cgroup;
>> -	unsigned long long min_limit, min_memsw_limit, tmp;
>> +	unsigned long long max_limit, max_memsw_limit, tmp;
>>   
>> -	min_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
>> -	min_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
>> +	max_limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
>> +	max_memsw_limit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
>>   	cgroup = memcg->css.cgroup;
>>   	if (!memcg->use_hierarchy)
>>   		goto out;
>> @@ -3943,13 +3943,13 @@ static void memcg_get_hierarchical_limit(struct mem_cgroup *memcg,
>>   		if (!memcg->use_hierarchy)
>>   			break;
>>   		tmp = res_counter_read_u64(&memcg->res, RES_LIMIT);
>> -		min_limit = min(min_limit, tmp);
>> +		max_limit = max(max_limit, tmp);
>>   		tmp = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
>> -		min_memsw_limit = min(min_memsw_limit, tmp);
>> +		max_memsw_limit = max(max_memsw_limit, tmp);
>>   	}
>>   out:
>> -	*mem_limit = min_limit;
>> -	*memsw_limit = min_memsw_limit;
>> +	*mem_limit = max_limit;
>> +	*memsw_limit = max_memsw_limit;
>>   }
>>   
>>   static int mem_cgroup_reset(struct cgroup *cont, unsigned int event)
>> 
>
>
>
>--
>To unsubscribe, send a message with 'unsubscribe linux-mm' in
>the body to majordomo@kvack.org.  For more info on Linux MM,
>see: http://www.linux-mm.org/ .
>Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm/memcg: wrap mem_cgroup_from_css function

[Kirill A. Shutemov]

On Wed, Jul 18, 2012 at 11:05:30AM +0800, Wanpeng Li wrote:
> wrap mem_cgroup_from_css function to clarify get mem cgroup
> from cgroup_subsys_state.
> 
> Signed-off-by: Wanpeng Li <liwanp@linux.vnet.ibm.com>
> Cc: Michal Hocko <mhocko@suse.cz>
> Cc: Johannes Weiner <hannes@cmpxchg.org>
> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Gavin Shan <shangw@linux.vnet.ibm.com>
> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com>
> Cc: linux-kernel@vger.kernel.org
> ---
>  mm/memcontrol.c |   14 ++++++++++----
>  1 files changed, 10 insertions(+), 4 deletions(-)
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 58a08fc..20f6a15 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -396,6 +396,12 @@ static void mem_cgroup_put(struct mem_cgroup *memcg);
>  #include <net/sock.h>
>  #include <net/ip.h>
>  
> +static inline
> +struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *s)
> +{
> +	return container_of(s, struct mem_cgroup, css);
> +}
> +
>  static bool mem_cgroup_is_root(struct mem_cgroup *memcg);
>  void sock_update_memcg(struct sock *sk)
>  {
> @@ -820,7 +826,7 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page)
>  
>  struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont)
>  {
> -	return container_of(cgroup_subsys_state(cont,
> +	return mem_cgroup_from_css(cgroup_subsys_state(cont,
>  				mem_cgroup_subsys_id), struct mem_cgroup,
>  				css);

Hm?.. Here and below too many args to mem_cgroup_from_css().
Have you tested the code?

>  }
> @@ -835,7 +841,7 @@ struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p)
>  	if (unlikely(!p))
>  		return NULL;
>  
> -	return container_of(task_subsys_state(p, mem_cgroup_subsys_id),
> +	return mem_cgroup_from_css(task_subsys_state(p, mem_cgroup_subsys_id),
>  				struct mem_cgroup, css);
>  }
>  
> @@ -922,7 +928,7 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root,
>  		css = css_get_next(&mem_cgroup_subsys, id + 1, &root->css, &id);
>  		if (css) {
>  			if (css == &root->css || css_tryget(css))
> -				memcg = container_of(css,
> +				memcg = mem_cgroup_from_css(css,
>  						     struct mem_cgroup, css);
>  		} else
>  			id = 0;
> @@ -2406,7 +2412,7 @@ static struct mem_cgroup *mem_cgroup_lookup(unsigned short id)
>  	css = css_lookup(&mem_cgroup_subsys, id);
>  	if (!css)
>  		return NULL;
> -	return container_of(css, struct mem_cgroup, css);
> +	return mem_cgroup_from_css(css, struct mem_cgroup, css);
>  }
>  
>  struct mem_cgroup *try_get_mem_cgroup_from_page(struct page *page)
> -- 
> 1.7.5.4
> 
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org.  For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

-- 
 Kirill A. Shutemov

Re: [PATCH] mm/memcg: wrap mem_cgroup_from_css function

[Wanpeng Li]

On Thu, Jul 19, 2012 at 12:14:20PM +0300, Kirill A. Shutemov wrote:
>On Wed, Jul 18, 2012 at 11:05:30AM +0800, Wanpeng Li wrote:
>> wrap mem_cgroup_from_css function to clarify get mem cgroup
>> from cgroup_subsys_state.
>> 
>> Signed-off-by: Wanpeng Li <liwanp@linux.vnet.ibm.com>
>> Cc: Michal Hocko <mhocko@suse.cz>
>> Cc: Johannes Weiner <hannes@cmpxchg.org>
>> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>> Cc: Andrew Morton <akpm@linux-foundation.org>
>> Cc: Gavin Shan <shangw@linux.vnet.ibm.com>
>> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com>
>> Cc: linux-kernel@vger.kernel.org
>> ---
>>  mm/memcontrol.c |   14 ++++++++++----
>>  1 files changed, 10 insertions(+), 4 deletions(-)
>> 
>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>> index 58a08fc..20f6a15 100644
>> --- a/mm/memcontrol.c
>> +++ b/mm/memcontrol.c
>> @@ -396,6 +396,12 @@ static void mem_cgroup_put(struct mem_cgroup *memcg);
>>  #include <net/sock.h>
>>  #include <net/ip.h>
>>  
>> +static inline
>> +struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *s)
>> +{
>> +	return container_of(s, struct mem_cgroup, css);
>> +}
>> +
>>  static bool mem_cgroup_is_root(struct mem_cgroup *memcg);
>>  void sock_update_memcg(struct sock *sk)
>>  {
>> @@ -820,7 +826,7 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page)
>>  
>>  struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont)
>>  {
>> -	return container_of(cgroup_subsys_state(cont,
>> +	return mem_cgroup_from_css(cgroup_subsys_state(cont,
>>  				mem_cgroup_subsys_id), struct mem_cgroup,
>>  				css);
>
>Hm?.. Here and below too many args to mem_cgroup_from_css().
>Have you tested the code?

Hi, what's the meaning of "two many"?

cgroup_subsys_state(cont, mem_cgroup_subsys_id) and 
task_subsys_state(p, mem_cgroup_subsys_id) both are 
just one arg in mem_cgroup_from_css. :-)

>
>>  }
>> @@ -835,7 +841,7 @@ struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p)
>>  	if (unlikely(!p))
>>  		return NULL;
>>  
>> -	return container_of(task_subsys_state(p, mem_cgroup_subsys_id),
>> +	return mem_cgroup_from_css(task_subsys_state(p, mem_cgroup_subsys_id),
>>  				struct mem_cgroup, css);
>>  }
>>  
>> @@ -922,7 +928,7 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root,
>>  		css = css_get_next(&mem_cgroup_subsys, id + 1, &root->css, &id);
>>  		if (css) {
>>  			if (css == &root->css || css_tryget(css))
>> -				memcg = container_of(css,
>> +				memcg = mem_cgroup_from_css(css,
>>  						     struct mem_cgroup, css);
>>  		} else
>>  			id = 0;
>> @@ -2406,7 +2412,7 @@ static struct mem_cgroup *mem_cgroup_lookup(unsigned short id)
>>  	css = css_lookup(&mem_cgroup_subsys, id);
>>  	if (!css)
>>  		return NULL;
>> -	return container_of(css, struct mem_cgroup, css);
>> +	return mem_cgroup_from_css(css, struct mem_cgroup, css);
>>  }
>>  
>>  struct mem_cgroup *try_get_mem_cgroup_from_page(struct page *page)
>> -- 
>> 1.7.5.4
>> 
>> --
>> To unsubscribe, send a message with 'unsubscribe linux-mm' in
>> the body to majordomo@kvack.org.  For more info on Linux MM,
>> see: http://www.linux-mm.org/ .
>> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
>
>-- 
> Kirill A. Shutemov

Re: [PATCH] mm/memcg: wrap mem_cgroup_from_css function

[Kirill A. Shutemov]

On Thu, Jul 19, 2012 at 05:23:09PM +0800, Wanpeng Li wrote:
> On Thu, Jul 19, 2012 at 12:14:20PM +0300, Kirill A. Shutemov wrote:
> >On Wed, Jul 18, 2012 at 11:05:30AM +0800, Wanpeng Li wrote:
> >> wrap mem_cgroup_from_css function to clarify get mem cgroup
> >> from cgroup_subsys_state.
> >> 
> >> Signed-off-by: Wanpeng Li <liwanp@linux.vnet.ibm.com>
> >> Cc: Michal Hocko <mhocko@suse.cz>
> >> Cc: Johannes Weiner <hannes@cmpxchg.org>
> >> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> >> Cc: Andrew Morton <akpm@linux-foundation.org>
> >> Cc: Gavin Shan <shangw@linux.vnet.ibm.com>
> >> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com>
> >> Cc: linux-kernel@vger.kernel.org
> >> ---
> >>  mm/memcontrol.c |   14 ++++++++++----
> >>  1 files changed, 10 insertions(+), 4 deletions(-)
> >> 
> >> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> >> index 58a08fc..20f6a15 100644
> >> --- a/mm/memcontrol.c
> >> +++ b/mm/memcontrol.c
> >> @@ -396,6 +396,12 @@ static void mem_cgroup_put(struct mem_cgroup *memcg);
> >>  #include <net/sock.h>
> >>  #include <net/ip.h>
> >>  
> >> +static inline
> >> +struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *s)
> >> +{
> >> +	return container_of(s, struct mem_cgroup, css);
> >> +}
> >> +
> >>  static bool mem_cgroup_is_root(struct mem_cgroup *memcg);
> >>  void sock_update_memcg(struct sock *sk)
> >>  {
> >> @@ -820,7 +826,7 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page)
> >>  
> >>  struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont)
> >>  {
> >> -	return container_of(cgroup_subsys_state(cont,
> >> +	return mem_cgroup_from_css(cgroup_subsys_state(cont,
> >>  				mem_cgroup_subsys_id), struct mem_cgroup,
> >>  				css);
> >
> >Hm?.. Here and below too many args to mem_cgroup_from_css().
> >Have you tested the code?
> 
> Hi, what's the meaning of "two many"?
> 
> cgroup_subsys_state(cont, mem_cgroup_subsys_id) and 
> task_subsys_state(p, mem_cgroup_subsys_id) both are 
> just one arg in mem_cgroup_from_css. :-)

Em.. I guess my eyes are broken. %)

By the patch I see that mem_cgroup_from_css() here has tree arguments:

1. cgroup_subsys_state(cont, mem_cgroup_subsys_id)
2. struct mem_cgroup
3. css

Where's my parser is broken?

-- 
 Kirill A. Shutemov

Re: [PATCH] mm/memcg: wrap mem_cgroup_from_css function

[Kirill A. Shutemov]

On Thu, Jul 19, 2012 at 05:38:35PM +0800, Gavin Shan wrote:
> On Thu, Jul 19, 2012 at 05:23:09PM +0800, Wanpeng Li wrote:
> >On Thu, Jul 19, 2012 at 12:14:20PM +0300, Kirill A. Shutemov wrote:
> >>On Wed, Jul 18, 2012 at 11:05:30AM +0800, Wanpeng Li wrote:
> >>> wrap mem_cgroup_from_css function to clarify get mem cgroup
> >>> from cgroup_subsys_state.
> >>> 
> >>> Signed-off-by: Wanpeng Li <liwanp@linux.vnet.ibm.com>
> >>> Cc: Michal Hocko <mhocko@suse.cz>
> >>> Cc: Johannes Weiner <hannes@cmpxchg.org>
> >>> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> >>> Cc: Andrew Morton <akpm@linux-foundation.org>
> >>> Cc: Gavin Shan <shangw@linux.vnet.ibm.com>
> >>> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com>
> >>> Cc: linux-kernel@vger.kernel.org
> >>> ---
> >>>  mm/memcontrol.c |   14 ++++++++++----
> >>>  1 files changed, 10 insertions(+), 4 deletions(-)
> >>> 
> >>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> >>> index 58a08fc..20f6a15 100644
> >>> --- a/mm/memcontrol.c
> >>> +++ b/mm/memcontrol.c
> >>> @@ -396,6 +396,12 @@ static void mem_cgroup_put(struct mem_cgroup *memcg);
> >>>  #include <net/sock.h>
> >>>  #include <net/ip.h>
> >>>  
> >>> +static inline
> >>> +struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *s)
> >>> +{
> >>> +	return container_of(s, struct mem_cgroup, css);
> >>> +}
> >>> +
> >>>  static bool mem_cgroup_is_root(struct mem_cgroup *memcg);
> >>>  void sock_update_memcg(struct sock *sk)
> >>>  {
> >>> @@ -820,7 +826,7 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page)
> >>>  
> >>>  struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont)
> >>>  {
> >>> -	return container_of(cgroup_subsys_state(cont,
> >>> +	return mem_cgroup_from_css(cgroup_subsys_state(cont,
> >>>  				mem_cgroup_subsys_id), struct mem_cgroup,
> >>>  				css);
> >>
> >>Hm?.. Here and below too many args to mem_cgroup_from_css().
> >>Have you tested the code?
> >
> >Hi, what's the meaning of "two many"?
> >
> 
> It might be the typo for "two" here.

Oops.. You're right.

-- 
 Kirill A. Shutemov

Re: [PATCH] mm/memcg: wrap mem_cgroup_from_css function

[Wanpeng Li]

[-- Attachment #1: Type: text/plain, Size: 4360 bytes --]

On Thu, Jul 19, 2012 at 05:38:35PM +0800, Gavin Shan wrote:
>On Thu, Jul 19, 2012 at 05:23:09PM +0800, Wanpeng Li wrote:
>>On Thu, Jul 19, 2012 at 12:14:20PM +0300, Kirill A. Shutemov wrote:
>>>On Wed, Jul 18, 2012 at 11:05:30AM +0800, Wanpeng Li wrote:
>>>> wrap mem_cgroup_from_css function to clarify get mem cgroup
>>>> from cgroup_subsys_state.
>>>> 
>>>> Signed-off-by: Wanpeng Li <liwanp@linux.vnet.ibm.com>
>>>> Cc: Michal Hocko <mhocko@suse.cz>
>>>> Cc: Johannes Weiner <hannes@cmpxchg.org>
>>>> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>>>> Cc: Andrew Morton <akpm@linux-foundation.org>
>>>> Cc: Gavin Shan <shangw@linux.vnet.ibm.com>
>>>> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com>
>>>> Cc: linux-kernel@vger.kernel.org
>>>> ---
>>>>  mm/memcontrol.c |   14 ++++++++++----
>>>>  1 files changed, 10 insertions(+), 4 deletions(-)
>>>> 
>>>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>>>> index 58a08fc..20f6a15 100644
>>>> --- a/mm/memcontrol.c
>>>> +++ b/mm/memcontrol.c
>>>> @@ -396,6 +396,12 @@ static void mem_cgroup_put(struct mem_cgroup *memcg);
>>>>  #include <net/sock.h>
>>>>  #include <net/ip.h>
>>>>  
>>>> +static inline
>>>> +struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *s)
>>>> +{
>>>> +	return container_of(s, struct mem_cgroup, css);
>>>> +}
>>>> +
>>>>  static bool mem_cgroup_is_root(struct mem_cgroup *memcg);
>>>>  void sock_update_memcg(struct sock *sk)
>>>>  {
>>>> @@ -820,7 +826,7 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page)
>>>>  
>>>>  struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont)
>>>>  {
>>>> -	return container_of(cgroup_subsys_state(cont,
>>>> +	return mem_cgroup_from_css(cgroup_subsys_state(cont,
>>>>  				mem_cgroup_subsys_id), struct mem_cgroup,
>>>>  				css);
>>>
>>>Hm?.. Here and below too many args to mem_cgroup_from_css().
>>>Have you tested the code?
>>
>>Hi, what's the meaning of "two many"?
>>
>
>It might be the typo for "two" here. I think it would be
>"too". However, it seems that you had pass "two" more arguments
>here to mem_cgroup_from_css() since the function only takes "one"
>parameter as you implemented before.
>
>+struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *s)
>
>+   return mem_cgroup_from_css(cgroup_subsys_state(cont,
>+		mem_cgroup_subsys_id), struct mem_cgroup,
>+		css);
>

Hi Andrew, 

Sorry for make a mistake. Please drop this patch from linux-mm 
and merged the new one. BTW, thanks Gavin and Kirill.

Best Regards,
Wanpeng Li


>Thanks,
>Gavin
>
>>cgroup_subsys_state(cont, mem_cgroup_subsys_id) and 
>>task_subsys_state(p, mem_cgroup_subsys_id) both are 
>>just one arg in mem_cgroup_from_css. :-)
>>
>>>
>>>>  }
>>>> @@ -835,7 +841,7 @@ struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p)
>>>>  	if (unlikely(!p))
>>>>  		return NULL;
>>>>  
>>>> -	return container_of(task_subsys_state(p, mem_cgroup_subsys_id),
>>>> +	return mem_cgroup_from_css(task_subsys_state(p, mem_cgroup_subsys_id),
>>>>  				struct mem_cgroup, css);
>>>>  }
>>>>  
>>>> @@ -922,7 +928,7 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root,
>>>>  		css = css_get_next(&mem_cgroup_subsys, id + 1, &root->css, &id);
>>>>  		if (css) {
>>>>  			if (css == &root->css || css_tryget(css))
>>>> -				memcg = container_of(css,
>>>> +				memcg = mem_cgroup_from_css(css,
>>>>  						     struct mem_cgroup, css);
>>>>  		} else
>>>>  			id = 0;
>>>> @@ -2406,7 +2412,7 @@ static struct mem_cgroup *mem_cgroup_lookup(unsigned short id)
>>>>  	css = css_lookup(&mem_cgroup_subsys, id);
>>>>  	if (!css)
>>>>  		return NULL;
>>>> -	return container_of(css, struct mem_cgroup, css);
>>>> +	return mem_cgroup_from_css(css, struct mem_cgroup, css);
>>>>  }
>>>>  
>>>>  struct mem_cgroup *try_get_mem_cgroup_from_page(struct page *page)
>>>> -- 
>>>> 1.7.5.4
>>>> 
>>>> --
>>>> To unsubscribe, send a message with 'unsubscribe linux-mm' in
>>>> the body to majordomo@kvack.org.  For more info on Linux MM,
>>>> see: http://www.linux-mm.org/ .
>>>> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
>>>
>>>-- 
>>> Kirill A. Shutemov
>
>--
>To unsubscribe, send a message with 'unsubscribe linux-mm' in
>the body to majordomo@kvack.org.  For more info on Linux MM,
>see: http://www.linux-mm.org/ .
>Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[-- Attachment #2: 0001-memcg-add-mem_cgroup_from_css-helper.patch --]
[-- Type: text/x-diff, Size: 2480 bytes --]

>From ca3849a5633d578d6a924817fad5602641707285 Mon Sep 17 00:00:00 2001
From: Wanpeng Li <liwanp@linux.vnet.ibm.com>
Date: Thu, 19 Jul 2012 18:08:31 +0800
Subject: [PATCH] memcg: add mem_cgroup_from_css() helper

Add a mem_cgroup_from_css() helper to replace open-coded invokations of
container_of().  To clarify the code and to add a little more type safety.

Signed-off-by: Wanpeng Li <liwanp@linux.vnet.ibm.com>
---
 mm/memcontrol.c |   19 +++++++++++--------
 1 files changed, 11 insertions(+), 8 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index f72b5e5..2136560 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -412,6 +412,12 @@ static void mem_cgroup_put(struct mem_cgroup *memcg);
 #include <net/sock.h>
 #include <net/ip.h>
 
+static inline
+struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *s)
+{
+	return container_of(s, struct mem_cgroup, css);
+}
+
 static bool mem_cgroup_is_root(struct mem_cgroup *memcg);
 void sock_update_memcg(struct sock *sk)
 {
@@ -864,9 +870,8 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page)
 
 struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont)
 {
-	return container_of(cgroup_subsys_state(cont,
-				mem_cgroup_subsys_id), struct mem_cgroup,
-				css);
+	return mem_cgroup_from_css(cgroup_subsys_state(cont,
+				mem_cgroup_subsys_id));
 }
 
 struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p)
@@ -879,8 +884,7 @@ struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p)
 	if (unlikely(!p))
 		return NULL;
 
-	return container_of(task_subsys_state(p, mem_cgroup_subsys_id),
-				struct mem_cgroup, css);
+	return mem_cgroup_from_css(task_subsys_state(p, mem_cgroup_subsys_id));
 }
 
 struct mem_cgroup *try_get_mem_cgroup_from_mm(struct mm_struct *mm)
@@ -966,8 +970,7 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root,
 		css = css_get_next(&mem_cgroup_subsys, id + 1, &root->css, &id);
 		if (css) {
 			if (css == &root->css || css_tryget(css))
-				memcg = container_of(css,
-						     struct mem_cgroup, css);
+				memcg = mem_cgroup_from_css(css);
 		} else
 			id = 0;
 		rcu_read_unlock();
@@ -2429,7 +2432,7 @@ static struct mem_cgroup *mem_cgroup_lookup(unsigned short id)
 	css = css_lookup(&mem_cgroup_subsys, id);
 	if (!css)
 		return NULL;
-	return container_of(css, struct mem_cgroup, css);
+	return mem_cgroup_from_css(css);
 }
 
 struct mem_cgroup *try_get_mem_cgroup_from_page(struct page *page)
-- 
1.7.7.6

[PATCH V2 00/35] CPUFreq: Implement light weight ->target(): for 3.13

[Viresh Kumar]

Currently prototype of cpufreq_drivers target routines is:

int target(struct cpufreq_policy *policy, unsigned int target_freq,
        unsigned int relation);

And most of the drivers call cpufreq_frequency_table_target() to get a valid
index of their frequency table which is closest to the target_freq. And they
don't use target_freq and relation after it.

So, it makes sense to just do this work in cpufreq core before calling
cpufreq_frequency_table_target() and simply pass index instead. But this can be
done only with drivers which expose their frequency table with cpufreq core. For
others we need to stick with the old prototype of target() until those drivers
are converted to expose frequency tables.

There are 7 drivers after this patchset which still use the heavy weight
version, i.e. target() and 44 drivers have adopted this new approach, i.e.
target_index().

Once those 7 drivers are also moved to use .target_index(), .target() will be
removed completely.

This is part 3 of my generic cpufreq cleanup stuff.. First two are posted here
and this one is rebased of them:

1: cpufreq: Introduce cpufreq_table_validate_and_show()
https://lkml.org/lkml/2013/8/8/263

2: cpufreq: define generic routines for cpufreq drivers
https://lkml.org/lkml/2013/8/10/48

All these are pushed here:
https://git.linaro.org/gitweb?p=people/vireshk/linux.git;a=shortlog;h=refs/heads/for-v3.13

V1->V2:
------
- Must be less ugly this time :)
- new interface is named as target_index() instead of target()
- old interface is kept as target() instead of target_old()
- few more drivers got converted to use this infrastructure (5)
- Documentation updates
- CONFIG_CPU_FREQ_TABLE removed completely as core depends on it now

Cc: Andrew Lunn <andrew@lunn.ch>
Cc: David S. Miller <davem@davemloft.net>
Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Cc: Eric Miao <eric.y.miao@gmail.com>
Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Cc: John Crispin <blogic@openwrt.org>
Cc: Kukjin Kim <kgene.kim@samsung.com>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: linux-cris-kernel@axis.com
Cc: Mikael Starvik <starvik@axis.com>
Cc: Santosh Shilimkar <santosh.shilimkar@ti.com>
Cc: Sekhar Nori <nsekhar@ti.com>
Cc: Shawn Guo <shawn.guo@linaro.org>
Cc: sparclinux@vger.kernel.org
Cc: Stephen Warren <swarren@nvidia.com>
Cc: Steven Miao <realmz6@gmail.com>
Cc: Tony Luck <tony.luck@intel.com>

Viresh Kumar (35):
  cpufreq: Implement light weight ->target_index() routine
  cpufreq: remove CONFIG_CPU_FREQ_TABLE
  cpufreq: acpi: Covert to light weight ->target_index() routine
  cpufreq: arm_big_little: Covert to light weight ->target_index()
    routine
  cpufreq: at32ap: Covert to light weight ->target_index() routine
  cpufreq: blackfin: Covert to light weight ->target_index() routine
  cpufreq: cpu0: Covert to light weight ->target_index() routine
  cpufreq: cris: Covert to light weight ->target_index() routine
  cpufreq: davinci: Covert to light weight ->target_index() routine
  cpufreq: dbx500: Covert to light weight ->target_index() routine
  cpufreq: e_powersaver: Covert to light weight ->target_index()
    routine
  cpufreq: elanfreq: Covert to light weight ->target_index() routine
  cpufreq: exynos: Covert to light weight ->target_index() routine
  cpufreq: ia64: Covert to light weight ->target_index() routine
  cpufreq: imx6q: Covert to light weight ->target_index() routine
  cpufreq: kirkwood: Covert to light weight ->target_index() routine
  cpufreq: longhaul: Covert to light weight ->target_index() routine
  cpufreq: loongson2: Covert to light weight ->target_index() routine
  cpufreq: maple: Covert to light weight ->target_index() routine
  cpufreq: omap: Covert to light weight ->target_index() routine
  cpufreq: p4: Covert to light weight ->target_index() routine
  cpufreq: pasemi: Covert to light weight ->target_index() routine
  cpufreq: pmac32: Covert to light weight ->target_index() routine
  cpufreq: powernow: Covert to light weight ->target_index() routine
  cpufreq: ppc: Covert to light weight ->target_index() routine
  cpufreq: pxa: Covert to light weight ->target_index() routine
  cpufreq: s3c2416: Covert to light weight ->target_index() routine
  cpufreq: s3c64xx: Covert to light weight ->target_index() routine
  cpufreq: s5pv210: Covert to light weight ->target_index() routine
  cpufreq: sa11x0: Covert to light weight ->target_index() routine
  cpufreq: sc520: Covert to light weight ->target_index() routine
  cpufreq: sparc: Covert to light weight ->target_index() routine
  cpufreq: SPEAr: Covert to light weight ->target_index() routine
  cpufreq: speedstep: Covert to light weight ->target_index() routine
  cpufreq: tegra: Covert to light weight ->target_index() routine

 Documentation/cpu-freq/cpu-drivers.txt | 27 ++++++++++------
 Documentation/cpu-freq/governors.txt   |  4 +--
 arch/arm/mach-davinci/Kconfig          |  1 -
 arch/arm/mach-pxa/Kconfig              |  3 --
 arch/arm/mach-sa1100/generic.c         | 20 ------------
 arch/arm/mach-sa1100/generic.h         |  2 --
 arch/arm/mach-ux500/Kconfig            |  1 -
 arch/blackfin/Kconfig                  |  1 -
 arch/cris/Kconfig                      |  2 --
 drivers/cpufreq/Kconfig                | 11 -------
 drivers/cpufreq/Kconfig.arm            | 11 -------
 drivers/cpufreq/Kconfig.powerpc        |  6 ----
 drivers/cpufreq/Kconfig.x86            | 13 --------
 drivers/cpufreq/Makefile               |  5 +--
 drivers/cpufreq/acpi-cpufreq.c         | 21 ++++---------
 drivers/cpufreq/arm_big_little.c       | 17 +++-------
 drivers/cpufreq/at32ap-cpufreq.c       | 23 +++-----------
 drivers/cpufreq/blackfin-cpufreq.c     | 17 +++-------
 drivers/cpufreq/cpufreq-cpu0.c         | 17 ++--------
 drivers/cpufreq/cpufreq.c              | 57 ++++++++++++++++++++++++++--------
 drivers/cpufreq/cris-artpec3-cpufreq.c | 18 ++---------
 drivers/cpufreq/cris-etraxfs-cpufreq.c | 17 ++--------
 drivers/cpufreq/davinci-cpufreq.c      | 16 ++--------
 drivers/cpufreq/dbx500-cpufreq.c       | 16 ++--------
 drivers/cpufreq/e_powersaver.c         | 17 ++--------
 drivers/cpufreq/elanfreq.c             | 34 ++------------------
 drivers/cpufreq/exynos-cpufreq.c       | 21 ++-----------
 drivers/cpufreq/exynos5440-cpufreq.c   | 13 ++------
 drivers/cpufreq/ia64-acpi-cpufreq.c    | 21 ++-----------
 drivers/cpufreq/imx6q-cpufreq.c        | 17 ++--------
 drivers/cpufreq/kirkwood-cpufreq.c     | 19 ++----------
 drivers/cpufreq/longhaul.c             | 13 ++------
 drivers/cpufreq/loongson2_cpufreq.c    | 21 +++----------
 drivers/cpufreq/maple-cpufreq.c        | 16 +++-------
 drivers/cpufreq/omap-cpufreq.c         | 31 ++----------------
 drivers/cpufreq/p4-clockmod.c          | 18 +++--------
 drivers/cpufreq/pasemi-cpufreq.c       | 12 ++-----
 drivers/cpufreq/pmac32-cpufreq.c       | 12 ++-----
 drivers/cpufreq/pmac64-cpufreq.c       | 17 +++-------
 drivers/cpufreq/powernow-k6.c          | 35 +++------------------
 drivers/cpufreq/powernow-k7.c          | 22 +++----------
 drivers/cpufreq/powernow-k8.c          | 24 +++++---------
 drivers/cpufreq/ppc-corenet-cpufreq.c  | 15 +++------
 drivers/cpufreq/ppc_cbe_cpufreq.c      | 12 ++-----
 drivers/cpufreq/pxa2xx-cpufreq.c       | 13 ++------
 drivers/cpufreq/pxa3xx-cpufreq.c       | 17 ++--------
 drivers/cpufreq/s3c2416-cpufreq.c      | 17 +++-------
 drivers/cpufreq/s3c64xx-cpufreq.c      | 18 +++--------
 drivers/cpufreq/s5pv210-cpufreq.c      | 54 +++++++++-----------------------
 drivers/cpufreq/sa1100-cpufreq.c       | 24 +++-----------
 drivers/cpufreq/sa1110-cpufreq.c       | 26 +++-------------
 drivers/cpufreq/sc520_freq.c           | 19 ++----------
 drivers/cpufreq/sparc-us2e-cpufreq.c   | 21 ++-----------
 drivers/cpufreq/sparc-us3-cpufreq.c    | 23 ++------------
 drivers/cpufreq/spear-cpufreq.c        | 12 +++----
 drivers/cpufreq/speedstep-centrino.c   | 26 +++++-----------
 drivers/cpufreq/speedstep-ich.c        | 24 ++++----------
 drivers/cpufreq/speedstep-smi.c        | 20 +++---------
 drivers/cpufreq/tegra-cpufreq.c        | 12 ++-----
 drivers/thermal/Kconfig                |  1 -
 include/linux/cpufreq.h                |  4 ++-
 61 files changed, 238 insertions(+), 809 deletions(-)

-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 01/35] cpufreq: Implement light weight ->target_index() routine

[Viresh Kumar]

Currently prototype of cpufreq_drivers target routines is:

int target(struct cpufreq_policy *policy, unsigned int target_freq,
		unsigned int relation);

And most of the drivers call cpufreq_frequency_table_target() to get a valid
index of their frequency table which is closest to the target_freq. And they
don't use target_freq and relation after it.

So, it makes sense to just do this work in cpufreq core before calling
cpufreq_frequency_table_target() and simply pass index instead. But this can be
done only with drivers which expose their frequency table with cpufreq core. For
others we need to stick with the old prototype of target() until those drivers
are converted to expose frequency tables.

This patch implements the new light weight prototype for target_index() routine.
It looks like this:

int target_index(struct cpufreq_policy *policy, unsigned int index);

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and pass index to it. Because CPUFreq core now requires to call routines
present in freq_table.c CONFIG_CPU_FREQ_TABLE must be enabled all the time.

This also marks target() interface as deprecated. So, that new drivers avoid
using it. And

Documentation is updated accordingly.

Cc: Andrew Lunn <andrew@lunn.ch>
Cc: David S. Miller <davem@davemloft.net>
Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Cc: Eric Miao <eric.y.miao@gmail.com>
Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Cc: John Crispin <blogic@openwrt.org>
Cc: Kukjin Kim <kgene.kim@samsung.com>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: linux-cris-kernel@axis.com
Cc: Mikael Starvik <starvik@axis.com>
Cc: Santosh Shilimkar <santosh.shilimkar@ti.com>
Cc: Sekhar Nori <nsekhar@ti.com>
Cc: Shawn Guo <shawn.guo@linaro.org>
Cc: sparclinux@vger.kernel.org
Cc: Stephen Warren <swarren@nvidia.com>
Cc: Steven Miao <realmz6@gmail.com>
Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 Documentation/cpu-freq/cpu-drivers.txt | 27 +++++++++++------
 Documentation/cpu-freq/governors.txt   |  4 +--
 drivers/cpufreq/Kconfig                |  1 +
 drivers/cpufreq/cpufreq.c              | 55 +++++++++++++++++++++++++++-------
 include/linux/cpufreq.h                |  4 ++-
 5 files changed, 68 insertions(+), 23 deletions(-)

diff --git a/Documentation/cpu-freq/cpu-drivers.txt b/Documentation/cpu-freq/cpu-drivers.txt
index 40282e6..8b1a445 100644
--- a/Documentation/cpu-freq/cpu-drivers.txt
+++ b/Documentation/cpu-freq/cpu-drivers.txt
@@ -23,8 +23,8 @@ Contents:
 1.1  Initialization
 1.2  Per-CPU Initialization
 1.3  verify
-1.4  target or setpolicy?
-1.5  target
+1.4  target/target_index or setpolicy?
+1.5  target/target_index
 1.6  setpolicy
 2.   Frequency Table Helpers
 
@@ -56,7 +56,8 @@ cpufreq_driver.init -		A pointer to the per-CPU initialization
 cpufreq_driver.verify -		A pointer to a "verification" function.
 
 cpufreq_driver.setpolicy _or_ 
-cpufreq_driver.target -		See below on the differences.
+cpufreq_driver.target/
+target_index		-	See below on the differences.
 
 And optionally
 
@@ -66,7 +67,7 @@ cpufreq_driver.resume -		A pointer to a per-CPU resume function
 				which is called with interrupts disabled
 				and _before_ the pre-suspend frequency
 				and/or policy is restored by a call to
-				->target or ->setpolicy.
+				->target/target_index or ->setpolicy.
 
 cpufreq_driver.attr -		A pointer to a NULL-terminated list of
 				"struct freq_attr" which allow to
@@ -103,8 +104,8 @@ policy->governor		must contain the "default policy" for
 				this CPU. A few moments later,
 				cpufreq_driver.verify and either
 				cpufreq_driver.setpolicy or
-				cpufreq_driver.target is called with
-				these values.
+				cpufreq_driver.target/target_index is called
+				with these values.
 
 For setting some of these values (cpuinfo.min[max]_freq, policy->min[max]), the
 frequency table helpers might be helpful. See the section 2 for more information
@@ -133,20 +134,28 @@ range) is within policy->min and policy->max. If necessary, increase
 policy->max first, and only if this is no solution, decrease policy->min.
 
 
-1.4 target or setpolicy?
+1.4 target/target_index or setpolicy?
 ----------------------------
 
 Most cpufreq drivers or even most cpu frequency scaling algorithms 
 only allow the CPU to be set to one frequency. For these, you use the
-->target call.
+->target/target_index call.
 
 Some cpufreq-capable processors switch the frequency between certain
 limits on their own. These shall use the ->setpolicy call
 
 
-1.4. target
+1.4. target/target_index
 -------------
 
+The target_index call has two arguments: struct cpufreq_policy *policy,
+and unsigned int index (into the exposed frequency table).
+
+The CPUfreq driver must set the new frequency when called here. The
+actual frequency must be determined by freq_table[index].frequency.
+
+Deprecated:
+----------
 The target call has three arguments: struct cpufreq_policy *policy,
 unsigned int target_frequency, unsigned int relation.
 
diff --git a/Documentation/cpu-freq/governors.txt b/Documentation/cpu-freq/governors.txt
index 219970b..77ec215 100644
--- a/Documentation/cpu-freq/governors.txt
+++ b/Documentation/cpu-freq/governors.txt
@@ -40,7 +40,7 @@ Most cpufreq drivers (in fact, all except one, longrun) or even most
 cpu frequency scaling algorithms only offer the CPU to be set to one
 frequency. In order to offer dynamic frequency scaling, the cpufreq
 core must be able to tell these drivers of a "target frequency". So
-these specific drivers will be transformed to offer a "->target"
+these specific drivers will be transformed to offer a "->target/target_index"
 call instead of the existing "->setpolicy" call. For "longrun", all
 stays the same, though.
 
@@ -71,7 +71,7 @@ CPU can be set to switch independently	 |	   CPU can only be set
 		    /			       the limits of policy->{min,max}
 		   /			            \
 		  /				     \
-	Using the ->setpolicy call,		 Using the ->target call,
+	Using the ->setpolicy call,		 Using the ->target/target_index call,
 	    the limits and the			  the frequency closest
 	     "policy" is set.			  to target_freq is set.
 						  It is assured that it
diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig
index 534fcb8..2d06754 100644
--- a/drivers/cpufreq/Kconfig
+++ b/drivers/cpufreq/Kconfig
@@ -2,6 +2,7 @@ menu "CPU Frequency scaling"
 
 config CPU_FREQ
 	bool "CPU Frequency scaling"
+	select CPU_FREQ_TABLE
 	help
 	  CPU Frequency scaling allows you to change the clock speed of 
 	  CPUs on the fly. This is a nice method to save power, because 
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 37a6874..f1b0e0f 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -47,6 +47,11 @@ static LIST_HEAD(cpufreq_policy_list);
 static DEFINE_PER_CPU(char[CPUFREQ_NAME_LEN], cpufreq_cpu_governor);
 #endif
 
+static inline bool has_target(void)
+{
+	return cpufreq_driver->target_index || cpufreq_driver->target;
+}
+
 /*
  * cpu_policy_rwsem is a per CPU reader-writer semaphore designed to cure
  * all cpufreq/hotplug/workqueue/etc related lock issues.
@@ -377,7 +382,7 @@ static int cpufreq_parse_governor(char *str_governor, unsigned int *policy,
 			*policy = CPUFREQ_POLICY_POWERSAVE;
 			err = 0;
 		}
-	} else if (cpufreq_driver->target) {
+	} else if (has_target()) {
 		struct cpufreq_governor *t;
 
 		mutex_lock(&cpufreq_governor_mutex);
@@ -539,7 +544,7 @@ static ssize_t show_scaling_available_governors(struct cpufreq_policy *policy,
 	ssize_t i = 0;
 	struct cpufreq_governor *t;
 
-	if (!cpufreq_driver->target) {
+	if (!has_target()) {
 		i += sprintf(buf, "performance powersave");
 		goto out;
 	}
@@ -822,7 +827,7 @@ static int cpufreq_add_dev_interface(struct cpufreq_policy *policy,
 		if (ret)
 			goto err_out_kobj_put;
 	}
-	if (cpufreq_driver->target) {
+	if (has_target()) {
 		ret = sysfs_create_file(&policy->kobj, &scaling_cur_freq.attr);
 		if (ret)
 			goto err_out_kobj_put;
@@ -871,10 +876,10 @@ static int cpufreq_add_policy_cpu(struct cpufreq_policy *policy,
 				  unsigned int cpu, struct device *dev,
 				  bool frozen)
 {
-	int ret = 0, has_target = !!cpufreq_driver->target;
+	int ret = 0;
 	unsigned long flags;
 
-	if (has_target) {
+	if (has_target()) {
 		ret = __cpufreq_governor(policy, CPUFREQ_GOV_STOP);
 		if (ret) {
 			pr_err("%s: Failed to stop governor\n", __func__);
@@ -893,7 +898,7 @@ static int cpufreq_add_policy_cpu(struct cpufreq_policy *policy,
 
 	unlock_policy_rwsem_write(policy->cpu);
 
-	if (has_target) {
+	if (has_target()) {
 		if ((ret = __cpufreq_governor(policy, CPUFREQ_GOV_START)) ||
 			(ret = __cpufreq_governor(policy, CPUFREQ_GOV_LIMITS))) {
 			pr_err("%s: Failed to start governor\n", __func__);
@@ -1204,7 +1209,7 @@ static int __cpufreq_remove_dev(struct device *dev,
 		return -EINVAL;
 	}
 
-	if (cpufreq_driver->target) {
+	if (has_target()) {
 		ret = __cpufreq_governor(policy, CPUFREQ_GOV_STOP);
 		if (ret) {
 			pr_err("%s: Failed to stop governor\n", __func__);
@@ -1244,7 +1249,7 @@ static int __cpufreq_remove_dev(struct device *dev,
 
 	/* If cpu is last user of policy, free policy */
 	if (cpus == 1) {
-		if (cpufreq_driver->target) {
+		if (has_target()) {
 			ret = __cpufreq_governor(policy,
 					CPUFREQ_GOV_POLICY_EXIT);
 			if (ret) {
@@ -1282,7 +1287,7 @@ static int __cpufreq_remove_dev(struct device *dev,
 		if (!frozen)
 			cpufreq_policy_free(policy);
 	} else {
-		if (cpufreq_driver->target) {
+		if (has_target()) {
 			if ((ret = __cpufreq_governor(policy, CPUFREQ_GOV_START)) ||
 					(ret = __cpufreq_governor(policy, CPUFREQ_GOV_LIMITS))) {
 				pr_err("%s: Failed to start governor\n",
@@ -1646,11 +1651,39 @@ int __cpufreq_driver_target(struct cpufreq_policy *policy,
 	pr_debug("target for CPU %u: %u kHz, relation %u, requested %u kHz\n",
 			policy->cpu, target_freq, relation, old_target_freq);
 
+	/*
+	 * This might look like a redundant call as we are checking it again
+	 * after finding index. But it is left intentionally for cases where
+	 * exactly same freq is called again and so we can save on few function
+	 * calls.
+	 */
 	if (target_freq == policy->cur)
 		return 0;
 
 	if (cpufreq_driver->target)
 		retval = cpufreq_driver->target(policy, target_freq, relation);
+	else if (cpufreq_driver->target_index) {
+		struct cpufreq_frequency_table *freq_table;
+		int index;
+
+		freq_table = cpufreq_frequency_get_table(policy->cpu);
+		if (unlikely(!freq_table)) {
+			pr_err("%s: Unable to find freq_table\n", __func__);
+			return retval;
+		}
+
+		retval = cpufreq_frequency_table_target(policy, freq_table,
+				target_freq, relation, &index);
+		if (unlikely(retval)) {
+			pr_err("%s: Unable to find matching freq\n", __func__);
+			return retval;
+		}
+
+		if (freq_table[index].frequency == policy->cur)
+			return 0;
+
+		retval = cpufreq_driver->target_index(policy, index);
+	}
 
 	return retval;
 }
@@ -1983,7 +2016,7 @@ int cpufreq_update_policy(unsigned int cpu)
 			pr_debug("Driver did not initialize current freq");
 			policy->cur = new_policy.cur;
 		} else {
-			if (policy->cur != new_policy.cur && cpufreq_driver->target)
+			if (policy->cur != new_policy.cur && has_target())
 				cpufreq_out_of_sync(cpu, policy->cur,
 								new_policy.cur);
 		}
@@ -2058,7 +2091,7 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data)
 		return -ENODEV;
 
 	if (!driver_data || !driver_data->verify || !driver_data->init ||
-	    ((!driver_data->setpolicy) && (!driver_data->target)))
+	    (!driver_data->setpolicy && !has_target()))
 		return -EINVAL;
 
 	pr_debug("trying to register driver %s\n", driver_data->name);
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 4907eb2..ff9c8df 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -195,9 +195,11 @@ struct cpufreq_driver {
 
 	/* define one out of two */
 	int	(*setpolicy)	(struct cpufreq_policy *policy);
-	int	(*target)	(struct cpufreq_policy *policy,
+	int	(*target)	(struct cpufreq_policy *policy,	/* Deprecated */
 				 unsigned int target_freq,
 				 unsigned int relation);
+	int	(*target_index)	(struct cpufreq_policy *policy,
+				 unsigned int index);
 
 	/* should be defined, if possible */
 	unsigned int	(*get)	(unsigned int cpu);
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 02/35] cpufreq: remove CONFIG_CPU_FREQ_TABLE

[Viresh Kumar]

CONFIG_CPU_FREQ_TABLE will be always enabled when cpufreq framework is used, as
cpufreq core depends on it. So, we don't need this CONFIG option anymore as it
is not configurable. Remove CONFIG_CPU_FREQ_TABLE and update its users.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 arch/arm/mach-davinci/Kconfig   |  1 -
 arch/arm/mach-pxa/Kconfig       |  3 ---
 arch/arm/mach-ux500/Kconfig     |  1 -
 arch/blackfin/Kconfig           |  1 -
 arch/cris/Kconfig               |  2 --
 drivers/cpufreq/Kconfig         | 12 ------------
 drivers/cpufreq/Kconfig.arm     | 11 -----------
 drivers/cpufreq/Kconfig.powerpc |  6 ------
 drivers/cpufreq/Kconfig.x86     | 13 -------------
 drivers/cpufreq/Makefile        |  5 +----
 drivers/cpufreq/cpufreq.c       |  2 --
 drivers/thermal/Kconfig         |  1 -
 12 files changed, 1 insertion(+), 57 deletions(-)

diff --git a/arch/arm/mach-davinci/Kconfig b/arch/arm/mach-davinci/Kconfig
index e026b19..a075b3e 100644
--- a/arch/arm/mach-davinci/Kconfig
+++ b/arch/arm/mach-davinci/Kconfig
@@ -40,7 +40,6 @@ config ARCH_DAVINCI_DA850
 	bool "DA850/OMAP-L138/AM18x based system"
 	select ARCH_DAVINCI_DA8XX
 	select ARCH_HAS_CPUFREQ
-	select CPU_FREQ_TABLE
 	select CP_INTC
 
 config ARCH_DAVINCI_DA8XX
diff --git a/arch/arm/mach-pxa/Kconfig b/arch/arm/mach-pxa/Kconfig
index a842711..96100db 100644
--- a/arch/arm/mach-pxa/Kconfig
+++ b/arch/arm/mach-pxa/Kconfig
@@ -615,14 +615,12 @@ endmenu
 config PXA25x
 	bool
 	select CPU_XSCALE
-	select CPU_FREQ_TABLE if CPU_FREQ
 	help
 	  Select code specific to PXA21x/25x/26x variants
 
 config PXA27x
 	bool
 	select CPU_XSCALE
-	select CPU_FREQ_TABLE if CPU_FREQ
 	help
 	  Select code specific to PXA27x variants
 
@@ -635,7 +633,6 @@ config CPU_PXA26x
 config PXA3xx
 	bool
 	select CPU_XSC3
-	select CPU_FREQ_TABLE if CPU_FREQ
 	help
 	  Select code specific to PXA3xx variants
 
diff --git a/arch/arm/mach-ux500/Kconfig b/arch/arm/mach-ux500/Kconfig
index b19b072..e6eaefa 100644
--- a/arch/arm/mach-ux500/Kconfig
+++ b/arch/arm/mach-ux500/Kconfig
@@ -34,7 +34,6 @@ config UX500_SOC_COMMON
 
 config UX500_SOC_DB8500
 	bool
-	select CPU_FREQ_TABLE if CPU_FREQ
 	select MFD_DB8500_PRCMU
 	select PINCTRL_DB8500
 	select PINCTRL_DB8540
diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index 3b6abc5..1ac474a 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -1430,7 +1430,6 @@ source "drivers/cpufreq/Kconfig"
 config BFIN_CPU_FREQ
 	bool
 	depends on CPU_FREQ
-	select CPU_FREQ_TABLE
 	default y
 
 config CPU_VOLTAGE
diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig
index 3201ddb..9f3c5436 100644
--- a/arch/cris/Kconfig
+++ b/arch/cris/Kconfig
@@ -134,13 +134,11 @@ config SVINTO_SIM
 
 config ETRAXFS
 	bool "ETRAX-FS-V32"
-	select CPU_FREQ_TABLE if CPU_FREQ
 	help
 	  Support CRIS V32.
 
 config CRIS_MACH_ARTPEC3
         bool "ARTPEC-3"
-	select CPU_FREQ_TABLE if CPU_FREQ
         help
           Support Axis ARTPEC-3.
 
diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig
index 2d06754..38093e2 100644
--- a/drivers/cpufreq/Kconfig
+++ b/drivers/cpufreq/Kconfig
@@ -2,7 +2,6 @@ menu "CPU Frequency scaling"
 
 config CPU_FREQ
 	bool "CPU Frequency scaling"
-	select CPU_FREQ_TABLE
 	help
 	  CPU Frequency scaling allows you to change the clock speed of 
 	  CPUs on the fly. This is a nice method to save power, because 
@@ -18,15 +17,11 @@ config CPU_FREQ
 
 if CPU_FREQ
 
-config CPU_FREQ_TABLE
-	tristate
-
 config CPU_FREQ_GOV_COMMON
 	bool
 
 config CPU_FREQ_STAT
 	tristate "CPU frequency translation statistics"
-	select CPU_FREQ_TABLE
 	default y
 	help
 	  This driver exports CPU frequency statistics information through sysfs
@@ -144,7 +139,6 @@ config CPU_FREQ_GOV_USERSPACE
 
 config CPU_FREQ_GOV_ONDEMAND
 	tristate "'ondemand' cpufreq policy governor"
-	select CPU_FREQ_TABLE
 	select CPU_FREQ_GOV_COMMON
 	help
 	  'ondemand' - This driver adds a dynamic cpufreq policy governor.
@@ -188,7 +182,6 @@ config CPU_FREQ_GOV_CONSERVATIVE
 config GENERIC_CPUFREQ_CPU0
 	tristate "Generic CPU0 cpufreq driver"
 	depends on HAVE_CLK && REGULATOR && PM_OPP && OF
-	select CPU_FREQ_TABLE
 	help
 	  This adds a generic cpufreq driver for CPU0 frequency management.
 	  It supports both uniprocessor (UP) and symmetric multiprocessor (SMP)
@@ -224,7 +217,6 @@ depends on IA64
 
 config IA64_ACPI_CPUFREQ
 	tristate "ACPI Processor P-States driver"
-	select CPU_FREQ_TABLE
 	depends on ACPI_PROCESSOR
 	help
 	This driver adds a CPUFreq driver which utilizes the ACPI
@@ -241,7 +233,6 @@ depends on MIPS
 
 config LOONGSON2_CPUFREQ
 	tristate "Loongson2 CPUFreq Driver"
-	select CPU_FREQ_TABLE
 	help
 	  This option adds a CPUFreq driver for loongson processors which
 	  support software configurable cpu frequency.
@@ -263,7 +254,6 @@ menu "SPARC CPU frequency scaling drivers"
 depends on SPARC64
 config SPARC_US3_CPUFREQ
 	tristate "UltraSPARC-III CPU Frequency driver"
-	select CPU_FREQ_TABLE
 	help
 	  This adds the CPUFreq driver for UltraSPARC-III processors.
 
@@ -273,7 +263,6 @@ config SPARC_US3_CPUFREQ
 
 config SPARC_US2E_CPUFREQ
 	tristate "UltraSPARC-IIe CPU Frequency driver"
-	select CPU_FREQ_TABLE
 	help
 	  This adds the CPUFreq driver for UltraSPARC-IIe processors.
 
@@ -286,7 +275,6 @@ menu "SH CPU Frequency scaling"
 depends on SUPERH
 config SH_CPU_FREQ
 	tristate "SuperH CPU Frequency driver"
-	select CPU_FREQ_TABLE
 	help
 	  This adds the cpufreq driver for SuperH. Any CPU that supports
 	  clock rate rounding through the clock framework can use this
diff --git a/drivers/cpufreq/Kconfig.arm b/drivers/cpufreq/Kconfig.arm
index de4d5d9..1bbb71e 100644
--- a/drivers/cpufreq/Kconfig.arm
+++ b/drivers/cpufreq/Kconfig.arm
@@ -5,7 +5,6 @@
 config ARM_BIG_LITTLE_CPUFREQ
 	tristate "Generic ARM big LITTLE CPUfreq driver"
 	depends on ARM_CPU_TOPOLOGY && PM_OPP && HAVE_CLK
-	select CPU_FREQ_TABLE
 	help
 	  This enables the Generic CPUfreq driver for ARM big.LITTLE platforms.
 
@@ -19,7 +18,6 @@ config ARM_DT_BL_CPUFREQ
 config ARM_EXYNOS_CPUFREQ
 	bool "SAMSUNG EXYNOS SoCs"
 	depends on ARCH_EXYNOS
-	select CPU_FREQ_TABLE
 	default y
 	help
 	  This adds the CPUFreq driver common part for Samsung
@@ -48,7 +46,6 @@ config ARM_EXYNOS5250_CPUFREQ
 config ARM_EXYNOS5440_CPUFREQ
 	def_bool SOC_EXYNOS5440
 	depends on HAVE_CLK && PM_OPP && OF
-	select CPU_FREQ_TABLE
 	help
 	  This adds the CPUFreq driver for Samsung EXYNOS5440
 	  SoC. The nature of exynos5440 clock controller is
@@ -73,7 +70,6 @@ config ARM_IMX6Q_CPUFREQ
 	tristate "Freescale i.MX6Q cpufreq support"
 	depends on SOC_IMX6Q
 	depends on REGULATOR_ANATOP
-	select CPU_FREQ_TABLE
 	help
 	  This adds cpufreq driver support for Freescale i.MX6Q SOC.
 
@@ -89,7 +85,6 @@ config ARM_INTEGRATOR
 
 config ARM_KIRKWOOD_CPUFREQ
 	def_bool ARCH_KIRKWOOD && OF
-	select CPU_FREQ_TABLE
 	help
 	  This adds the CPUFreq driver for Marvell Kirkwood
 	  SoCs.
@@ -98,7 +93,6 @@ config ARM_OMAP2PLUS_CPUFREQ
 	bool "TI OMAP2+"
 	depends on ARCH_OMAP2PLUS
 	default ARCH_OMAP2PLUS
-	select CPU_FREQ_TABLE
 
 config ARM_S3C_CPUFREQ
 	bool
@@ -153,7 +147,6 @@ config ARM_S3C2412_CPUFREQ
 config ARM_S3C2416_CPUFREQ
 	bool "S3C2416 CPU Frequency scaling support"
 	depends on CPU_S3C2416
-	select CPU_FREQ_TABLE
 	help
 	  This adds the CPUFreq driver for the Samsung S3C2416 and
 	  S3C2450 SoC. The S3C2416 supports changing the rate of the
@@ -184,7 +177,6 @@ config ARM_S3C2440_CPUFREQ
 config ARM_S3C64XX_CPUFREQ
 	bool "Samsung S3C64XX"
 	depends on CPU_S3C6410
-	select CPU_FREQ_TABLE
 	default y
 	help
 	  This adds the CPUFreq driver for Samsung S3C6410 SoC.
@@ -194,7 +186,6 @@ config ARM_S3C64XX_CPUFREQ
 config ARM_S5PV210_CPUFREQ
 	bool "Samsung S5PV210 and S5PC110"
 	depends on CPU_S5PV210
-	select CPU_FREQ_TABLE
 	default y
 	help
 	  This adds the CPUFreq driver for Samsung S5PV210 and
@@ -211,7 +202,6 @@ config ARM_SA1110_CPUFREQ
 config ARM_SPEAR_CPUFREQ
 	bool "SPEAr CPUFreq support"
 	depends on PLAT_SPEAR
-	select CPU_FREQ_TABLE
 	default y
 	help
 	  This adds the CPUFreq driver support for SPEAr SOCs.
@@ -219,7 +209,6 @@ config ARM_SPEAR_CPUFREQ
 config ARM_TEGRA_CPUFREQ
 	bool "TEGRA CPUFreq support"
 	depends on ARCH_TEGRA
-	select CPU_FREQ_TABLE
 	default y
 	help
 	  This adds the CPUFreq driver support for TEGRA SOCs.
diff --git a/drivers/cpufreq/Kconfig.powerpc b/drivers/cpufreq/Kconfig.powerpc
index 25ca9db..ca0021a 100644
--- a/drivers/cpufreq/Kconfig.powerpc
+++ b/drivers/cpufreq/Kconfig.powerpc
@@ -1,7 +1,6 @@
 config CPU_FREQ_CBE
 	tristate "CBE frequency scaling"
 	depends on CBE_RAS && PPC_CELL
-	select CPU_FREQ_TABLE
 	default m
 	help
 	  This adds the cpufreq driver for Cell BE processors.
@@ -20,7 +19,6 @@ config CPU_FREQ_CBE_PMI
 config CPU_FREQ_MAPLE
 	bool "Support for Maple 970FX Evaluation Board"
 	depends on PPC_MAPLE
-	select CPU_FREQ_TABLE
 	help
 	  This adds support for frequency switching on Maple 970FX
 	  Evaluation Board and compatible boards (IBM JS2x blades).
@@ -28,7 +26,6 @@ config CPU_FREQ_MAPLE
 config PPC_CORENET_CPUFREQ
 	tristate "CPU frequency scaling driver for Freescale E500MC SoCs"
 	depends on PPC_E500MC && OF && COMMON_CLK
-	select CPU_FREQ_TABLE
 	select CLK_PPC_CORENET
 	help
 	  This adds the CPUFreq driver support for Freescale e500mc,
@@ -38,7 +35,6 @@ config PPC_CORENET_CPUFREQ
 config CPU_FREQ_PMAC
 	bool "Support for Apple PowerBooks"
 	depends on ADB_PMU && PPC32
-	select CPU_FREQ_TABLE
 	help
 	  This adds support for frequency switching on Apple PowerBooks,
 	  this currently includes some models of iBook & Titanium
@@ -47,7 +43,6 @@ config CPU_FREQ_PMAC
 config CPU_FREQ_PMAC64
 	bool "Support for some Apple G5s"
 	depends on PPC_PMAC && PPC64
-	select CPU_FREQ_TABLE
 	help
 	  This adds support for frequency switching on Apple iMac G5,
 	  and some of the more recent desktop G5 machines as well.
@@ -55,7 +50,6 @@ config CPU_FREQ_PMAC64
 config PPC_PASEMI_CPUFREQ
 	bool "Support for PA Semi PWRficient"
 	depends on PPC_PASEMI
-	select CPU_FREQ_TABLE
 	default y
 	help
 	  This adds the support for frequency switching on PA Semi
diff --git a/drivers/cpufreq/Kconfig.x86 b/drivers/cpufreq/Kconfig.x86
index e2b6eab..6897ad8 100644
--- a/drivers/cpufreq/Kconfig.x86
+++ b/drivers/cpufreq/Kconfig.x86
@@ -31,7 +31,6 @@ config X86_PCC_CPUFREQ
 
 config X86_ACPI_CPUFREQ
 	tristate "ACPI Processor P-States driver"
-	select CPU_FREQ_TABLE
 	depends on ACPI_PROCESSOR
 	help
 	  This driver adds a CPUFreq driver which utilizes the ACPI
@@ -60,7 +59,6 @@ config X86_ACPI_CPUFREQ_CPB
 
 config ELAN_CPUFREQ
 	tristate "AMD Elan SC400 and SC410"
-	select CPU_FREQ_TABLE
 	depends on MELAN
 	---help---
 	  This adds the CPUFreq driver for AMD Elan SC400 and SC410
@@ -76,7 +74,6 @@ config ELAN_CPUFREQ
 
 config SC520_CPUFREQ
 	tristate "AMD Elan SC520"
-	select CPU_FREQ_TABLE
 	depends on MELAN
 	---help---
 	  This adds the CPUFreq driver for AMD Elan SC520 processor.
@@ -88,7 +85,6 @@ config SC520_CPUFREQ
 
 config X86_POWERNOW_K6
 	tristate "AMD Mobile K6-2/K6-3 PowerNow!"
-	select CPU_FREQ_TABLE
 	depends on X86_32
 	help
 	  This adds the CPUFreq driver for mobile AMD K6-2+ and mobile
@@ -100,7 +96,6 @@ config X86_POWERNOW_K6
 
 config X86_POWERNOW_K7
 	tristate "AMD Mobile Athlon/Duron PowerNow!"
-	select CPU_FREQ_TABLE
 	depends on X86_32
 	help
 	  This adds the CPUFreq driver for mobile AMD K7 mobile processors.
@@ -118,7 +113,6 @@ config X86_POWERNOW_K7_ACPI
 
 config X86_POWERNOW_K8
 	tristate "AMD Opteron/Athlon64 PowerNow!"
-	select CPU_FREQ_TABLE
 	depends on ACPI && ACPI_PROCESSOR && X86_ACPI_CPUFREQ
 	help
 	  This adds the CPUFreq driver for K8/early Opteron/Athlon64 processors.
@@ -132,7 +126,6 @@ config X86_POWERNOW_K8
 config X86_AMD_FREQ_SENSITIVITY
 	tristate "AMD frequency sensitivity feedback powersave bias"
 	depends on CPU_FREQ_GOV_ONDEMAND && X86_ACPI_CPUFREQ && CPU_SUP_AMD
-	select CPU_FREQ_TABLE
 	help
 	  This adds AMD-specific powersave bias function to the ondemand
 	  governor, which allows it to make more power-conscious frequency
@@ -160,7 +153,6 @@ config X86_GX_SUSPMOD
 
 config X86_SPEEDSTEP_CENTRINO
 	tristate "Intel Enhanced SpeedStep (deprecated)"
-	select CPU_FREQ_TABLE
 	select X86_SPEEDSTEP_CENTRINO_TABLE if X86_32
 	depends on X86_32 || (X86_64 && ACPI_PROCESSOR)
 	help
@@ -190,7 +182,6 @@ config X86_SPEEDSTEP_CENTRINO_TABLE
 
 config X86_SPEEDSTEP_ICH
 	tristate "Intel Speedstep on ICH-M chipsets (ioport interface)"
-	select CPU_FREQ_TABLE
 	depends on X86_32
 	help
 	  This adds the CPUFreq driver for certain mobile Intel Pentium III
@@ -204,7 +195,6 @@ config X86_SPEEDSTEP_ICH
 
 config X86_SPEEDSTEP_SMI
 	tristate "Intel SpeedStep on 440BX/ZX/MX chipsets (SMI interface)"
-	select CPU_FREQ_TABLE
 	depends on X86_32
 	help
 	  This adds the CPUFreq driver for certain mobile Intel Pentium III
@@ -217,7 +207,6 @@ config X86_SPEEDSTEP_SMI
 
 config X86_P4_CLOCKMOD
 	tristate "Intel Pentium 4 clock modulation"
-	select CPU_FREQ_TABLE
 	help
 	  This adds the CPUFreq driver for Intel Pentium 4 / XEON
 	  processors.  When enabled it will lower CPU temperature by skipping
@@ -259,7 +248,6 @@ config X86_LONGRUN
 
 config X86_LONGHAUL
 	tristate "VIA Cyrix III Longhaul"
-	select CPU_FREQ_TABLE
 	depends on X86_32 && ACPI_PROCESSOR
 	help
 	  This adds the CPUFreq driver for VIA Samuel/CyrixIII,
@@ -272,7 +260,6 @@ config X86_LONGHAUL
 
 config X86_E_POWERSAVER
 	tristate "VIA C7 Enhanced PowerSaver (DANGEROUS)"
-	select CPU_FREQ_TABLE
 	depends on X86_32 && ACPI_PROCESSOR
 	help
 	  This adds the CPUFreq driver for VIA C7 processors.  However, this driver
diff --git a/drivers/cpufreq/Makefile b/drivers/cpufreq/Makefile
index ad5866c..b7948bb 100644
--- a/drivers/cpufreq/Makefile
+++ b/drivers/cpufreq/Makefile
@@ -1,5 +1,5 @@
 # CPUfreq core
-obj-$(CONFIG_CPU_FREQ)			+= cpufreq.o
+obj-$(CONFIG_CPU_FREQ)			+= cpufreq.o freq_table.o
 # CPUfreq stats
 obj-$(CONFIG_CPU_FREQ_STAT)             += cpufreq_stats.o
 
@@ -11,9 +11,6 @@ obj-$(CONFIG_CPU_FREQ_GOV_ONDEMAND)	+= cpufreq_ondemand.o
 obj-$(CONFIG_CPU_FREQ_GOV_CONSERVATIVE)	+= cpufreq_conservative.o
 obj-$(CONFIG_CPU_FREQ_GOV_COMMON)		+= cpufreq_governor.o
 
-# CPUfreq cross-arch helpers
-obj-$(CONFIG_CPU_FREQ_TABLE)		+= freq_table.o
-
 obj-$(CONFIG_GENERIC_CPUFREQ_CPU0)	+= cpufreq-cpu0.o
 
 ##################################################################################
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index f1b0e0f..4d37306 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -1130,9 +1130,7 @@ static void update_policy_cpu(struct cpufreq_policy *policy, unsigned int cpu)
 	for_each_cpu(j, policy->cpus)
 		per_cpu(cpufreq_policy_cpu, j) = cpu;
 
-#ifdef CONFIG_CPU_FREQ_TABLE
 	cpufreq_frequency_table_update_policy_cpu(policy);
-#endif
 	blocking_notifier_call_chain(&cpufreq_policy_notifier_list,
 			CPUFREQ_UPDATE_POLICY_CPU, policy);
 }
diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
index e988c81..6a5b948 100644
--- a/drivers/thermal/Kconfig
+++ b/drivers/thermal/Kconfig
@@ -69,7 +69,6 @@ config THERMAL_GOV_USER_SPACE
 config CPU_THERMAL
 	bool "generic cpu cooling support"
 	depends on CPU_FREQ
-	select CPU_FREQ_TABLE
 	help
 	  This implements the generic cpu cooling mechanism through frequency
 	  reduction. An ACPI version of this already exists
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 03/35] cpufreq: acpi: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/acpi-cpufreq.c | 21 ++++++---------------
 1 file changed, 6 insertions(+), 15 deletions(-)

diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c
index cd5badb..f69b4c8 100644
--- a/drivers/cpufreq/acpi-cpufreq.c
+++ b/drivers/cpufreq/acpi-cpufreq.c
@@ -424,17 +424,17 @@ static unsigned int check_freqs(const struct cpumask *mask, unsigned int freq,
 }
 
 static int acpi_cpufreq_target(struct cpufreq_policy *policy,
-			       unsigned int target_freq, unsigned int relation)
+			       unsigned int index)
 {
 	struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu);
 	struct acpi_processor_performance *perf;
 	struct cpufreq_freqs freqs;
 	struct drv_cmd cmd;
-	unsigned int next_state = 0; /* Index into freq_table */
 	unsigned int next_perf_state = 0; /* Index into perf table */
 	int result = 0;
 
-	pr_debug("acpi_cpufreq_target %d (%d)\n", target_freq, policy->cpu);
+	pr_debug("acpi_cpufreq_target %d (%d)\n",
+			data->freq_table[index].frequency, policy->cpu);
 
 	if (unlikely(data == NULL ||
 	     data->acpi_data == NULL || data->freq_table == NULL)) {
@@ -442,16 +442,7 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
 	}
 
 	perf = data->acpi_data;
-	result = cpufreq_frequency_table_target(policy,
-						data->freq_table,
-						target_freq,
-						relation, &next_state);
-	if (unlikely(result)) {
-		result = -ENODEV;
-		goto out;
-	}
-
-	next_perf_state = data->freq_table[next_state].driver_data;
+	next_perf_state = data->freq_table[index].driver_data;
 	if (perf->state == next_perf_state) {
 		if (unlikely(data->resume)) {
 			pr_debug("Called after resume, resetting to P%d\n",
@@ -493,7 +484,7 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
 		cmd.mask = cpumask_of(policy->cpu);
 
 	freqs.old = perf->states[perf->state].core_frequency * 1000;
-	freqs.new = data->freq_table[next_state].frequency;
+	freqs.new = data->freq_table[index].frequency;
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 
 	drv_write(&cmd);
@@ -919,7 +910,7 @@ static struct freq_attr *acpi_cpufreq_attr[] = {
 
 static struct cpufreq_driver acpi_cpufreq_driver = {
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= acpi_cpufreq_target,
+	.target_index	= acpi_cpufreq_target,
 	.bios_limit	= acpi_processor_get_bios_limit,
 	.init		= acpi_cpufreq_cpu_init,
 	.exit		= acpi_cpufreq_cpu_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 04/35] cpufreq: arm_big_little: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/arm_big_little.c | 17 +++++------------
 1 file changed, 5 insertions(+), 12 deletions(-)

diff --git a/drivers/cpufreq/arm_big_little.c b/drivers/cpufreq/arm_big_little.c
index 7e92586..4bba5ad 100644
--- a/drivers/cpufreq/arm_big_little.c
+++ b/drivers/cpufreq/arm_big_little.c
@@ -49,28 +49,21 @@ static unsigned int bL_cpufreq_get(unsigned int cpu)
 
 /* Set clock frequency */
 static int bL_cpufreq_set_target(struct cpufreq_policy *policy,
-		unsigned int target_freq, unsigned int relation)
+		unsigned int index)
 {
 	struct cpufreq_freqs freqs;
-	u32 cpu = policy->cpu, freq_tab_idx, cur_cluster;
+	u32 cpu = policy->cpu, cur_cluster;
 	int ret = 0;
 
 	cur_cluster = cpu_to_cluster(policy->cpu);
 
 	freqs.old = bL_cpufreq_get(policy->cpu);
-
-	/* Determine valid target frequency using freq_table */
-	cpufreq_frequency_table_target(policy, freq_table[cur_cluster],
-			target_freq, relation, &freq_tab_idx);
-	freqs.new = freq_table[cur_cluster][freq_tab_idx].frequency;
+	freqs.new = freq_table[cur_cluster][index].frequency;
 
 	pr_debug("%s: cpu: %d, cluster: %d, oldfreq: %d, target freq: %d, new freq: %d\n",
-			__func__, cpu, cur_cluster, freqs.old, target_freq,
+			__func__, cpu, cur_cluster, freqs.old, freqs.new,
 			freqs.new);
 
-	if (freqs.old == freqs.new)
-		return 0;
-
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 
 	ret = clk_set_rate(clk[cur_cluster], freqs.new * 1000);
@@ -201,7 +194,7 @@ static struct cpufreq_driver bL_cpufreq_driver = {
 	.name			= "arm-big-little",
 	.flags			= CPUFREQ_STICKY,
 	.verify			= cpufreq_generic_frequency_table_verify,
-	.target			= bL_cpufreq_set_target,
+	.target_index		= bL_cpufreq_set_target,
 	.get			= bL_cpufreq_get,
 	.init			= bL_cpufreq_init,
 	.exit			= bL_cpufreq_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 05/35] cpufreq: at32ap: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/at32ap-cpufreq.c | 23 +++++------------------
 1 file changed, 5 insertions(+), 18 deletions(-)

diff --git a/drivers/cpufreq/at32ap-cpufreq.c b/drivers/cpufreq/at32ap-cpufreq.c
index 788f7e7..2e964a7 100644
--- a/drivers/cpufreq/at32ap-cpufreq.c
+++ b/drivers/cpufreq/at32ap-cpufreq.c
@@ -35,25 +35,12 @@ static unsigned int at32_get_speed(unsigned int cpu)
 static unsigned int	ref_freq;
 static unsigned long	loops_per_jiffy_ref;
 
-static int at32_set_target(struct cpufreq_policy *policy,
-			  unsigned int target_freq,
-			  unsigned int relation)
+static int at32_set_target(struct cpufreq_policy *policy, unsigned int index)
 {
 	struct cpufreq_freqs freqs;
-	long freq;
-
-	/* Convert target_freq from kHz to Hz */
-	freq = clk_round_rate(cpuclk, target_freq * 1000);
-
-	/* Check if policy->min <= new_freq <= policy->max */
-	if(freq < (policy->min * 1000) || freq > (policy->max * 1000))
-		return -EINVAL;
-
-	pr_debug("cpufreq: requested frequency %u Hz\n", target_freq * 1000);
 
 	freqs.old = at32_get_speed(0);
-	freqs.new = (freq + 500) / 1000;
-	freqs.flags = 0;
+	freqs.new = freq_table[index].frequency;
 
 	if (!ref_freq) {
 		ref_freq = freqs.old;
@@ -64,13 +51,13 @@ static int at32_set_target(struct cpufreq_policy *policy,
 	if (freqs.old < freqs.new)
 		boot_cpu_data.loops_per_jiffy = cpufreq_scale(
 				loops_per_jiffy_ref, ref_freq, freqs.new);
-	clk_set_rate(cpuclk, freq);
+	clk_set_rate(cpuclk, freqs.new * 1000);
 	if (freqs.new < freqs.old)
 		boot_cpu_data.loops_per_jiffy = cpufreq_scale(
 				loops_per_jiffy_ref, ref_freq, freqs.new);
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
 
-	pr_debug("cpufreq: set frequency %lu Hz\n", freq);
+	pr_debug("cpufreq: set frequency %lu Hz\n", freqs.new * 1000);
 
 	return 0;
 }
@@ -143,7 +130,7 @@ static struct cpufreq_driver at32_driver = {
 	.name		= "at32ap",
 	.init		= at32_cpufreq_driver_init,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= at32_set_target,
+	.target_index	= at32_set_target,
 	.get		= at32_get_speed,
 	.flags		= CPUFREQ_STICKY,
 };
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 06/35] cpufreq: blackfin: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Steven Miao <realmz6@gmail.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/blackfin-cpufreq.c | 17 ++++-------------
 1 file changed, 4 insertions(+), 13 deletions(-)

diff --git a/drivers/cpufreq/blackfin-cpufreq.c b/drivers/cpufreq/blackfin-cpufreq.c
index 48888cf..b343c7b 100644
--- a/drivers/cpufreq/blackfin-cpufreq.c
+++ b/drivers/cpufreq/blackfin-cpufreq.c
@@ -127,14 +127,11 @@ unsigned long cpu_set_cclk(int cpu, unsigned long new)
 }
 #endif
 
-static int bfin_target(struct cpufreq_policy *policy,
-			unsigned int target_freq, unsigned int relation)
+static int bfin_target(struct cpufreq_policy *policy, unsigned int index)
 {
 #ifndef CONFIG_BF60x
 	unsigned int plldiv;
 #endif
-	unsigned int index;
-	unsigned long cclk_hz;
 	struct cpufreq_freqs freqs;
 	static unsigned long lpj_ref;
 	static unsigned int  lpj_ref_freq;
@@ -144,17 +141,11 @@ static int bfin_target(struct cpufreq_policy *policy,
 	cycles_t cycles;
 #endif
 
-	if (cpufreq_frequency_table_target(policy, bfin_freq_table, target_freq,
-				relation, &index))
-		return -EINVAL;
-
-	cclk_hz = bfin_freq_table[index].frequency;
-
 	freqs.old = bfin_getfreq_khz(0);
-	freqs.new = cclk_hz;
+	freqs.new = bfin_freq_table[index].frequency;
 
 	pr_debug("cpufreq: changing cclk to %lu; target = %u, oldfreq = %u\n",
-			cclk_hz, target_freq, freqs.old);
+			freqs.new, freqs.new, freqs.old);
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 #ifndef CONFIG_BF60x
@@ -210,7 +201,7 @@ static int __bfin_cpu_init(struct cpufreq_policy *policy)
 
 static struct cpufreq_driver bfin_driver = {
 	.verify = cpufreq_generic_frequency_table_verify,
-	.target = bfin_target,
+	.target_index = bfin_target,
 	.get = bfin_getfreq_khz,
 	.init = __bfin_cpu_init,
 	.exit = cpufreq_generic_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 07/35] cpufreq: cpu0: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/cpufreq-cpu0.c | 17 ++---------------
 1 file changed, 2 insertions(+), 15 deletions(-)

diff --git a/drivers/cpufreq/cpufreq-cpu0.c b/drivers/cpufreq/cpufreq-cpu0.c
index 3d24e7b..fb07a40 100644
--- a/drivers/cpufreq/cpufreq-cpu0.c
+++ b/drivers/cpufreq/cpufreq-cpu0.c
@@ -34,24 +34,14 @@ static unsigned int cpu0_get_speed(unsigned int cpu)
 	return clk_get_rate(cpu_clk) / 1000;
 }
 
-static int cpu0_set_target(struct cpufreq_policy *policy,
-			   unsigned int target_freq, unsigned int relation)
+static int cpu0_set_target(struct cpufreq_policy *policy, unsigned int index)
 {
 	struct cpufreq_freqs freqs;
 	struct opp *opp;
 	unsigned long volt = 0, volt_old = 0, tol = 0;
 	long freq_Hz, freq_exact;
-	unsigned int index;
 	int ret;
 
-	ret = cpufreq_frequency_table_target(policy, freq_table, target_freq,
-					     relation, &index);
-	if (ret) {
-		pr_err("failed to match target freqency %d: %d\n",
-		       target_freq, ret);
-		return ret;
-	}
-
 	freq_Hz = clk_round_rate(cpu_clk, freq_table[index].frequency * 1000);
 	if (freq_Hz < 0)
 		freq_Hz = freq_table[index].frequency * 1000;
@@ -59,9 +49,6 @@ static int cpu0_set_target(struct cpufreq_policy *policy,
 	freqs.new = freq_Hz / 1000;
 	freqs.old = clk_get_rate(cpu_clk) / 1000;
 
-	if (freqs.old == freqs.new)
-		return 0;
-
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 
 	if (cpu_reg) {
@@ -145,7 +132,7 @@ static int cpu0_cpufreq_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver cpu0_cpufreq_driver = {
 	.flags = CPUFREQ_STICKY,
 	.verify = cpufreq_generic_frequency_table_verify,
-	.target = cpu0_set_target,
+	.target_index = cpu0_set_target,
 	.get = cpu0_get_speed,
 	.init = cpu0_cpufreq_init,
 	.exit = cpufreq_generic_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 08/35] cpufreq: cris: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Cc: Mikael Starvik <starvik@axis.com>
Cc: linux-cris-kernel@axis.com
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/cris-artpec3-cpufreq.c | 18 ++----------------
 drivers/cpufreq/cris-etraxfs-cpufreq.c | 17 ++---------------
 2 files changed, 4 insertions(+), 31 deletions(-)

diff --git a/drivers/cpufreq/cris-artpec3-cpufreq.c b/drivers/cpufreq/cris-artpec3-cpufreq.c
index d26f4e4..1488277 100644
--- a/drivers/cpufreq/cris-artpec3-cpufreq.c
+++ b/drivers/cpufreq/cris-artpec3-cpufreq.c
@@ -27,8 +27,7 @@ static unsigned int cris_freq_get_cpu_frequency(unsigned int cpu)
 	return clk_ctrl.pll ? 200000 : 6000;
 }
 
-static void cris_freq_set_cpu_state(struct cpufreq_policy *policy,
-		unsigned int state)
+static int cris_freq_target(struct cpufreq_policy *policy, unsigned int state)
 {
 	struct cpufreq_freqs freqs;
 	reg_clkgen_rw_clk_ctrl clk_ctrl;
@@ -52,19 +51,6 @@ static void cris_freq_set_cpu_state(struct cpufreq_policy *policy,
 	local_irq_enable();
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
-};
-
-static int cris_freq_target(struct cpufreq_policy *policy,
-			    unsigned int target_freq,
-			    unsigned int relation)
-{
-	unsigned int newstate = 0;
-
-	if (cpufreq_frequency_table_target(policy, cris_freq_table,
-			target_freq, relation, &newstate))
-		return -EINVAL;
-
-	cris_freq_set_cpu_state(policy, newstate);
 
 	return 0;
 }
@@ -82,7 +68,7 @@ static int cris_freq_cpu_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver cris_freq_driver = {
 	.get	= cris_freq_get_cpu_frequency,
 	.verify	= cpufreq_generic_frequency_table_verify,
-	.target	= cris_freq_target,
+	.target_index = cris_freq_target,
 	.init	= cris_freq_cpu_init,
 	.exit	= cpufreq_generic_exit,
 	.name	= "cris_freq",
diff --git a/drivers/cpufreq/cris-etraxfs-cpufreq.c b/drivers/cpufreq/cris-etraxfs-cpufreq.c
index d384e63..4e3e9c7 100644
--- a/drivers/cpufreq/cris-etraxfs-cpufreq.c
+++ b/drivers/cpufreq/cris-etraxfs-cpufreq.c
@@ -27,8 +27,7 @@ static unsigned int cris_freq_get_cpu_frequency(unsigned int cpu)
 	return clk_ctrl.pll ? 200000 : 6000;
 }
 
-static void cris_freq_set_cpu_state(struct cpufreq_policy *policy,
-		unsigned int state)
+static int cris_freq_target(struct cpufreq_policy *policy, unsigned int state)
 {
 	struct cpufreq_freqs freqs;
 	reg_config_rw_clk_ctrl clk_ctrl;
@@ -52,18 +51,6 @@ static void cris_freq_set_cpu_state(struct cpufreq_policy *policy,
 	local_irq_enable();
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
-};
-
-static int cris_freq_target(struct cpufreq_policy *policy,
-			    unsigned int target_freq, unsigned int relation)
-{
-	unsigned int newstate = 0;
-
-	if (cpufreq_frequency_table_target
-	    (policy, cris_freq_table, target_freq, relation, &newstate))
-		return -EINVAL;
-
-	cris_freq_set_cpu_state(policy, newstate);
 
 	return 0;
 }
@@ -80,7 +67,7 @@ static int cris_freq_cpu_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver cris_freq_driver = {
 	.get = cris_freq_get_cpu_frequency,
 	.verify = cpufreq_generic_frequency_table_verify,
-	.target = cris_freq_target,
+	.target_index = cris_freq_target,
 	.init = cris_freq_cpu_init,
 	.exit = cpufreq_generic_exit,
 	.name = "cris_freq",
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 09/35] cpufreq: davinci: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/davinci-cpufreq.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/drivers/cpufreq/davinci-cpufreq.c b/drivers/cpufreq/davinci-cpufreq.c
index 33e9460..822100b 100644
--- a/drivers/cpufreq/davinci-cpufreq.c
+++ b/drivers/cpufreq/davinci-cpufreq.c
@@ -68,28 +68,18 @@ static unsigned int davinci_getspeed(unsigned int cpu)
 	return clk_get_rate(cpufreq.armclk) / 1000;
 }
 
-static int davinci_target(struct cpufreq_policy *policy,
-				unsigned int target_freq, unsigned int relation)
+static int davinci_target(struct cpufreq_policy *policy, unsigned int idx)
 {
 	int ret = 0;
-	unsigned int idx;
 	struct cpufreq_freqs freqs;
 	struct davinci_cpufreq_config *pdata = cpufreq.dev->platform_data;
 	struct clk *armclk = cpufreq.armclk;
 
 	freqs.old = davinci_getspeed(0);
-	freqs.new = clk_round_rate(armclk, target_freq * 1000) / 1000;
-
-	if (freqs.old == freqs.new)
-		return ret;
+	freqs.new = pdata->freq_table[idx].frequency;
 
 	dev_dbg(cpufreq.dev, "transition: %u --> %u\n", freqs.old, freqs.new);
 
-	ret = cpufreq_frequency_table_target(policy, pdata->freq_table,
-						freqs.new, relation, &idx);
-	if (ret)
-		return -EINVAL;
-
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 
 	/* if moving to higher frequency, up the voltage beforehand */
@@ -160,7 +150,7 @@ static int davinci_cpu_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver davinci_driver = {
 	.flags		= CPUFREQ_STICKY,
 	.verify		= davinci_verify_speed,
-	.target		= davinci_target,
+	.target_index	= davinci_target,
 	.get		= davinci_getspeed,
 	.init		= davinci_cpu_init,
 	.exit		= cpufreq_generic_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 10/35] cpufreq: dbx500: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/dbx500-cpufreq.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/drivers/cpufreq/dbx500-cpufreq.c b/drivers/cpufreq/dbx500-cpufreq.c
index 7e2f9c0..28deaf0 100644
--- a/drivers/cpufreq/dbx500-cpufreq.c
+++ b/drivers/cpufreq/dbx500-cpufreq.c
@@ -20,23 +20,13 @@ static struct cpufreq_frequency_table *freq_table;
 static struct clk *armss_clk;
 
 static int dbx500_cpufreq_target(struct cpufreq_policy *policy,
-				unsigned int target_freq,
-				unsigned int relation)
+				unsigned int index)
 {
 	struct cpufreq_freqs freqs;
-	unsigned int idx;
 	int ret;
 
-	/* Lookup the next frequency */
-	if (cpufreq_frequency_table_target(policy, freq_table, target_freq,
-					relation, &idx))
-		return -EINVAL;
-
 	freqs.old = policy->cur;
-	freqs.new = freq_table[idx].frequency;
-
-	if (freqs.old == freqs.new)
-		return 0;
+	freqs.new = freq_table[index].frequency;
 
 	/* pre-change notification */
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
@@ -104,7 +94,7 @@ static int dbx500_cpufreq_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver dbx500_cpufreq_driver = {
 	.flags  = CPUFREQ_STICKY | CPUFREQ_CONST_LOOPS,
 	.verify = cpufreq_generic_frequency_table_verify,
-	.target = dbx500_cpufreq_target,
+	.target_index = dbx500_cpufreq_target,
 	.get    = dbx500_cpufreq_getspeed,
 	.init   = dbx500_cpufreq_init,
 	.name   = "DBX500",
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 11/35] cpufreq: e_powersaver: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/e_powersaver.c | 17 +++--------------
 1 file changed, 3 insertions(+), 14 deletions(-)

diff --git a/drivers/cpufreq/e_powersaver.c b/drivers/cpufreq/e_powersaver.c
index a8cbbd0..60cd576 100644
--- a/drivers/cpufreq/e_powersaver.c
+++ b/drivers/cpufreq/e_powersaver.c
@@ -168,12 +168,9 @@ postchange:
 	return err;
 }
 
-static int eps_target(struct cpufreq_policy *policy,
-			       unsigned int target_freq,
-			       unsigned int relation)
+static int eps_target(struct cpufreq_policy *policy, unsigned int index)
 {
 	struct eps_cpu_data *centaur;
-	unsigned int newstate = 0;
 	unsigned int cpu = policy->cpu;
 	unsigned int dest_state;
 	int ret;
@@ -182,16 +179,8 @@ static int eps_target(struct cpufreq_policy *policy,
 		return -ENODEV;
 	centaur = eps_cpu[cpu];
 
-	if (unlikely(cpufreq_frequency_table_target(policy,
-			&eps_cpu[cpu]->freq_table[0],
-			target_freq,
-			relation,
-			&newstate))) {
-		return -EINVAL;
-	}
-
 	/* Make frequency transition */
-	dest_state = centaur->freq_table[newstate].driver_data & 0xffff;
+	dest_state = centaur->freq_table[index].driver_data & 0xffff;
 	ret = eps_set_state(centaur, policy, dest_state);
 	if (ret)
 		printk(KERN_ERR "eps: Timeout!\n");
@@ -419,7 +408,7 @@ static int eps_cpu_exit(struct cpufreq_policy *policy)
 
 static struct cpufreq_driver eps_driver = {
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= eps_target,
+	.target_index	= eps_target,
 	.init		= eps_cpu_init,
 	.exit		= eps_cpu_exit,
 	.get		= eps_get,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 12/35] cpufreq: elanfreq: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/elanfreq.c | 34 +++-------------------------------
 1 file changed, 3 insertions(+), 31 deletions(-)

diff --git a/drivers/cpufreq/elanfreq.c b/drivers/cpufreq/elanfreq.c
index fe7053c..0d133a7 100644
--- a/drivers/cpufreq/elanfreq.c
+++ b/drivers/cpufreq/elanfreq.c
@@ -105,20 +105,8 @@ static unsigned int elanfreq_get_cpu_frequency(unsigned int cpu)
 }
 
 
-/**
- *	elanfreq_set_cpu_frequency: Change the CPU core frequency
- *	@cpu: cpu number
- *	@freq: frequency in kHz
- *
- *	This function takes a frequency value and changes the CPU frequency
- *	according to this. Note that the frequency has to be checked by
- *	elanfreq_validatespeed() for correctness!
- *
- *	There is no return value.
- */
-
-static void elanfreq_set_cpu_state(struct cpufreq_policy *policy,
-		unsigned int state)
+static int elanfreq_target(struct cpufreq_policy *policy,
+			    unsigned int state)
 {
 	struct cpufreq_freqs    freqs;
 
@@ -162,25 +150,9 @@ static void elanfreq_set_cpu_state(struct cpufreq_policy *policy,
 	local_irq_enable();
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
-};
-
-
-static int elanfreq_target(struct cpufreq_policy *policy,
-			    unsigned int target_freq,
-			    unsigned int relation)
-{
-	unsigned int newstate = 0;
-
-	if (cpufreq_frequency_table_target(policy, &elanfreq_table[0],
-				target_freq, relation, &newstate))
-		return -EINVAL;
-
-	elanfreq_set_cpu_state(policy, newstate);
 
 	return 0;
 }
-
-
 /*
  *	Module init and exit code
  */
@@ -238,7 +210,7 @@ __setup("elanfreq=", elanfreq_setup);
 static struct cpufreq_driver elanfreq_driver = {
 	.get		= elanfreq_get_cpu_frequency,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= elanfreq_target,
+	.target_index	= elanfreq_target,
 	.init		= elanfreq_cpu_init,
 	.exit		= cpufreq_generic_exit,
 	.name		= "elanfreq",
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 13/35] cpufreq: exynos: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/exynos-cpufreq.c     | 21 +++------------------
 drivers/cpufreq/exynos5440-cpufreq.c | 13 +++----------
 2 files changed, 6 insertions(+), 28 deletions(-)

diff --git a/drivers/cpufreq/exynos-cpufreq.c b/drivers/cpufreq/exynos-cpufreq.c
index 7663a96..8735b42 100644
--- a/drivers/cpufreq/exynos-cpufreq.c
+++ b/drivers/cpufreq/exynos-cpufreq.c
@@ -65,9 +65,6 @@ static int exynos_cpufreq_scale(unsigned int target_freq)
 	freqs.old = policy->cur;
 	freqs.new = target_freq;
 
-	if (freqs.new == freqs.old)
-		goto out;
-
 	/*
 	 * The policy max have been changed so that we cannot get proper
 	 * old_index with cpufreq_frequency_table_target(). Thus, ignore
@@ -151,13 +148,9 @@ out:
 	return ret;
 }
 
-static int exynos_target(struct cpufreq_policy *policy,
-			  unsigned int target_freq,
-			  unsigned int relation)
+static int exynos_target(struct cpufreq_policy *policy, unsigned int index)
 {
 	struct cpufreq_frequency_table *freq_table = exynos_info->freq_table;
-	unsigned int index;
-	unsigned int new_freq;
 	int ret = 0;
 
 	mutex_lock(&cpufreq_lock);
@@ -165,15 +158,7 @@ static int exynos_target(struct cpufreq_policy *policy,
 	if (frequency_locked)
 		goto out;
 
-	if (cpufreq_frequency_table_target(policy, freq_table,
-					   target_freq, relation, &index)) {
-		ret = -EINVAL;
-		goto out;
-	}
-
-	new_freq = freq_table[index].frequency;
-
-	ret = exynos_cpufreq_scale(new_freq);
+	ret = exynos_cpufreq_scale(freq_table[index].frequency);
 
 out:
 	mutex_unlock(&cpufreq_lock);
@@ -254,7 +239,7 @@ static int exynos_cpufreq_cpu_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver exynos_driver = {
 	.flags		= CPUFREQ_STICKY,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= exynos_target,
+	.target_index	= exynos_target,
 	.get		= exynos_getspeed,
 	.init		= exynos_cpufreq_cpu_init,
 	.exit		= cpufreq_generic_exit,
diff --git a/drivers/cpufreq/exynos5440-cpufreq.c b/drivers/cpufreq/exynos5440-cpufreq.c
index f139b3b..9a5ed485 100644
--- a/drivers/cpufreq/exynos5440-cpufreq.c
+++ b/drivers/cpufreq/exynos5440-cpufreq.c
@@ -214,21 +214,14 @@ static unsigned int exynos_getspeed(unsigned int cpu)
 	return dvfs_info->cur_frequency;
 }
 
-static int exynos_target(struct cpufreq_policy *policy,
-			  unsigned int target_freq,
-			  unsigned int relation)
+static int exynos_target(struct cpufreq_policy *policy, unsigned int index)
 {
-	unsigned int index, tmp;
+	unsigned int tmp;
 	int ret = 0, i;
 	struct cpufreq_frequency_table *freq_table = dvfs_info->freq_table;
 
 	mutex_lock(&cpufreq_lock);
 
-	ret = cpufreq_frequency_table_target(policy, freq_table,
-					   target_freq, relation, &index);
-	if (ret)
-		goto out;
-
 	freqs.old = dvfs_info->cur_frequency;
 	freqs.new = freq_table[index].frequency;
 
@@ -333,7 +326,7 @@ static int exynos_cpufreq_cpu_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver exynos_driver = {
 	.flags		= CPUFREQ_STICKY,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= exynos_target,
+	.target_index	= exynos_target,
 	.get		= exynos_getspeed,
 	.init		= exynos_cpufreq_cpu_init,
 	.exit		= cpufreq_generic_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 14/35] cpufreq: ia64: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/ia64-acpi-cpufreq.c | 21 +++------------------
 1 file changed, 3 insertions(+), 18 deletions(-)

diff --git a/drivers/cpufreq/ia64-acpi-cpufreq.c b/drivers/cpufreq/ia64-acpi-cpufreq.c
index b958bdb..371c63d 100644
--- a/drivers/cpufreq/ia64-acpi-cpufreq.c
+++ b/drivers/cpufreq/ia64-acpi-cpufreq.c
@@ -227,26 +227,11 @@ acpi_cpufreq_get (
 static int
 acpi_cpufreq_target (
 	struct cpufreq_policy   *policy,
-	unsigned int target_freq,
-	unsigned int relation)
+	unsigned int index)
 {
-	struct cpufreq_acpi_io *data = acpi_io_data[policy->cpu];
-	unsigned int next_state = 0;
-	unsigned int result = 0;
-
-	pr_debug("acpi_cpufreq_setpolicy\n");
-
-	result = cpufreq_frequency_table_target(policy,
-			data->freq_table, target_freq, relation, &next_state);
-	if (result)
-		return (result);
-
-	result = processor_set_freq(data, policy, next_state);
-
-	return (result);
+	return processor_set_freq(acpi_io_data[policy->cpu], policy, index);
 }
 
-
 static int
 acpi_cpufreq_cpu_init (
 	struct cpufreq_policy   *policy)
@@ -380,7 +365,7 @@ acpi_cpufreq_cpu_exit (
 
 static struct cpufreq_driver acpi_cpufreq_driver = {
 	.verify 	= cpufreq_generic_frequency_table_verify,
-	.target 	= acpi_cpufreq_target,
+	.target_index	= acpi_cpufreq_target,
 	.get 		= acpi_cpufreq_get,
 	.init		= acpi_cpufreq_cpu_init,
 	.exit		= acpi_cpufreq_cpu_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 15/35] cpufreq: imx6q: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/imx6q-cpufreq.c | 17 ++---------------
 1 file changed, 2 insertions(+), 15 deletions(-)

diff --git a/drivers/cpufreq/imx6q-cpufreq.c b/drivers/cpufreq/imx6q-cpufreq.c
index 81d1727..f6f877c 100644
--- a/drivers/cpufreq/imx6q-cpufreq.c
+++ b/drivers/cpufreq/imx6q-cpufreq.c
@@ -39,30 +39,17 @@ static unsigned int imx6q_get_speed(unsigned int cpu)
 	return clk_get_rate(arm_clk) / 1000;
 }
 
-static int imx6q_set_target(struct cpufreq_policy *policy,
-			    unsigned int target_freq, unsigned int relation)
+static int imx6q_set_target(struct cpufreq_policy *policy, unsigned int index)
 {
 	struct cpufreq_freqs freqs;
 	struct opp *opp;
 	unsigned long freq_hz, volt, volt_old;
-	unsigned int index;
 	int ret;
 
-	ret = cpufreq_frequency_table_target(policy, freq_table, target_freq,
-					     relation, &index);
-	if (ret) {
-		dev_err(cpu_dev, "failed to match target frequency %d: %d\n",
-			target_freq, ret);
-		return ret;
-	}
-
 	freqs.new = freq_table[index].frequency;
 	freq_hz = freqs.new * 1000;
 	freqs.old = clk_get_rate(arm_clk) / 1000;
 
-	if (freqs.old == freqs.new)
-		return 0;
-
 	rcu_read_lock();
 	opp = opp_find_freq_ceil(cpu_dev, &freq_hz);
 	if (IS_ERR(opp)) {
@@ -187,7 +174,7 @@ static int imx6q_cpufreq_init(struct cpufreq_policy *policy)
 
 static struct cpufreq_driver imx6q_cpufreq_driver = {
 	.verify = cpufreq_generic_frequency_table_verify,
-	.target = imx6q_set_target,
+	.target_index = imx6q_set_target,
 	.get = imx6q_get_speed,
 	.init = imx6q_cpufreq_init,
 	.exit = cpufreq_generic_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 16/35] cpufreq: kirkwood: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/kirkwood-cpufreq.c | 19 +++----------------
 1 file changed, 3 insertions(+), 16 deletions(-)

diff --git a/drivers/cpufreq/kirkwood-cpufreq.c b/drivers/cpufreq/kirkwood-cpufreq.c
index 9018c4d..d0d107a 100644
--- a/drivers/cpufreq/kirkwood-cpufreq.c
+++ b/drivers/cpufreq/kirkwood-cpufreq.c
@@ -55,8 +55,8 @@ static unsigned int kirkwood_cpufreq_get_cpu_frequency(unsigned int cpu)
 	return kirkwood_freq_table[0].frequency;
 }
 
-static void kirkwood_cpufreq_set_cpu_state(struct cpufreq_policy *policy,
-		unsigned int index)
+static int kirkwood_cpufreq_target(struct cpufreq_policy *policy,
+			    unsigned int index)
 {
 	struct cpufreq_freqs freqs;
 	unsigned int state = kirkwood_freq_table[index].driver_data;
@@ -100,19 +100,6 @@ static void kirkwood_cpufreq_set_cpu_state(struct cpufreq_policy *policy,
 		local_irq_enable();
 	}
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
-};
-
-static int kirkwood_cpufreq_target(struct cpufreq_policy *policy,
-			    unsigned int target_freq,
-			    unsigned int relation)
-{
-	unsigned int index = 0;
-
-	if (cpufreq_frequency_table_target(policy, kirkwood_freq_table,
-				target_freq, relation, &index))
-		return -EINVAL;
-
-	kirkwood_cpufreq_set_cpu_state(policy, index);
 
 	return 0;
 }
@@ -130,7 +117,7 @@ static int kirkwood_cpufreq_cpu_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver kirkwood_cpufreq_driver = {
 	.get	= kirkwood_cpufreq_get_cpu_frequency,
 	.verify	= cpufreq_generic_frequency_table_verify,
-	.target	= kirkwood_cpufreq_target,
+	.target_index = kirkwood_cpufreq_target,
 	.init	= kirkwood_cpufreq_cpu_init,
 	.exit	= cpufreq_generic_exit,
 	.name	= "kirkwood-cpufreq",
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 17/35] cpufreq: longhaul: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/longhaul.c | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/drivers/cpufreq/longhaul.c b/drivers/cpufreq/longhaul.c
index 57d7b02..e566776 100644
--- a/drivers/cpufreq/longhaul.c
+++ b/drivers/cpufreq/longhaul.c
@@ -626,21 +626,12 @@ static void longhaul_setup_voltagescaling(void)
 
 
 static int longhaul_target(struct cpufreq_policy *policy,
-			    unsigned int target_freq, unsigned int relation)
+			    unsigned int table_index)
 {
-	unsigned int table_index = 0;
 	unsigned int i;
 	unsigned int dir = 0;
 	u8 vid, current_vid;
 
-	if (cpufreq_frequency_table_target(policy, longhaul_table, target_freq,
-				relation, &table_index))
-		return -EINVAL;
-
-	/* Don't set same frequency again */
-	if (longhaul_index == table_index)
-		return 0;
-
 	if (!can_scale_voltage)
 		longhaul_setstate(policy, table_index);
 	else {
@@ -920,7 +911,7 @@ static int longhaul_cpu_init(struct cpufreq_policy *policy)
 
 static struct cpufreq_driver longhaul_driver = {
 	.verify	= cpufreq_generic_frequency_table_verify,
-	.target	= longhaul_target,
+	.target_index = longhaul_target,
 	.get	= longhaul_get,
 	.init	= longhaul_cpu_init,
 	.exit	= cpufreq_generic_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 18/35] cpufreq: loongson2: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: John Crispin <blogic@openwrt.org>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/loongson2_cpufreq.c | 21 +++++----------------
 1 file changed, 5 insertions(+), 16 deletions(-)

diff --git a/drivers/cpufreq/loongson2_cpufreq.c b/drivers/cpufreq/loongson2_cpufreq.c
index ed7fbe2..009c29c 100644
--- a/drivers/cpufreq/loongson2_cpufreq.c
+++ b/drivers/cpufreq/loongson2_cpufreq.c
@@ -53,11 +53,9 @@ static unsigned int loongson2_cpufreq_get(unsigned int cpu)
  * Here we notify other drivers of the proposed change and the final change.
  */
 static int loongson2_cpufreq_target(struct cpufreq_policy *policy,
-				     unsigned int target_freq,
-				     unsigned int relation)
+				     unsigned int index)
 {
 	unsigned int cpu = policy->cpu;
-	unsigned int newstate = 0;
 	cpumask_t cpus_allowed;
 	struct cpufreq_freqs freqs;
 	unsigned int freq;
@@ -65,26 +63,17 @@ static int loongson2_cpufreq_target(struct cpufreq_policy *policy,
 	cpus_allowed = current->cpus_allowed;
 	set_cpus_allowed_ptr(current, cpumask_of(cpu));
 
-	if (cpufreq_frequency_table_target
-	    (policy, &loongson2_clockmod_table[0], target_freq, relation,
-	     &newstate))
-		return -EINVAL;
-
 	freq =
 	    ((cpu_clock_freq / 1000) *
-	     loongson2_clockmod_table[newstate].driver_data) / 8;
-	if (freq < policy->min || freq > policy->max)
-		return -EINVAL;
+	     loongson2_clockmod_table[index].driver_data) / 8;
 
-	pr_debug("cpufreq: requested frequency %u Hz\n", target_freq * 1000);
+	pr_debug("cpufreq: requested frequency %u Hz\n",
+			loongson2_clockmod_table[index].frequency * 1000);
 
 	freqs.old = loongson2_cpufreq_get(cpu);
 	freqs.new = freq;
 	freqs.flags = 0;
 
-	if (freqs.new == freqs.old)
-		return 0;
-
 	/* notifiers */
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 
@@ -148,7 +137,7 @@ static struct cpufreq_driver loongson2_cpufreq_driver = {
 	.name = "loongson2",
 	.init = loongson2_cpufreq_cpu_init,
 	.verify = cpufreq_generic_frequency_table_verify,
-	.target = loongson2_cpufreq_target,
+	.target_index = loongson2_cpufreq_target,
 	.get = loongson2_cpufreq_get,
 	.exit = loongson2_cpufreq_exit,
 	.attr = cpufreq_generic_attr,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 19/35] cpufreq: maple: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/maple-cpufreq.c | 16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)

diff --git a/drivers/cpufreq/maple-cpufreq.c b/drivers/cpufreq/maple-cpufreq.c
index 7720670..1c24219 100644
--- a/drivers/cpufreq/maple-cpufreq.c
+++ b/drivers/cpufreq/maple-cpufreq.c
@@ -131,26 +131,18 @@ static int maple_scom_query_freq(void)
  */
 
 static int maple_cpufreq_target(struct cpufreq_policy *policy,
-	unsigned int target_freq, unsigned int relation)
+	unsigned int index)
 {
-	unsigned int newstate = 0;
 	struct cpufreq_freqs freqs;
 	int rc;
 
-	if (cpufreq_frequency_table_target(policy, maple_cpu_freqs,
-			target_freq, relation, &newstate))
-		return -EINVAL;
-
-	if (maple_pmode_cur == newstate)
-		return 0;
-
 	mutex_lock(&maple_switch_mutex);
 
 	freqs.old = maple_cpu_freqs[maple_pmode_cur].frequency;
-	freqs.new = maple_cpu_freqs[newstate].frequency;
+	freqs.new = maple_cpu_freqs[index].frequency;
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
-	rc = maple_scom_switch_freq(newstate);
+	rc = maple_scom_switch_freq(index);
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
 
 	mutex_unlock(&maple_switch_mutex);
@@ -181,7 +173,7 @@ static struct cpufreq_driver maple_cpufreq_driver = {
 	.flags		= CPUFREQ_CONST_LOOPS,
 	.init		= maple_cpufreq_cpu_init,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= maple_cpufreq_target,
+	.target_index	= maple_cpufreq_target,
 	.get		= maple_cpufreq_get_speed,
 	.attr		= cpufreq_generic_attr,
 };
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 20/35] cpufreq: omap: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Santosh Shilimkar <santosh.shilimkar@ti.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/omap-cpufreq.c | 31 +++----------------------------
 1 file changed, 3 insertions(+), 28 deletions(-)

diff --git a/drivers/cpufreq/omap-cpufreq.c b/drivers/cpufreq/omap-cpufreq.c
index 48020b5..69bf7d8 100644
--- a/drivers/cpufreq/omap-cpufreq.c
+++ b/drivers/cpufreq/omap-cpufreq.c
@@ -51,40 +51,15 @@ static unsigned int omap_getspeed(unsigned int cpu)
 	return rate;
 }
 
-static int omap_target(struct cpufreq_policy *policy,
-		       unsigned int target_freq,
-		       unsigned int relation)
+static int omap_target(struct cpufreq_policy *policy, unsigned int index)
 {
-	unsigned int i;
 	int r, ret = 0;
 	struct cpufreq_freqs freqs;
 	struct opp *opp;
 	unsigned long freq, volt = 0, volt_old = 0, tol = 0;
 
-	if (!freq_table) {
-		dev_err(mpu_dev, "%s: cpu%d: no freq table!\n", __func__,
-				policy->cpu);
-		return -EINVAL;
-	}
-
-	ret = cpufreq_frequency_table_target(policy, freq_table, target_freq,
-			relation, &i);
-	if (ret) {
-		dev_dbg(mpu_dev, "%s: cpu%d: no freq match for %d(ret=%d)\n",
-			__func__, policy->cpu, target_freq, ret);
-		return ret;
-	}
-	freqs.new = freq_table[i].frequency;
-	if (!freqs.new) {
-		dev_err(mpu_dev, "%s: cpu%d: no match for freq %d\n", __func__,
-			policy->cpu, target_freq);
-		return -EINVAL;
-	}
-
 	freqs.old = omap_getspeed(policy->cpu);
-
-	if (freqs.old == freqs.new && policy->cur == freqs.new)
-		return ret;
+	freqs.new = freq_table[index].frequency;
 
 	freq = freqs.new * 1000;
 	ret = clk_round_rate(mpu_clk, freq);
@@ -223,7 +198,7 @@ static int omap_cpu_exit(struct cpufreq_policy *policy)
 static struct cpufreq_driver omap_driver = {
 	.flags		= CPUFREQ_STICKY,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= omap_target,
+	.target_index	= omap_target,
 	.get		= omap_getspeed,
 	.init		= omap_cpu_init,
 	.exit		= omap_cpu_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 21/35] cpufreq: p4: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/p4-clockmod.c | 18 ++++--------------
 1 file changed, 4 insertions(+), 14 deletions(-)

diff --git a/drivers/cpufreq/p4-clockmod.c b/drivers/cpufreq/p4-clockmod.c
index 4fe6d4c..5a6263e 100644
--- a/drivers/cpufreq/p4-clockmod.c
+++ b/drivers/cpufreq/p4-clockmod.c
@@ -105,23 +105,13 @@ static struct cpufreq_frequency_table p4clockmod_table[] = {
 };
 
 
-static int cpufreq_p4_target(struct cpufreq_policy *policy,
-			     unsigned int target_freq,
-			     unsigned int relation)
+static int cpufreq_p4_target(struct cpufreq_policy *policy, unsigned int index)
 {
-	unsigned int    newstate = DC_RESV;
 	struct cpufreq_freqs freqs;
 	int i;
 
-	if (cpufreq_frequency_table_target(policy, &p4clockmod_table[0],
-				target_freq, relation, &newstate))
-		return -EINVAL;
-
 	freqs.old = cpufreq_p4_get(policy->cpu);
-	freqs.new = stock_freq * p4clockmod_table[newstate].driver_data / 8;
-
-	if (freqs.new == freqs.old)
-		return 0;
+	freqs.new = stock_freq * p4clockmod_table[index].driver_data / 8;
 
 	/* notifiers */
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
@@ -131,7 +121,7 @@ static int cpufreq_p4_target(struct cpufreq_policy *policy,
 	 * Developer's Manual, Volume 3
 	 */
 	for_each_cpu(i, policy->cpus)
-		cpufreq_p4_setdc(i, p4clockmod_table[newstate].driver_data);
+		cpufreq_p4_setdc(i, p4clockmod_table[index].driver_data);
 
 	/* notifiers */
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
@@ -256,7 +246,7 @@ static unsigned int cpufreq_p4_get(unsigned int cpu)
 
 static struct cpufreq_driver p4clockmod_driver = {
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= cpufreq_p4_target,
+	.target_index	= cpufreq_p4_target,
 	.init		= cpufreq_p4_cpu_init,
 	.exit		= cpufreq_generic_exit,
 	.get		= cpufreq_p4_get,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 22/35] cpufreq: pasemi: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/pasemi-cpufreq.c | 12 ++----------
 1 file changed, 2 insertions(+), 10 deletions(-)

diff --git a/drivers/cpufreq/pasemi-cpufreq.c b/drivers/cpufreq/pasemi-cpufreq.c
index 16f2508..aaee5fa 100644
--- a/drivers/cpufreq/pasemi-cpufreq.c
+++ b/drivers/cpufreq/pasemi-cpufreq.c
@@ -247,19 +247,11 @@ static int pas_cpufreq_cpu_exit(struct cpufreq_policy *policy)
 }
 
 static int pas_cpufreq_target(struct cpufreq_policy *policy,
-			      unsigned int target_freq,
-			      unsigned int relation)
+			      unsigned int pas_astate_new)
 {
 	struct cpufreq_freqs freqs;
-	int pas_astate_new;
 	int i;
 
-	cpufreq_frequency_table_target(policy,
-				       pas_freqs,
-				       target_freq,
-				       relation,
-				       &pas_astate_new);
-
 	freqs.old = policy->cur;
 	freqs.new = pas_freqs[pas_astate_new].frequency;
 
@@ -289,7 +281,7 @@ static struct cpufreq_driver pas_cpufreq_driver = {
 	.init		= pas_cpufreq_cpu_init,
 	.exit		= pas_cpufreq_cpu_exit,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= pas_cpufreq_target,
+	.target_index	= pas_cpufreq_target,
 	.attr		= cpufreq_generic_attr,
 };
 
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 23/35] cpufreq: pmac32: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/pmac32-cpufreq.c | 12 +++---------
 drivers/cpufreq/pmac64-cpufreq.c | 17 ++++-------------
 2 files changed, 7 insertions(+), 22 deletions(-)

diff --git a/drivers/cpufreq/pmac32-cpufreq.c b/drivers/cpufreq/pmac32-cpufreq.c
index 3f8efd2..b1ba708 100644
--- a/drivers/cpufreq/pmac32-cpufreq.c
+++ b/drivers/cpufreq/pmac32-cpufreq.c
@@ -373,17 +373,11 @@ static unsigned int pmac_cpufreq_get_speed(unsigned int cpu)
 }
 
 static int pmac_cpufreq_target(	struct cpufreq_policy *policy,
-					unsigned int target_freq,
-					unsigned int relation)
+					unsigned int index)
 {
-	unsigned int    newstate = 0;
 	int		rc;
 
-	if (cpufreq_frequency_table_target(policy, pmac_cpu_freqs,
-			target_freq, relation, &newstate))
-		return -EINVAL;
-
-	rc = do_set_cpu_speed(policy, newstate, 1);
+	rc = do_set_cpu_speed(policy, index, 1);
 
 	ppc_proc_freq = cur_freq * 1000ul;
 	return rc;
@@ -458,7 +452,7 @@ static int pmac_cpufreq_resume(struct cpufreq_policy *policy)
 
 static struct cpufreq_driver pmac_cpufreq_driver = {
 	.verify 	= cpufreq_generic_frequency_table_verify,
-	.target 	= pmac_cpufreq_target,
+	.target_index 	= pmac_cpufreq_target,
 	.get		= pmac_cpufreq_get_speed,
 	.init		= pmac_cpufreq_cpu_init,
 	.suspend	= pmac_cpufreq_suspend,
diff --git a/drivers/cpufreq/pmac64-cpufreq.c b/drivers/cpufreq/pmac64-cpufreq.c
index 0eb9313..7679990 100644
--- a/drivers/cpufreq/pmac64-cpufreq.c
+++ b/drivers/cpufreq/pmac64-cpufreq.c
@@ -311,27 +311,18 @@ static int g5_pfunc_query_freq(void)
  * Common interface to the cpufreq core
  */
 
-static int g5_cpufreq_target(struct cpufreq_policy *policy,
-	unsigned int target_freq, unsigned int relation)
+static int g5_cpufreq_target(struct cpufreq_policy *policy, unsigned int index)
 {
-	unsigned int newstate = 0;
 	struct cpufreq_freqs freqs;
 	int rc;
 
-	if (cpufreq_frequency_table_target(policy, g5_cpu_freqs,
-			target_freq, relation, &newstate))
-		return -EINVAL;
-
-	if (g5_pmode_cur == newstate)
-		return 0;
-
 	mutex_lock(&g5_switch_mutex);
 
 	freqs.old = g5_cpu_freqs[g5_pmode_cur].frequency;
-	freqs.new = g5_cpu_freqs[newstate].frequency;
+	freqs.new = g5_cpu_freqs[index].frequency;
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
-	rc = g5_switch_freq(newstate);
+	rc = g5_switch_freq(index);
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
 
 	mutex_unlock(&g5_switch_mutex);
@@ -362,7 +353,7 @@ static struct cpufreq_driver g5_cpufreq_driver = {
 	.flags		= CPUFREQ_CONST_LOOPS,
 	.init		= g5_cpufreq_cpu_init,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= g5_cpufreq_target,
+	.target_index	= g5_cpufreq_target,
 	.get		= g5_cpufreq_get_speed,
 	.attr 		= cpufreq_generic_attr,
 };
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 24/35] cpufreq: powernow: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/powernow-k6.c | 35 +++++------------------------------
 drivers/cpufreq/powernow-k7.c | 22 ++++------------------
 drivers/cpufreq/powernow-k8.c | 24 ++++++++----------------
 3 files changed, 17 insertions(+), 64 deletions(-)

diff --git a/drivers/cpufreq/powernow-k6.c b/drivers/cpufreq/powernow-k6.c
index ff05d28..cb19fb8 100644
--- a/drivers/cpufreq/powernow-k6.c
+++ b/drivers/cpufreq/powernow-k6.c
@@ -63,12 +63,12 @@ static int powernow_k6_get_cpu_multiplier(void)
 
 
 /**
- * powernow_k6_set_state - set the PowerNow! multiplier
+ * powernow_k6_target - set the PowerNow! multiplier
  * @best_i: clock_ratio[best_i] is the target multiplier
  *
  *   Tries to change the PowerNow! multiplier
  */
-static void powernow_k6_set_state(struct cpufreq_policy *policy,
+static int powernow_k6_target(struct cpufreq_policy *policy,
 		unsigned int best_i)
 {
 	unsigned long outvalue = 0, invalue = 0;
@@ -77,7 +77,7 @@ static void powernow_k6_set_state(struct cpufreq_policy *policy,
 
 	if (clock_ratio[best_i].driver_data > max_multiplier) {
 		printk(KERN_ERR PFX "invalid target frequency\n");
-		return;
+		return -EINVAL;
 	}
 
 	freqs.old = busfreq * powernow_k6_get_cpu_multiplier();
@@ -100,31 +100,6 @@ static void powernow_k6_set_state(struct cpufreq_policy *policy,
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
 
-	return;
-}
-
-
-/**
- * powernow_k6_setpolicy - sets a new CPUFreq policy
- * @policy: new policy
- * @target_freq: the target frequency
- * @relation: how that frequency relates to achieved frequency
- *  (CPUFREQ_RELATION_L or CPUFREQ_RELATION_H)
- *
- * sets a new CPUFreq policy
- */
-static int powernow_k6_target(struct cpufreq_policy *policy,
-			       unsigned int target_freq,
-			       unsigned int relation)
-{
-	unsigned int newstate = 0;
-
-	if (cpufreq_frequency_table_target(policy, &clock_ratio[0],
-				target_freq, relation, &newstate))
-		return -EINVAL;
-
-	powernow_k6_set_state(policy, newstate);
-
 	return 0;
 }
 
@@ -162,7 +137,7 @@ static int powernow_k6_cpu_exit(struct cpufreq_policy *policy)
 	unsigned int i;
 	for (i = 0; i < 8; i++) {
 		if (i == max_multiplier)
-			powernow_k6_set_state(policy, i);
+			powernow_k6_target(policy, i);
 	}
 	cpufreq_frequency_table_put_attr(policy->cpu);
 	return 0;
@@ -177,7 +152,7 @@ static unsigned int powernow_k6_get(unsigned int cpu)
 
 static struct cpufreq_driver powernow_k6_driver = {
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= powernow_k6_target,
+	.target_index	= powernow_k6_target,
 	.init		= powernow_k6_cpu_init,
 	.exit		= powernow_k6_cpu_exit,
 	.get		= powernow_k6_get,
diff --git a/drivers/cpufreq/powernow-k7.c b/drivers/cpufreq/powernow-k7.c
index 14cd98f..7c76f03 100644
--- a/drivers/cpufreq/powernow-k7.c
+++ b/drivers/cpufreq/powernow-k7.c
@@ -248,7 +248,7 @@ static void change_VID(int vid)
 }
 
 
-static void change_speed(struct cpufreq_policy *policy, unsigned int index)
+static int powernow_target(struct cpufreq_policy *policy, unsigned int index)
 {
 	u8 fid, vid;
 	struct cpufreq_freqs freqs;
@@ -291,6 +291,8 @@ static void change_speed(struct cpufreq_policy *policy, unsigned int index)
 		local_irq_enable();
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
+
+	return 0;
 }
 
 
@@ -533,22 +535,6 @@ static int powernow_decode_bios(int maxfid, int startvid)
 }
 
 
-static int powernow_target(struct cpufreq_policy *policy,
-			    unsigned int target_freq,
-			    unsigned int relation)
-{
-	unsigned int newstate;
-
-	if (cpufreq_frequency_table_target(policy, powernow_table, target_freq,
-				relation, &newstate))
-		return -EINVAL;
-
-	change_speed(policy, newstate);
-
-	return 0;
-}
-
-
 /*
  * We use the fact that the bus frequency is somehow
  * a multiple of 100000/3 khz, then we compute sgtc according
@@ -696,7 +682,7 @@ static int powernow_cpu_exit(struct cpufreq_policy *policy)
 
 static struct cpufreq_driver powernow_driver = {
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= powernow_target,
+	.target_index	= powernow_target,
 	.get		= powernow_get,
 #ifdef CONFIG_X86_POWERNOW_K7_ACPI
 	.bios_limit	= acpi_processor_get_bios_limit,
diff --git a/drivers/cpufreq/powernow-k8.c b/drivers/cpufreq/powernow-k8.c
index 1e6f68a..80bc606 100644
--- a/drivers/cpufreq/powernow-k8.c
+++ b/drivers/cpufreq/powernow-k8.c
@@ -977,20 +977,17 @@ static int transition_frequency_fidvid(struct powernow_k8_data *data,
 
 struct powernowk8_target_arg {
 	struct cpufreq_policy		*pol;
-	unsigned			targfreq;
-	unsigned			relation;
+	unsigned			newstate;
 };
 
 static long powernowk8_target_fn(void *arg)
 {
 	struct powernowk8_target_arg *pta = arg;
 	struct cpufreq_policy *pol = pta->pol;
-	unsigned targfreq = pta->targfreq;
-	unsigned relation = pta->relation;
+	unsigned newstate = pta->newstate;
 	struct powernow_k8_data *data = per_cpu(powernow_data, pol->cpu);
 	u32 checkfid;
 	u32 checkvid;
-	unsigned int newstate;
 	int ret;
 
 	if (!data)
@@ -1004,8 +1001,9 @@ static long powernowk8_target_fn(void *arg)
 		return -EIO;
 	}
 
-	pr_debug("targ: cpu %d, %d kHz, min %d, max %d, relation %d\n",
-		pol->cpu, targfreq, pol->min, pol->max, relation);
+	pr_debug("targ: cpu %d, %d kHz, min %d, max %d\n",
+		pol->cpu, data->powernow_table[newstate].frequency, pol->min,
+		pol->max);
 
 	if (query_current_values_with_pending_wait(data))
 		return -EIO;
@@ -1021,10 +1019,6 @@ static long powernowk8_target_fn(void *arg)
 		       checkvid, data->currvid);
 	}
 
-	if (cpufreq_frequency_table_target(pol, data->powernow_table,
-				targfreq, relation, &newstate))
-		return -EIO;
-
 	mutex_lock(&fidvid_mutex);
 
 	powernow_k8_acpi_pst_values(data, newstate);
@@ -1044,11 +1038,9 @@ static long powernowk8_target_fn(void *arg)
 }
 
 /* Driver entry point to switch to the target frequency */
-static int powernowk8_target(struct cpufreq_policy *pol,
-		unsigned targfreq, unsigned relation)
+static int powernowk8_target(struct cpufreq_policy *pol, unsigned index)
 {
-	struct powernowk8_target_arg pta = { .pol = pol, .targfreq = targfreq,
-					     .relation = relation };
+	struct powernowk8_target_arg pta = { .pol = pol, .newstate = index };
 
 	return work_on_cpu(pol->cpu, powernowk8_target_fn, &pta);
 }
@@ -1216,7 +1208,7 @@ out:
 
 static struct cpufreq_driver cpufreq_amd64_driver = {
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= powernowk8_target,
+	.target_index	= powernowk8_target,
 	.bios_limit	= acpi_processor_get_bios_limit,
 	.init		= powernowk8_cpu_init,
 	.exit		= powernowk8_cpu_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 25/35] cpufreq: ppc: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/ppc-corenet-cpufreq.c | 15 ++++-----------
 drivers/cpufreq/ppc_cbe_cpufreq.c     | 12 ++----------
 2 files changed, 6 insertions(+), 21 deletions(-)

diff --git a/drivers/cpufreq/ppc-corenet-cpufreq.c b/drivers/cpufreq/ppc-corenet-cpufreq.c
index befd489..243a396 100644
--- a/drivers/cpufreq/ppc-corenet-cpufreq.c
+++ b/drivers/cpufreq/ppc-corenet-cpufreq.c
@@ -253,27 +253,20 @@ static int __exit corenet_cpufreq_cpu_exit(struct cpufreq_policy *policy)
 }
 
 static int corenet_cpufreq_target(struct cpufreq_policy *policy,
-		unsigned int target_freq, unsigned int relation)
+		unsigned int index)
 {
 	struct cpufreq_freqs freqs;
-	unsigned int new;
 	struct clk *parent;
 	int ret;
 	struct cpu_data *data = per_cpu(cpu_data, policy->cpu);
 
-	cpufreq_frequency_table_target(policy, data->table,
-			target_freq, relation, &new);
-
-	if (policy->cur == data->table[new].frequency)
-		return 0;
-
 	freqs.old = policy->cur;
-	freqs.new = data->table[new].frequency;
+	freqs.new = data->table[index].frequency;
 
 	mutex_lock(&cpufreq_lock);
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 
-	parent = of_clk_get(data->parent, data->table[new].driver_data);
+	parent = of_clk_get(data->parent, data->table[index].driver_data);
 	ret = clk_set_parent(data->clk, parent);
 	if (ret)
 		freqs.new = freqs.old;
@@ -290,7 +283,7 @@ static struct cpufreq_driver ppc_corenet_cpufreq_driver = {
 	.init		= corenet_cpufreq_cpu_init,
 	.exit		= __exit_p(corenet_cpufreq_cpu_exit),
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= corenet_cpufreq_target,
+	.target_index	= corenet_cpufreq_target,
 	.get		= corenet_cpufreq_get_speed,
 	.attr		= cpufreq_generic_attr,
 };
diff --git a/drivers/cpufreq/ppc_cbe_cpufreq.c b/drivers/cpufreq/ppc_cbe_cpufreq.c
index 38540d1..52f707d 100644
--- a/drivers/cpufreq/ppc_cbe_cpufreq.c
+++ b/drivers/cpufreq/ppc_cbe_cpufreq.c
@@ -129,18 +129,10 @@ static int cbe_cpufreq_cpu_init(struct cpufreq_policy *policy)
 }
 
 static int cbe_cpufreq_target(struct cpufreq_policy *policy,
-			      unsigned int target_freq,
-			      unsigned int relation)
+			      unsigned int cbe_pmode_new)
 {
 	int rc;
 	struct cpufreq_freqs freqs;
-	unsigned int cbe_pmode_new;
-
-	cpufreq_frequency_table_target(policy,
-				       cbe_freqs,
-				       target_freq,
-				       relation,
-				       &cbe_pmode_new);
 
 	freqs.old = policy->cur;
 	freqs.new = cbe_freqs[cbe_pmode_new].frequency;
@@ -164,7 +156,7 @@ static int cbe_cpufreq_target(struct cpufreq_policy *policy,
 
 static struct cpufreq_driver cbe_cpufreq_driver = {
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= cbe_cpufreq_target,
+	.target_index	= cbe_cpufreq_target,
 	.init		= cbe_cpufreq_cpu_init,
 	.exit		= cpufreq_generic_exit,
 	.name		= "cbe-cpufreq",
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 26/35] cpufreq: pxa: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Eric Miao <eric.y.miao@gmail.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/pxa2xx-cpufreq.c | 13 ++-----------
 drivers/cpufreq/pxa3xx-cpufreq.c | 17 +++--------------
 2 files changed, 5 insertions(+), 25 deletions(-)

diff --git a/drivers/cpufreq/pxa2xx-cpufreq.c b/drivers/cpufreq/pxa2xx-cpufreq.c
index 5a72bf3..17326e1 100644
--- a/drivers/cpufreq/pxa2xx-cpufreq.c
+++ b/drivers/cpufreq/pxa2xx-cpufreq.c
@@ -267,14 +267,11 @@ static unsigned int pxa_cpufreq_get(unsigned int cpu)
 	return get_clk_frequency_khz(0);
 }
 
-static int pxa_set_target(struct cpufreq_policy *policy,
-			  unsigned int target_freq,
-			  unsigned int relation)
+static int pxa_set_target(struct cpufreq_policy *policy, unsigned int idx)
 {
 	struct cpufreq_frequency_table *pxa_freqs_table;
 	pxa_freqs_t *pxa_freq_settings;
 	struct cpufreq_freqs freqs;
-	unsigned int idx;
 	unsigned long flags;
 	unsigned int new_freq_cpu, new_freq_mem;
 	unsigned int unused, preset_mdrefr, postset_mdrefr, cclkcfg;
@@ -283,12 +280,6 @@ static int pxa_set_target(struct cpufreq_policy *policy,
 	/* Get the current policy */
 	find_freq_tables(&pxa_freqs_table, &pxa_freq_settings);
 
-	/* Lookup the next frequency */
-	if (cpufreq_frequency_table_target(policy, pxa_freqs_table,
-					   target_freq, relation, &idx)) {
-		return -EINVAL;
-	}
-
 	new_freq_cpu = pxa_freq_settings[idx].khz;
 	new_freq_mem = pxa_freq_settings[idx].membus;
 	freqs.old = policy->cur;
@@ -450,7 +441,7 @@ static int pxa_cpufreq_init(struct cpufreq_policy *policy)
 
 static struct cpufreq_driver pxa_cpufreq_driver = {
 	.verify	= cpufreq_generic_frequency_table_verify,
-	.target	= pxa_set_target,
+	.target_index = pxa_set_target,
 	.init	= pxa_cpufreq_init,
 	.exit	= cpufreq_generic_exit,
 	.get	= pxa_cpufreq_get,
diff --git a/drivers/cpufreq/pxa3xx-cpufreq.c b/drivers/cpufreq/pxa3xx-cpufreq.c
index 2837fd6..66e8cd0 100644
--- a/drivers/cpufreq/pxa3xx-cpufreq.c
+++ b/drivers/cpufreq/pxa3xx-cpufreq.c
@@ -155,24 +155,16 @@ static unsigned int pxa3xx_cpufreq_get(unsigned int cpu)
 	return pxa3xx_get_clk_frequency_khz(0);
 }
 
-static int pxa3xx_cpufreq_set(struct cpufreq_policy *policy,
-			      unsigned int target_freq,
-			      unsigned int relation)
+static int pxa3xx_cpufreq_set(struct cpufreq_policy *policy, unsigned int index)
 {
 	struct pxa3xx_freq_info *next;
 	struct cpufreq_freqs freqs;
 	unsigned long flags;
-	int idx;
 
 	if (policy->cpu != 0)
 		return -EINVAL;
 
-	/* Lookup the next frequency */
-	if (cpufreq_frequency_table_target(policy, pxa3xx_freqs_table,
-				target_freq, relation, &idx))
-		return -EINVAL;
-
-	next = &pxa3xx_freqs[idx];
+	next = &pxa3xx_freqs[index];
 
 	freqs.old = policy->cur;
 	freqs.new = next->cpufreq_mhz * 1000;
@@ -181,9 +173,6 @@ static int pxa3xx_cpufreq_set(struct cpufreq_policy *policy,
 			freqs.old / 1000, freqs.new / 1000,
 			(freqs.old == freqs.new) ? " (skipped)" : "");
 
-	if (freqs.old == target_freq)
-		return 0;
-
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 
 	local_irq_save(flags);
@@ -224,7 +213,7 @@ static int pxa3xx_cpufreq_init(struct cpufreq_policy *policy)
 
 static struct cpufreq_driver pxa3xx_cpufreq_driver = {
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= pxa3xx_cpufreq_set,
+	.target_index	= pxa3xx_cpufreq_set,
 	.init		= pxa3xx_cpufreq_init,
 	.exit		= cpufreq_generic_exit,
 	.get		= pxa3xx_cpufreq_get,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 27/35] cpufreq: s3c2416: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/s3c2416-cpufreq.c | 17 ++++-------------
 1 file changed, 4 insertions(+), 13 deletions(-)

diff --git a/drivers/cpufreq/s3c2416-cpufreq.c b/drivers/cpufreq/s3c2416-cpufreq.c
index 8c57f10..90efbf5 100644
--- a/drivers/cpufreq/s3c2416-cpufreq.c
+++ b/drivers/cpufreq/s3c2416-cpufreq.c
@@ -217,24 +217,15 @@ static int s3c2416_cpufreq_leave_dvs(struct s3c2416_data *s3c_freq, int idx)
 }
 
 static int s3c2416_cpufreq_set_target(struct cpufreq_policy *policy,
-				      unsigned int target_freq,
-				      unsigned int relation)
+				      unsigned int index)
 {
 	struct s3c2416_data *s3c_freq = &s3c2416_cpufreq;
 	struct cpufreq_freqs freqs;
 	int idx, ret, to_dvs = 0;
-	unsigned int i;
 
 	mutex_lock(&cpufreq_lock);
 
-	pr_debug("cpufreq: to %dKHz, relation %d\n", target_freq, relation);
-
-	ret = cpufreq_frequency_table_target(policy, s3c_freq->freq_table,
-					     target_freq, relation, &i);
-	if (ret != 0)
-		goto out;
-
-	idx = s3c_freq->freq_table[i].driver_data;
+	idx = s3c_freq->freq_table[index].driver_data;
 
 	if (idx == SOURCE_HCLK)
 		to_dvs = 1;
@@ -256,7 +247,7 @@ static int s3c2416_cpufreq_set_target(struct cpufreq_policy *policy,
 	 */
 	freqs.new = (s3c_freq->is_dvs && !to_dvs)
 				? clk_get_rate(s3c_freq->hclk) / 1000
-				: s3c_freq->freq_table[i].frequency;
+				: s3c_freq->freq_table[index].frequency;
 
 	pr_debug("cpufreq: Transition %d-%dkHz\n", freqs.old, freqs.new);
 
@@ -509,7 +500,7 @@ err_hclk:
 static struct cpufreq_driver s3c2416_cpufreq_driver = {
 	.flags          = 0,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= s3c2416_cpufreq_set_target,
+	.target_index	= s3c2416_cpufreq_set_target,
 	.get		= s3c2416_cpufreq_get_speed,
 	.init		= s3c2416_cpufreq_driver_init,
 	.name		= "s3c2416",
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 28/35] cpufreq: s3c64xx: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/s3c64xx-cpufreq.c | 18 ++++--------------
 1 file changed, 4 insertions(+), 14 deletions(-)

diff --git a/drivers/cpufreq/s3c64xx-cpufreq.c b/drivers/cpufreq/s3c64xx-cpufreq.c
index 99fbc49..72f733b 100644
--- a/drivers/cpufreq/s3c64xx-cpufreq.c
+++ b/drivers/cpufreq/s3c64xx-cpufreq.c
@@ -63,26 +63,16 @@ static unsigned int s3c64xx_cpufreq_get_speed(unsigned int cpu)
 }
 
 static int s3c64xx_cpufreq_set_target(struct cpufreq_policy *policy,
-				      unsigned int target_freq,
-				      unsigned int relation)
+				      unsigned int index)
 {
 	int ret;
-	unsigned int i;
 	struct cpufreq_freqs freqs;
 	struct s3c64xx_dvfs *dvfs;
 
-	ret = cpufreq_frequency_table_target(policy, s3c64xx_freq_table,
-					     target_freq, relation, &i);
-	if (ret != 0)
-		return ret;
-
 	freqs.old = clk_get_rate(armclk) / 1000;
-	freqs.new = s3c64xx_freq_table[i].frequency;
+	freqs.new = s3c64xx_freq_table[index].frequency;
 	freqs.flags = 0;
-	dvfs = &s3c64xx_dvfs_table[s3c64xx_freq_table[i].driver_data];
-
-	if (freqs.old == freqs.new)
-		return 0;
+	dvfs = &s3c64xx_dvfs_table[s3c64xx_freq_table[index].driver_data];
 
 	pr_debug("Transition %d-%dkHz\n", freqs.old, freqs.new);
 
@@ -257,7 +247,7 @@ static int s3c64xx_cpufreq_driver_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver s3c64xx_cpufreq_driver = {
 	.flags          = 0,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= s3c64xx_cpufreq_set_target,
+	.target_index	= s3c64xx_cpufreq_set_target,
 	.get		= s3c64xx_cpufreq_get_speed,
 	.init		= s3c64xx_cpufreq_driver_init,
 	.name		= "s3c",
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 29/35] cpufreq: s5pv210: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

This drivers wasn't as straight forward as other ones. It was doing some funny
stuff to disable driver while going into suspend mode. This part is simplified
as well to get this converted to ->target_index().

Cc: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/s5pv210-cpufreq.c | 54 ++++++++++-----------------------------
 1 file changed, 14 insertions(+), 40 deletions(-)

diff --git a/drivers/cpufreq/s5pv210-cpufreq.c b/drivers/cpufreq/s5pv210-cpufreq.c
index 0eafc52..bdffc3d 100644
--- a/drivers/cpufreq/s5pv210-cpufreq.c
+++ b/drivers/cpufreq/s5pv210-cpufreq.c
@@ -36,16 +36,7 @@ static DEFINE_MUTEX(set_freq_lock);
 /* Use 800MHz when entering sleep mode */
 #define SLEEP_FREQ	(800 * 1000)
 
-/*
- * relation has an additional symantics other than the standard of cpufreq
- * DISALBE_FURTHER_CPUFREQ: disable further access to target
- * ENABLE_FURTUER_CPUFREQ: enable access to target
- */
-enum cpufreq_access {
-	DISABLE_FURTHER_CPUFREQ = 0x10,
-	ENABLE_FURTHER_CPUFREQ = 0x20,
-};
-
+/* Tracks if cpu freqency can be updated anymore */
 static bool no_cpufreq_access;
 
 /*
@@ -182,12 +173,10 @@ static unsigned int s5pv210_getspeed(unsigned int cpu)
 	return clk_get_rate(cpu_clk) / 1000;
 }
 
-static int s5pv210_target(struct cpufreq_policy *policy,
-			  unsigned int target_freq,
-			  unsigned int relation)
+static int s5pv210_target(struct cpufreq_policy *policy, unsigned int index)
 {
 	unsigned long reg;
-	unsigned int index, priv_index;
+	unsigned int priv_index;
 	unsigned int pll_changing = 0;
 	unsigned int bus_speed_changing = 0;
 	int arm_volt, int_volt;
@@ -195,9 +184,6 @@ static int s5pv210_target(struct cpufreq_policy *policy,
 
 	mutex_lock(&set_freq_lock);
 
-	if (relation & ENABLE_FURTHER_CPUFREQ)
-		no_cpufreq_access = false;
-
 	if (no_cpufreq_access) {
 #ifdef CONFIG_PM_VERBOSE
 		pr_err("%s:%d denied access to %s as it is disabled"
@@ -207,27 +193,13 @@ static int s5pv210_target(struct cpufreq_policy *policy,
 		goto exit;
 	}
 
-	if (relation & DISABLE_FURTHER_CPUFREQ)
-		no_cpufreq_access = true;
-
-	relation &= ~(ENABLE_FURTHER_CPUFREQ | DISABLE_FURTHER_CPUFREQ);
-
 	freqs.old = s5pv210_getspeed(0);
-
-	if (cpufreq_frequency_table_target(policy, s5pv210_freq_table,
-					   target_freq, relation, &index)) {
-		ret = -EINVAL;
-		goto exit;
-	}
-
 	freqs.new = s5pv210_freq_table[index].frequency;
 
-	if (freqs.new == freqs.old)
-		goto exit;
-
 	/* Finding current running level index */
 	if (cpufreq_frequency_table_target(policy, s5pv210_freq_table,
-					   freqs.old, relation, &priv_index)) {
+					   freqs.old, CPUFREQ_RELATION_H,
+					   &priv_index)) {
 		ret = -EINVAL;
 		goto exit;
 	}
@@ -563,16 +535,18 @@ static int s5pv210_cpufreq_notifier_event(struct notifier_block *this,
 
 	switch (event) {
 	case PM_SUSPEND_PREPARE:
-		ret = cpufreq_driver_target(cpufreq_cpu_get(0), SLEEP_FREQ,
-					    DISABLE_FURTHER_CPUFREQ);
+		ret = cpufreq_driver_target(cpufreq_cpu_get(0), SLEEP_FREQ, 0);
 		if (ret < 0)
 			return NOTIFY_BAD;
 
+		/* Disable updation of cpu frequency */
+		no_cpufreq_access = true;
 		return NOTIFY_OK;
 	case PM_POST_RESTORE:
 	case PM_POST_SUSPEND:
-		cpufreq_driver_target(cpufreq_cpu_get(0), SLEEP_FREQ,
-				      ENABLE_FURTHER_CPUFREQ);
+		/* Enable updation of cpu frequency */
+		no_cpufreq_access = false;
+		cpufreq_driver_target(cpufreq_cpu_get(0), SLEEP_FREQ, 0);
 
 		return NOTIFY_OK;
 	}
@@ -585,18 +559,18 @@ static int s5pv210_cpufreq_reboot_notifier_event(struct notifier_block *this,
 {
 	int ret;
 
-	ret = cpufreq_driver_target(cpufreq_cpu_get(0), SLEEP_FREQ,
-				    DISABLE_FURTHER_CPUFREQ);
+	ret = cpufreq_driver_target(cpufreq_cpu_get(0), SLEEP_FREQ, 0);
 	if (ret < 0)
 		return NOTIFY_BAD;
 
+	no_cpufreq_access = true;
 	return NOTIFY_DONE;
 }
 
 static struct cpufreq_driver s5pv210_driver = {
 	.flags		= CPUFREQ_STICKY,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= s5pv210_target,
+	.target_index	= s5pv210_target,
 	.get		= s5pv210_getspeed,
 	.init		= s5pv210_cpu_init,
 	.name		= "s5pv210",
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 30/35] cpufreq: sa11x0: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Some existing routines are no more required and so are removed now.

Cc: Russell King <linux@arm.linux.org.uk>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 arch/arm/mach-sa1100/generic.c   | 20 --------------------
 arch/arm/mach-sa1100/generic.h   |  2 --
 drivers/cpufreq/sa1100-cpufreq.c | 24 ++++--------------------
 drivers/cpufreq/sa1110-cpufreq.c | 26 ++++----------------------
 4 files changed, 8 insertions(+), 64 deletions(-)

diff --git a/arch/arm/mach-sa1100/generic.c b/arch/arm/mach-sa1100/generic.c
index cb4b2ca..d4ea142 100644
--- a/arch/arm/mach-sa1100/generic.c
+++ b/arch/arm/mach-sa1100/generic.c
@@ -62,26 +62,6 @@ struct cpufreq_frequency_table sa11x0_freq_table[NR_FREQS+1] = {
 	{ .frequency = CPUFREQ_TABLE_END, },
 };
 
-/* rounds up(!)  */
-unsigned int sa11x0_freq_to_ppcr(unsigned int khz)
-{
-	int i;
-
-	for (i = 0; i < NR_FREQS; i++)
-		if (sa11x0_freq_table[i].frequency >= khz)
-			break;
-
-	return i;
-}
-
-unsigned int sa11x0_ppcr_to_freq(unsigned int idx)
-{
-	unsigned int freq = 0;
-	if (idx < NR_FREQS)
-		freq = sa11x0_freq_table[idx].frequency;
-	return freq;
-}
-
 unsigned int sa11x0_getspeed(unsigned int cpu)
 {
 	if (cpu)
diff --git a/arch/arm/mach-sa1100/generic.h b/arch/arm/mach-sa1100/generic.h
index 39d56a67..84505d5 100644
--- a/arch/arm/mach-sa1100/generic.h
+++ b/arch/arm/mach-sa1100/generic.h
@@ -22,9 +22,7 @@ extern void sa1110_mb_disable(void);
 struct cpufreq_policy;
 
 extern struct cpufreq_frequency_table sa11x0_freq_table[];
-extern unsigned int sa11x0_freq_to_ppcr(unsigned int khz);
 extern unsigned int sa11x0_getspeed(unsigned int cpu);
-extern unsigned int sa11x0_ppcr_to_freq(unsigned int idx);
 
 struct flash_platform_data;
 struct resource;
diff --git a/drivers/cpufreq/sa1100-cpufreq.c b/drivers/cpufreq/sa1100-cpufreq.c
index 02fe8b2..7e42bce 100644
--- a/drivers/cpufreq/sa1100-cpufreq.c
+++ b/drivers/cpufreq/sa1100-cpufreq.c
@@ -177,36 +177,20 @@ static void sa1100_update_dram_timings(int current_speed, int new_speed)
 	}
 }
 
-static int sa1100_target(struct cpufreq_policy *policy,
-			 unsigned int target_freq,
-			 unsigned int relation)
+static int sa1100_target(struct cpufreq_policy *policy, unsigned int ppcr)
 {
 	unsigned int cur = sa11x0_getspeed(0);
-	unsigned int new_ppcr;
 	struct cpufreq_freqs freqs;
 
-	new_ppcr = sa11x0_freq_to_ppcr(target_freq);
-	switch (relation) {
-	case CPUFREQ_RELATION_L:
-		if (sa11x0_ppcr_to_freq(new_ppcr) > policy->max)
-			new_ppcr--;
-		break;
-	case CPUFREQ_RELATION_H:
-		if ((sa11x0_ppcr_to_freq(new_ppcr) > target_freq) &&
-		    (sa11x0_ppcr_to_freq(new_ppcr - 1) >= policy->min))
-			new_ppcr--;
-		break;
-	}
-
 	freqs.old = cur;
-	freqs.new = sa11x0_ppcr_to_freq(new_ppcr);
+	freqs.new = sa11x0_freq_table[ppcr].frequency;
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 
 	if (freqs.new > cur)
 		sa1100_update_dram_timings(cur, freqs.new);
 
-	PPCR = new_ppcr;
+	PPCR = ppcr;
 
 	if (freqs.new < cur)
 		sa1100_update_dram_timings(cur, freqs.new);
@@ -229,7 +213,7 @@ static int __init sa1100_cpu_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver sa1100_driver __refdata = {
 	.flags		= CPUFREQ_STICKY,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= sa1100_target,
+	.target_index	= sa1100_target,
 	.get		= sa11x0_getspeed,
 	.init		= sa1100_cpu_init,
 	.name		= "sa1100",
diff --git a/drivers/cpufreq/sa1110-cpufreq.c b/drivers/cpufreq/sa1110-cpufreq.c
index a38d904..fb69553 100644
--- a/drivers/cpufreq/sa1110-cpufreq.c
+++ b/drivers/cpufreq/sa1110-cpufreq.c
@@ -229,34 +229,16 @@ sdram_update_refresh(u_int cpu_khz, struct sdram_params *sdram)
 /*
  * Ok, set the CPU frequency.
  */
-static int sa1110_target(struct cpufreq_policy *policy,
-			 unsigned int target_freq,
-			 unsigned int relation)
+static int sa1110_target(struct cpufreq_policy *policy, unsigned int ppcr)
 {
 	struct sdram_params *sdram = &sdram_params;
 	struct cpufreq_freqs freqs;
 	struct sdram_info sd;
 	unsigned long flags;
-	unsigned int ppcr, unused;
-
-	switch (relation) {
-	case CPUFREQ_RELATION_L:
-		ppcr = sa11x0_freq_to_ppcr(target_freq);
-		if (sa11x0_ppcr_to_freq(ppcr) > policy->max)
-			ppcr--;
-		break;
-	case CPUFREQ_RELATION_H:
-		ppcr = sa11x0_freq_to_ppcr(target_freq);
-		if (ppcr && (sa11x0_ppcr_to_freq(ppcr) > target_freq) &&
-		    (sa11x0_ppcr_to_freq(ppcr-1) >= policy->min))
-			ppcr--;
-		break;
-	default:
-		return -EINVAL;
-	}
+	unsigned int unused;
 
 	freqs.old = sa11x0_getspeed(0);
-	freqs.new = sa11x0_ppcr_to_freq(ppcr);
+	freqs.new = sa11x0_freq_table[ppcr].frequency;
 
 	sdram_calculate_timing(&sd, freqs.new, sdram);
 
@@ -345,7 +327,7 @@ static int __init sa1110_cpu_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver sa1110_driver __refdata = {
 	.flags		= CPUFREQ_STICKY,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= sa1110_target,
+	.target_index	= sa1110_target,
 	.get		= sa11x0_getspeed,
 	.init		= sa1110_cpu_init,
 	.name		= "sa1110",
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 31/35] cpufreq: sc520: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/sc520_freq.c | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)

diff --git a/drivers/cpufreq/sc520_freq.c b/drivers/cpufreq/sc520_freq.c
index 8556225..cd62472 100644
--- a/drivers/cpufreq/sc520_freq.c
+++ b/drivers/cpufreq/sc520_freq.c
@@ -53,8 +53,7 @@ static unsigned int sc520_freq_get_cpu_frequency(unsigned int cpu)
 	}
 }
 
-static void sc520_freq_set_cpu_state(struct cpufreq_policy *policy,
-		unsigned int state)
+static int sc520_freq_target(struct cpufreq_policy *policy, unsigned int state)
 {
 
 	struct cpufreq_freqs	freqs;
@@ -76,24 +75,10 @@ static void sc520_freq_set_cpu_state(struct cpufreq_policy *policy,
 	local_irq_enable();
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
-};
-
-static int sc520_freq_target(struct cpufreq_policy *policy,
-			    unsigned int target_freq,
-			    unsigned int relation)
-{
-	unsigned int newstate = 0;
-
-	if (cpufreq_frequency_table_target(policy, sc520_freq_table,
-				target_freq, relation, &newstate))
-		return -EINVAL;
-
-	sc520_freq_set_cpu_state(policy, newstate);
 
 	return 0;
 }
 
-
 /*
  *	Module init and exit code
  */
@@ -118,7 +103,7 @@ static int sc520_freq_cpu_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver sc520_freq_driver = {
 	.get	= sc520_freq_get_cpu_frequency,
 	.verify	= cpufreq_generic_frequency_table_verify,
-	.target	= sc520_freq_target,
+	.target_index = sc520_freq_target,
 	.init	= sc520_freq_cpu_init,
 	.exit	= cpufreq_generic_exit,
 	.name	= "sc520_freq",
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 32/35] cpufreq: sparc: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: David S. Miller <davem@davemloft.net>
Cc: sparclinux@vger.kernel.org
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/sparc-us2e-cpufreq.c | 21 +++------------------
 drivers/cpufreq/sparc-us3-cpufreq.c  | 23 +++--------------------
 2 files changed, 6 insertions(+), 38 deletions(-)

diff --git a/drivers/cpufreq/sparc-us2e-cpufreq.c b/drivers/cpufreq/sparc-us2e-cpufreq.c
index 291688c..3bf5b8f 100644
--- a/drivers/cpufreq/sparc-us2e-cpufreq.c
+++ b/drivers/cpufreq/sparc-us2e-cpufreq.c
@@ -245,8 +245,7 @@ static unsigned int us2e_freq_get(unsigned int cpu)
 	return clock_tick / estar_to_divisor(estar);
 }
 
-static void us2e_set_cpu_divider_index(struct cpufreq_policy *policy,
-		unsigned int index)
+static int us2e_freq_target(struct cpufreq_policy *policy, unsigned int index)
 {
 	unsigned int cpu = policy->cpu;
 	unsigned long new_bits, new_freq;
@@ -277,20 +276,6 @@ static void us2e_set_cpu_divider_index(struct cpufreq_policy *policy,
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
 
 	set_cpus_allowed_ptr(current, &cpus_allowed);
-}
-
-static int us2e_freq_target(struct cpufreq_policy *policy,
-			  unsigned int target_freq,
-			  unsigned int relation)
-{
-	unsigned int new_index = 0;
-
-	if (cpufreq_frequency_table_target(policy,
-					   &us2e_freq_table[policy->cpu].table[0],
-					   target_freq, relation, &new_index))
-		return -EINVAL;
-
-	us2e_set_cpu_divider_index(policy, new_index);
 
 	return 0;
 }
@@ -325,7 +310,7 @@ static int us2e_freq_cpu_exit(struct cpufreq_policy *policy)
 {
 	if (cpufreq_us2e_driver) {
 		cpufreq_frequency_table_put_attr(policy->cpu);
-		us2e_set_cpu_divider_index(policy, 0);
+		us2e_freq_target(policy, 0);
 	}
 
 	return 0;
@@ -358,7 +343,7 @@ static int __init us2e_freq_init(void)
 
 		driver->init = us2e_freq_cpu_init;
 		driver->verify = cpufreq_generic_frequency_table_verify;
-		driver->target = us2e_freq_target;
+		driver->target_index = us2e_freq_target;
 		driver->get = us2e_freq_get;
 		driver->exit = us2e_freq_cpu_exit;
 		strcpy(driver->name, "UltraSPARC-IIe");
diff --git a/drivers/cpufreq/sparc-us3-cpufreq.c b/drivers/cpufreq/sparc-us3-cpufreq.c
index 9b3dbd3..2e54d55 100644
--- a/drivers/cpufreq/sparc-us3-cpufreq.c
+++ b/drivers/cpufreq/sparc-us3-cpufreq.c
@@ -93,8 +93,7 @@ static unsigned int us3_freq_get(unsigned int cpu)
 	return ret;
 }
 
-static void us3_set_cpu_divider_index(struct cpufreq_policy *policy,
-		unsigned int index)
+static int us3_freq_target(struct cpufreq_policy *policy, unsigned int index)
 {
 	unsigned int cpu = policy->cpu;
 	unsigned long new_bits, new_freq, reg;
@@ -136,22 +135,6 @@ static void us3_set_cpu_divider_index(struct cpufreq_policy *policy,
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
 
 	set_cpus_allowed_ptr(current, &cpus_allowed);
-}
-
-static int us3_freq_target(struct cpufreq_policy *policy,
-			  unsigned int target_freq,
-			  unsigned int relation)
-{
-	unsigned int new_index = 0;
-
-	if (cpufreq_frequency_table_target(policy,
-					   &us3_freq_table[policy->cpu].table[0],
-					   target_freq,
-					   relation,
-					   &new_index))
-		return -EINVAL;
-
-	us3_set_cpu_divider_index(policy, new_index);
 
 	return 0;
 }
@@ -182,7 +165,7 @@ static int us3_freq_cpu_exit(struct cpufreq_policy *policy)
 {
 	if (cpufreq_us3_driver) {
 		cpufreq_frequency_table_put_attr(policy->cpu);
-		us3_set_cpu_divider_index(policy, 0);
+		us3_freq_target(policy, 0);
 	}
 
 	return 0;
@@ -219,7 +202,7 @@ static int __init us3_freq_init(void)
 
 		driver->init = us3_freq_cpu_init;
 		driver->verify = cpufreq_generic_frequency_table_verify;
-		driver->target = us3_freq_target;
+		driver->target_index = us3_freq_target;
 		driver->get = us3_freq_get;
 		driver->exit = us3_freq_cpu_exit;
 		strcpy(driver->name, "UltraSPARC-III");
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 33/35] cpufreq: SPEAr: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/spear-cpufreq.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/drivers/cpufreq/spear-cpufreq.c b/drivers/cpufreq/spear-cpufreq.c
index d31accc..88aa5cf 100644
--- a/drivers/cpufreq/spear-cpufreq.c
+++ b/drivers/cpufreq/spear-cpufreq.c
@@ -105,20 +105,16 @@ static int spear1340_set_cpu_rate(struct clk *sys_pclk, unsigned long newfreq)
 }
 
 static int spear_cpufreq_target(struct cpufreq_policy *policy,
-		unsigned int target_freq, unsigned int relation)
+		unsigned int index)
 {
 	struct cpufreq_freqs freqs;
 	unsigned long newfreq;
 	struct clk *srcclk;
-	int index, ret, mult = 1;
-
-	if (cpufreq_frequency_table_target(policy, spear_cpufreq.freq_tbl,
-				target_freq, relation, &index))
-		return -EINVAL;
+	int ret, mult = 1;
 
 	freqs.old = spear_cpufreq_get(0);
-
 	newfreq = spear_cpufreq.freq_tbl[index].frequency * 1000;
+
 	if (of_machine_is_compatible("st,spear1340")) {
 		/*
 		 * SPEAr1340 is special in the sense that due to the possibility
@@ -191,7 +187,7 @@ static struct cpufreq_driver spear_cpufreq_driver = {
 	.name		= "cpufreq-spear",
 	.flags		= CPUFREQ_STICKY,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= spear_cpufreq_target,
+	.target_index	= spear_cpufreq_target,
 	.get		= spear_cpufreq_get,
 	.init		= spear_cpufreq_init,
 	.exit		= cpufreq_generic_exit,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 34/35] cpufreq: speedstep: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/speedstep-centrino.c | 26 +++++++-------------------
 drivers/cpufreq/speedstep-ich.c      | 24 ++++++------------------
 drivers/cpufreq/speedstep-smi.c      | 20 +++++---------------
 3 files changed, 18 insertions(+), 52 deletions(-)

diff --git a/drivers/cpufreq/speedstep-centrino.c b/drivers/cpufreq/speedstep-centrino.c
index c7c14ae..b309774 100644
--- a/drivers/cpufreq/speedstep-centrino.c
+++ b/drivers/cpufreq/speedstep-centrino.c
@@ -422,21 +422,17 @@ static int centrino_cpu_exit(struct cpufreq_policy *policy)
 /**
  * centrino_setpolicy - set a new CPUFreq policy
  * @policy: new policy
- * @target_freq: the target frequency
- * @relation: how that frequency relates to achieved frequency
- *	(CPUFREQ_RELATION_L or CPUFREQ_RELATION_H)
+ * @index: index of target frequency
  *
  * Sets a new CPUFreq policy.
  */
-static int centrino_target (struct cpufreq_policy *policy,
-			    unsigned int target_freq,
-			    unsigned int relation)
+static int centrino_target(struct cpufreq_policy *policy, unsigned int index)
 {
-	unsigned int    newstate = 0;
 	unsigned int	msr, oldmsr = 0, h = 0, cpu = policy->cpu;
 	struct cpufreq_freqs	freqs;
 	int			retval = 0;
 	unsigned int		j, first_cpu, tmp;
+	struct cpufreq_frequency_table *op_points;
 	cpumask_var_t covered_cpus;
 
 	if (unlikely(!zalloc_cpumask_var(&covered_cpus, GFP_KERNEL)))
@@ -447,16 +443,8 @@ static int centrino_target (struct cpufreq_policy *policy,
 		goto out;
 	}
 
-	if (unlikely(cpufreq_frequency_table_target(policy,
-			per_cpu(centrino_model, cpu)->op_points,
-			target_freq,
-			relation,
-			&newstate))) {
-		retval = -EINVAL;
-		goto out;
-	}
-
 	first_cpu = 1;
+	op_points = &per_cpu(centrino_model, cpu)->op_points[index];
 	for_each_cpu(j, policy->cpus) {
 		int good_cpu;
 
@@ -480,7 +468,7 @@ static int centrino_target (struct cpufreq_policy *policy,
 			break;
 		}
 
-		msr = per_cpu(centrino_model, cpu)->op_points[newstate].driver_data;
+		msr = op_points->driver_data;
 
 		if (first_cpu) {
 			rdmsr_on_cpu(good_cpu, MSR_IA32_PERF_CTL, &oldmsr, &h);
@@ -495,7 +483,7 @@ static int centrino_target (struct cpufreq_policy *policy,
 			freqs.new = extract_clock(msr, cpu, 0);
 
 			pr_debug("target=%dkHz old=%d new=%d msr=%04x\n",
-				target_freq, freqs.old, freqs.new, msr);
+				op_points->frequency, freqs.old, freqs.new, msr);
 
 			cpufreq_notify_transition(policy, &freqs,
 					CPUFREQ_PRECHANGE);
@@ -546,7 +534,7 @@ static struct cpufreq_driver centrino_driver = {
 	.init		= centrino_cpu_init,
 	.exit		= centrino_cpu_exit,
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= centrino_target,
+	.target_index	= centrino_target,
 	.get		= get_cur_freq,
 	.attr		= cpufreq_generic_attr,
 };
diff --git a/drivers/cpufreq/speedstep-ich.c b/drivers/cpufreq/speedstep-ich.c
index 929a4f0..be63486 100644
--- a/drivers/cpufreq/speedstep-ich.c
+++ b/drivers/cpufreq/speedstep-ich.c
@@ -251,36 +251,24 @@ static unsigned int speedstep_get(unsigned int cpu)
 /**
  * speedstep_target - set a new CPUFreq policy
  * @policy: new policy
- * @target_freq: the target frequency
- * @relation: how that frequency relates to achieved frequency
- *	(CPUFREQ_RELATION_L or CPUFREQ_RELATION_H)
+ * @index: index of target frequency
  *
  * Sets a new CPUFreq policy.
  */
-static int speedstep_target(struct cpufreq_policy *policy,
-			     unsigned int target_freq,
-			     unsigned int relation)
+static int speedstep_target(struct cpufreq_policy *policy, unsigned int index)
 {
-	unsigned int newstate = 0, policy_cpu;
+	unsigned int policy_cpu;
 	struct cpufreq_freqs freqs;
 
-	if (cpufreq_frequency_table_target(policy, &speedstep_freqs[0],
-				target_freq, relation, &newstate))
-		return -EINVAL;
-
 	policy_cpu = cpumask_any_and(policy->cpus, cpu_online_mask);
 	freqs.old = speedstep_get(policy_cpu);
-	freqs.new = speedstep_freqs[newstate].frequency;
+	freqs.new = speedstep_freqs[index].frequency;
 
 	pr_debug("transiting from %u to %u kHz\n", freqs.old, freqs.new);
 
-	/* no transition necessary */
-	if (freqs.old == freqs.new)
-		return 0;
-
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 
-	smp_call_function_single(policy_cpu, _speedstep_set_state, &newstate,
+	smp_call_function_single(policy_cpu, _speedstep_set_state, &index,
 				 true);
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
@@ -343,7 +331,7 @@ static int speedstep_cpu_init(struct cpufreq_policy *policy)
 static struct cpufreq_driver speedstep_driver = {
 	.name	= "speedstep-ich",
 	.verify	= cpufreq_generic_frequency_table_verify,
-	.target	= speedstep_target,
+	.target_index = speedstep_target,
 	.init	= speedstep_cpu_init,
 	.exit	= cpufreq_generic_exit,
 	.get	= speedstep_get,
diff --git a/drivers/cpufreq/speedstep-smi.c b/drivers/cpufreq/speedstep-smi.c
index b3dfba0..878e64b 100644
--- a/drivers/cpufreq/speedstep-smi.c
+++ b/drivers/cpufreq/speedstep-smi.c
@@ -235,29 +235,19 @@ static void speedstep_set_state(unsigned int state)
 /**
  * speedstep_target - set a new CPUFreq policy
  * @policy: new policy
- * @target_freq: new freq
- * @relation:
+ * @index: index of new freq
  *
  * Sets a new CPUFreq policy/freq.
  */
-static int speedstep_target(struct cpufreq_policy *policy,
-			unsigned int target_freq, unsigned int relation)
+static int speedstep_target(struct cpufreq_policy *policy, unsigned int index)
 {
-	unsigned int newstate = 0;
 	struct cpufreq_freqs freqs;
 
-	if (cpufreq_frequency_table_target(policy, &speedstep_freqs[0],
-				target_freq, relation, &newstate))
-		return -EINVAL;
-
 	freqs.old = speedstep_freqs[speedstep_get_state()].frequency;
-	freqs.new = speedstep_freqs[newstate].frequency;
-
-	if (freqs.old == freqs.new)
-		return 0;
+	freqs.new = speedstep_freqs[index].frequency;
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
-	speedstep_set_state(newstate);
+	speedstep_set_state(index);
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
 
 	return 0;
@@ -340,7 +330,7 @@ static int speedstep_resume(struct cpufreq_policy *policy)
 static struct cpufreq_driver speedstep_driver = {
 	.name		= "speedstep-smi",
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= speedstep_target,
+	.target_index	= speedstep_target,
 	.init		= speedstep_cpu_init,
 	.exit		= cpufreq_generic_exit,
 	.get		= speedstep_get,
-- 
1.7.12.rc2.18.g61b472e

[PATCH V2 35/35] cpufreq: tegra: Covert to light weight ->target_index() routine

[Viresh Kumar]

This patch converts existing .target() to newly defined light weight
.target_index() routine for this driver.

CPUFreq core will call cpufreq_frequency_table_target() before calling this
routine and will pass index to it.

Cc: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/tegra-cpufreq.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/drivers/cpufreq/tegra-cpufreq.c b/drivers/cpufreq/tegra-cpufreq.c
index affb294..ee2cf8f3 100644
--- a/drivers/cpufreq/tegra-cpufreq.c
+++ b/drivers/cpufreq/tegra-cpufreq.c
@@ -150,11 +150,8 @@ static unsigned long tegra_cpu_highest_speed(void)
 	return rate;
 }
 
-static int tegra_target(struct cpufreq_policy *policy,
-		       unsigned int target_freq,
-		       unsigned int relation)
+static int tegra_target(struct cpufreq_policy *policy, unsigned int index)
 {
-	unsigned int idx;
 	unsigned int freq;
 	int ret = 0;
 
@@ -165,10 +162,7 @@ static int tegra_target(struct cpufreq_policy *policy,
 		goto out;
 	}
 
-	cpufreq_frequency_table_target(policy, freq_table, target_freq,
-		relation, &idx);
-
-	freq = freq_table[idx].frequency;
+	freq = freq_table[index].frequency;
 
 	target_cpu_speed[policy->cpu] = freq;
 
@@ -234,7 +228,7 @@ static int tegra_cpu_exit(struct cpufreq_policy *policy)
 
 static struct cpufreq_driver tegra_cpufreq_driver = {
 	.verify		= cpufreq_generic_frequency_table_verify,
-	.target		= tegra_target,
+	.target_index	= tegra_target,
 	.get		= tegra_getspeed,
 	.init		= tegra_cpu_init,
 	.exit		= tegra_cpu_exit,
-- 
1.7.12.rc2.18.g61b472e

Re: [PATCH V2 00/35] CPUFreq: Implement light weight ->target(): for 3.13

[Viresh Kumar]

On 13 August 2013 19:02, Viresh Kumar <viresh.kumar@linaro.org> wrote:
> Currently prototype of cpufreq_drivers target routines is:
>
> int target(struct cpufreq_policy *policy, unsigned int target_freq,
>         unsigned int relation);
>
> And most of the drivers call cpufreq_frequency_table_target() to get a valid
> index of their frequency table which is closest to the target_freq. And they
> don't use target_freq and relation after it.

I just came to know from a friend that I have written "covert" instead of
"convert" in subjects of all the patches.

Will fix it in my repo for now.

Thanks Sudeep :)

--
viresh

Re: [PATCH V2 05/35] cpufreq: at32ap: Covert to light weight ->target_index() routine

[Hans-Christian Egtvedt]

Around Tue 13 Aug 2013 19:02:18 +0530 or thereabout, Viresh Kumar wrote:
> This patch converts existing .target() to newly defined light weight
> .target_index() routine for this driver.
> 
> CPUFreq core will call cpufreq_frequency_table_target() before calling this
> routine and will pass index to it.
> 
> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no>
> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>

Acked-by: Hans-Christian Egtvedt <egtvedt@samfundet.no>

> ---
>  drivers/cpufreq/at32ap-cpufreq.c | 23 +++++------------------
>  1 file changed, 5 insertions(+), 18 deletions(-)

<snipp diff>

-- 
mvh
Hans-Christian Egtvedt

Re: [PATCH V2 01/35] cpufreq: Implement light weight ->target_index() routine

[amit daniel kachhap]

Hi Viresh,

On Tue, Aug 13, 2013 at 7:02 PM, Viresh Kumar <viresh.kumar@linaro.org> wrote:
> Currently prototype of cpufreq_drivers target routines is:
>
> int target(struct cpufreq_policy *policy, unsigned int target_freq,
>                 unsigned int relation);
>
> And most of the drivers call cpufreq_frequency_table_target() to get a valid
> index of their frequency table which is closest to the target_freq. And they
> don't use target_freq and relation after it.
>
> So, it makes sense to just do this work in cpufreq core before calling
> cpufreq_frequency_table_target() and simply pass index instead. But this can be
> done only with drivers which expose their frequency table with cpufreq core. For
> others we need to stick with the old prototype of target() until those drivers
> are converted to expose frequency tables.
>
> This patch implements the new light weight prototype for target_index() routine.
> It looks like this:
>
> int target_index(struct cpufreq_policy *policy, unsigned int index);
This new API is fine but I have another idea.
Say During the registration of the frequency table cpufreq_policy can
be registered as SCALE_DIRECT or SCALE_STEPS. With SCALE_DIRECT flag,
valid frequency will be requested. With this flags the governor itself
can  can figure out if frequency scaling is required or not and very
few calls to __cpufreq_driver_target will happen.
But i agree that in this approach cpufreq_frequency_table_target is
still required but again it can be optimized by binary search as
currently the search is linear.

Thanks,
Amit
>
> CPUFreq core will call cpufreq_frequency_table_target() before calling this
> routine and pass index to it. Because CPUFreq core now requires to call routines
> present in freq_table.c CONFIG_CPU_FREQ_TABLE must be enabled all the time.
>
> This also marks target() interface as deprecated. So, that new drivers avoid
> using it. And
>
> Documentation is updated accordingly.
>
> Cc: Andrew Lunn <andrew@lunn.ch>
> Cc: David S. Miller <davem@davemloft.net>
> Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
> Cc: Eric Miao <eric.y.miao@gmail.com>
> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no>
> Cc: Jesper Nilsson <jesper.nilsson@axis.com>
> Cc: John Crispin <blogic@openwrt.org>
> Cc: Kukjin Kim <kgene.kim@samsung.com>
> Cc: Linus Walleij <linus.walleij@linaro.org>
> Cc: linux-cris-kernel@axis.com
> Cc: Mikael Starvik <starvik@axis.com>
> Cc: Santosh Shilimkar <santosh.shilimkar@ti.com>
> Cc: Sekhar Nori <nsekhar@ti.com>
> Cc: Shawn Guo <shawn.guo@linaro.org>
> Cc: sparclinux@vger.kernel.org
> Cc: Stephen Warren <swarren@nvidia.com>
> Cc: Steven Miao <realmz6@gmail.com>
> Cc: Tony Luck <tony.luck@intel.com>
> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
> ---
>  Documentation/cpu-freq/cpu-drivers.txt | 27 +++++++++++------
>  Documentation/cpu-freq/governors.txt   |  4 +--
>  drivers/cpufreq/Kconfig                |  1 +
>  drivers/cpufreq/cpufreq.c              | 55 +++++++++++++++++++++++++++-------
>  include/linux/cpufreq.h                |  4 ++-
>  5 files changed, 68 insertions(+), 23 deletions(-)
>
> diff --git a/Documentation/cpu-freq/cpu-drivers.txt b/Documentation/cpu-freq/cpu-drivers.txt
> index 40282e6..8b1a445 100644
> --- a/Documentation/cpu-freq/cpu-drivers.txt
> +++ b/Documentation/cpu-freq/cpu-drivers.txt
> @@ -23,8 +23,8 @@ Contents:
>  1.1  Initialization
>  1.2  Per-CPU Initialization
>  1.3  verify
> -1.4  target or setpolicy?
> -1.5  target
> +1.4  target/target_index or setpolicy?
> +1.5  target/target_index
>  1.6  setpolicy
>  2.   Frequency Table Helpers
>
> @@ -56,7 +56,8 @@ cpufreq_driver.init -         A pointer to the per-CPU initialization
>  cpufreq_driver.verify -                A pointer to a "verification" function.
>
>  cpufreq_driver.setpolicy _or_
> -cpufreq_driver.target -                See below on the differences.
> +cpufreq_driver.target/
> +target_index           -       See below on the differences.
>
>  And optionally
>
> @@ -66,7 +67,7 @@ cpufreq_driver.resume -               A pointer to a per-CPU resume function
>                                 which is called with interrupts disabled
>                                 and _before_ the pre-suspend frequency
>                                 and/or policy is restored by a call to
> -                               ->target or ->setpolicy.
> +                               ->target/target_index or ->setpolicy.
>
>  cpufreq_driver.attr -          A pointer to a NULL-terminated list of
>                                 "struct freq_attr" which allow to
> @@ -103,8 +104,8 @@ policy->governor            must contain the "default policy" for
>                                 this CPU. A few moments later,
>                                 cpufreq_driver.verify and either
>                                 cpufreq_driver.setpolicy or
> -                               cpufreq_driver.target is called with
> -                               these values.
> +                               cpufreq_driver.target/target_index is called
> +                               with these values.
>
>  For setting some of these values (cpuinfo.min[max]_freq, policy->min[max]), the
>  frequency table helpers might be helpful. See the section 2 for more information
> @@ -133,20 +134,28 @@ range) is within policy->min and policy->max. If necessary, increase
>  policy->max first, and only if this is no solution, decrease policy->min.
>
>
> -1.4 target or setpolicy?
> +1.4 target/target_index or setpolicy?
>  ----------------------------
>
>  Most cpufreq drivers or even most cpu frequency scaling algorithms
>  only allow the CPU to be set to one frequency. For these, you use the
> -->target call.
> +->target/target_index call.
>
>  Some cpufreq-capable processors switch the frequency between certain
>  limits on their own. These shall use the ->setpolicy call
>
>
> -1.4. target
> +1.4. target/target_index
>  -------------
>
> +The target_index call has two arguments: struct cpufreq_policy *policy,
> +and unsigned int index (into the exposed frequency table).
> +
> +The CPUfreq driver must set the new frequency when called here. The
> +actual frequency must be determined by freq_table[index].frequency.
> +
> +Deprecated:
> +----------
>  The target call has three arguments: struct cpufreq_policy *policy,
>  unsigned int target_frequency, unsigned int relation.
>
> diff --git a/Documentation/cpu-freq/governors.txt b/Documentation/cpu-freq/governors.txt
> index 219970b..77ec215 100644
> --- a/Documentation/cpu-freq/governors.txt
> +++ b/Documentation/cpu-freq/governors.txt
> @@ -40,7 +40,7 @@ Most cpufreq drivers (in fact, all except one, longrun) or even most
>  cpu frequency scaling algorithms only offer the CPU to be set to one
>  frequency. In order to offer dynamic frequency scaling, the cpufreq
>  core must be able to tell these drivers of a "target frequency". So
> -these specific drivers will be transformed to offer a "->target"
> +these specific drivers will be transformed to offer a "->target/target_index"
>  call instead of the existing "->setpolicy" call. For "longrun", all
>  stays the same, though.
>
> @@ -71,7 +71,7 @@ CPU can be set to switch independently         |         CPU can only be set
>                     /                          the limits of policy->{min,max}
>                    /                                \
>                   /                                  \
> -       Using the ->setpolicy call,              Using the ->target call,
> +       Using the ->setpolicy call,              Using the ->target/target_index call,
>             the limits and the                    the frequency closest
>              "policy" is set.                     to target_freq is set.
>                                                   It is assured that it
> diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig
> index 534fcb8..2d06754 100644
> --- a/drivers/cpufreq/Kconfig
> +++ b/drivers/cpufreq/Kconfig
> @@ -2,6 +2,7 @@ menu "CPU Frequency scaling"
>
>  config CPU_FREQ
>         bool "CPU Frequency scaling"
> +       select CPU_FREQ_TABLE
>         help
>           CPU Frequency scaling allows you to change the clock speed of
>           CPUs on the fly. This is a nice method to save power, because
> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> index 37a6874..f1b0e0f 100644
> --- a/drivers/cpufreq/cpufreq.c
> +++ b/drivers/cpufreq/cpufreq.c
> @@ -47,6 +47,11 @@ static LIST_HEAD(cpufreq_policy_list);
>  static DEFINE_PER_CPU(char[CPUFREQ_NAME_LEN], cpufreq_cpu_governor);
>  #endif
>
> +static inline bool has_target(void)
> +{
> +       return cpufreq_driver->target_index || cpufreq_driver->target;
> +}
> +
>  /*
>   * cpu_policy_rwsem is a per CPU reader-writer semaphore designed to cure
>   * all cpufreq/hotplug/workqueue/etc related lock issues.
> @@ -377,7 +382,7 @@ static int cpufreq_parse_governor(char *str_governor, unsigned int *policy,
>                         *policy = CPUFREQ_POLICY_POWERSAVE;
>                         err = 0;
>                 }
> -       } else if (cpufreq_driver->target) {
> +       } else if (has_target()) {
>                 struct cpufreq_governor *t;
>
>                 mutex_lock(&cpufreq_governor_mutex);
> @@ -539,7 +544,7 @@ static ssize_t show_scaling_available_governors(struct cpufreq_policy *policy,
>         ssize_t i = 0;
>         struct cpufreq_governor *t;
>
> -       if (!cpufreq_driver->target) {
> +       if (!has_target()) {
>                 i += sprintf(buf, "performance powersave");
>                 goto out;
>         }
> @@ -822,7 +827,7 @@ static int cpufreq_add_dev_interface(struct cpufreq_policy *policy,
>                 if (ret)
>                         goto err_out_kobj_put;
>         }
> -       if (cpufreq_driver->target) {
> +       if (has_target()) {
>                 ret = sysfs_create_file(&policy->kobj, &scaling_cur_freq.attr);
>                 if (ret)
>                         goto err_out_kobj_put;
> @@ -871,10 +876,10 @@ static int cpufreq_add_policy_cpu(struct cpufreq_policy *policy,
>                                   unsigned int cpu, struct device *dev,
>                                   bool frozen)
>  {
> -       int ret = 0, has_target = !!cpufreq_driver->target;
> +       int ret = 0;
>         unsigned long flags;
>
> -       if (has_target) {
> +       if (has_target()) {
>                 ret = __cpufreq_governor(policy, CPUFREQ_GOV_STOP);
>                 if (ret) {
>                         pr_err("%s: Failed to stop governor\n", __func__);
> @@ -893,7 +898,7 @@ static int cpufreq_add_policy_cpu(struct cpufreq_policy *policy,
>
>         unlock_policy_rwsem_write(policy->cpu);
>
> -       if (has_target) {
> +       if (has_target()) {
>                 if ((ret = __cpufreq_governor(policy, CPUFREQ_GOV_START)) ||
>                         (ret = __cpufreq_governor(policy, CPUFREQ_GOV_LIMITS))) {
>                         pr_err("%s: Failed to start governor\n", __func__);
> @@ -1204,7 +1209,7 @@ static int __cpufreq_remove_dev(struct device *dev,
>                 return -EINVAL;
>         }
>
> -       if (cpufreq_driver->target) {
> +       if (has_target()) {
>                 ret = __cpufreq_governor(policy, CPUFREQ_GOV_STOP);
>                 if (ret) {
>                         pr_err("%s: Failed to stop governor\n", __func__);
> @@ -1244,7 +1249,7 @@ static int __cpufreq_remove_dev(struct device *dev,
>
>         /* If cpu is last user of policy, free policy */
>         if (cpus == 1) {
> -               if (cpufreq_driver->target) {
> +               if (has_target()) {
>                         ret = __cpufreq_governor(policy,
>                                         CPUFREQ_GOV_POLICY_EXIT);
>                         if (ret) {
> @@ -1282,7 +1287,7 @@ static int __cpufreq_remove_dev(struct device *dev,
>                 if (!frozen)
>                         cpufreq_policy_free(policy);
>         } else {
> -               if (cpufreq_driver->target) {
> +               if (has_target()) {
>                         if ((ret = __cpufreq_governor(policy, CPUFREQ_GOV_START)) ||
>                                         (ret = __cpufreq_governor(policy, CPUFREQ_GOV_LIMITS))) {
>                                 pr_err("%s: Failed to start governor\n",
> @@ -1646,11 +1651,39 @@ int __cpufreq_driver_target(struct cpufreq_policy *policy,
>         pr_debug("target for CPU %u: %u kHz, relation %u, requested %u kHz\n",
>                         policy->cpu, target_freq, relation, old_target_freq);
>
> +       /*
> +        * This might look like a redundant call as we are checking it again
> +        * after finding index. But it is left intentionally for cases where
> +        * exactly same freq is called again and so we can save on few function
> +        * calls.
> +        */
>         if (target_freq == policy->cur)
>                 return 0;
>
>         if (cpufreq_driver->target)
>                 retval = cpufreq_driver->target(policy, target_freq, relation);
> +       else if (cpufreq_driver->target_index) {
> +               struct cpufreq_frequency_table *freq_table;
> +               int index;
> +
> +               freq_table = cpufreq_frequency_get_table(policy->cpu);
> +               if (unlikely(!freq_table)) {
> +                       pr_err("%s: Unable to find freq_table\n", __func__);
> +                       return retval;
> +               }
> +
> +               retval = cpufreq_frequency_table_target(policy, freq_table,
> +                               target_freq, relation, &index);
> +               if (unlikely(retval)) {
> +                       pr_err("%s: Unable to find matching freq\n", __func__);
> +                       return retval;
> +               }
> +
> +               if (freq_table[index].frequency == policy->cur)
> +                       return 0;
> +
> +               retval = cpufreq_driver->target_index(policy, index);
> +       }
>
>         return retval;
>  }
> @@ -1983,7 +2016,7 @@ int cpufreq_update_policy(unsigned int cpu)
>                         pr_debug("Driver did not initialize current freq");
>                         policy->cur = new_policy.cur;
>                 } else {
> -                       if (policy->cur != new_policy.cur && cpufreq_driver->target)
> +                       if (policy->cur != new_policy.cur && has_target())
>                                 cpufreq_out_of_sync(cpu, policy->cur,
>                                                                 new_policy.cur);
>                 }
> @@ -2058,7 +2091,7 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data)
>                 return -ENODEV;
>
>         if (!driver_data || !driver_data->verify || !driver_data->init ||
> -           ((!driver_data->setpolicy) && (!driver_data->target)))
> +           (!driver_data->setpolicy && !has_target()))
>                 return -EINVAL;
>
>         pr_debug("trying to register driver %s\n", driver_data->name);
> diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
> index 4907eb2..ff9c8df 100644
> --- a/include/linux/cpufreq.h
> +++ b/include/linux/cpufreq.h
> @@ -195,9 +195,11 @@ struct cpufreq_driver {
>
>         /* define one out of two */
>         int     (*setpolicy)    (struct cpufreq_policy *policy);
> -       int     (*target)       (struct cpufreq_policy *policy,
> +       int     (*target)       (struct cpufreq_policy *policy, /* Deprecated */
>                                  unsigned int target_freq,
>                                  unsigned int relation);
> +       int     (*target_index) (struct cpufreq_policy *policy,
> +                                unsigned int index);
>
>         /* should be defined, if possible */
>         unsigned int    (*get)  (unsigned int cpu);
> --
> 1.7.12.rc2.18.g61b472e
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-pm" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html