2014-04-14 21:08:13

by Stratos Karafotis

Subject: [PATCH 02/20] cpufreq: acpi-cpufreq: Use cpufreq_for_each_entry macro for iteration

The cpufreq core supports the cpufreq_for_each_entry macro helper
for iteration over the cpufreq_frequency_table, so use it.

It should have no functional changes.

Signed-off-by: Stratos Karafotis <[email protected]>
---
drivers/cpufreq/acpi-cpufreq.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c
index 000e4e0..12a5750 100644
--- a/drivers/cpufreq/acpi-cpufreq.c
+++ b/drivers/cpufreq/acpi-cpufreq.c
@@ -213,7 +213,7 @@ static unsigned extract_io(u32 value, struct acpi_cpufreq_data *data)

static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data)
{
- int i;
+ struct cpufreq_frequency_table *pos;
struct acpi_processor_performance *perf;

if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
@@ -223,11 +223,11 @@ static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data)

perf = data->acpi_data;

- for (i = 0; data->freq_table[i].frequency != CPUFREQ_TABLE_END; i++) {
- if (msr == perf->states[data->freq_table[i].driver_data].status)
- return data->freq_table[i].frequency;
+ cpufreq_for_each_entry(pos, data->freq_table) {
+ if (msr == perf->states[pos->driver_data].status)
+ return pos->frequency;
}
- return data->freq_table[0].frequency;
+ return data->freq_table->frequency;
}

static unsigned extract_freq(u32 val, struct acpi_cpufreq_data *data)
--
1.9.0

2014-04-15 05:32:46

by Viresh Kumar

Subject: Re: [PATCH 02/20] cpufreq: acpi-cpufreq: Use cpufreq_for_each_entry macro for iteration

On 15 April 2014 02:38, Stratos Karafotis <[email protected]> wrote:
> The cpufreq core supports the cpufreq_for_each_entry macro helper
> for iteration over the cpufreq_frequency_table, so use it.
>
> It should have no functional changes.
>
> Signed-off-by: Stratos Karafotis <[email protected]>
> ---
> drivers/cpufreq/acpi-cpufreq.c | 10 +++++-----
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c
> index 000e4e0..12a5750 100644
> --- a/drivers/cpufreq/acpi-cpufreq.c
> +++ b/drivers/cpufreq/acpi-cpufreq.c
> @@ -213,7 +213,7 @@ static unsigned extract_io(u32 value, struct acpi_cpufreq_data *data)
>
> static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data)
> {
> - int i;
> + struct cpufreq_frequency_table *pos;
> struct acpi_processor_performance *perf;
>
> if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
> @@ -223,11 +223,11 @@ static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data)
>
> perf = data->acpi_data;
>
> - for (i = 0; data->freq_table[i].frequency != CPUFREQ_TABLE_END; i++) {
> - if (msr == perf->states[data->freq_table[i].driver_data].status)
> - return data->freq_table[i].frequency;
> + cpufreq_for_each_entry(pos, data->freq_table) {
> + if (msr == perf->states[pos->driver_data].status)
> + return pos->frequency;
> }
> - return data->freq_table[0].frequency;
> + return data->freq_table->frequency;

This isn't a related change, isn't it? Also it make it less readable.
So probably
just leave it as is.

2014-04-15 11:00:19

by Stratos Karafotis

Subject: Re: [PATCH 02/20] cpufreq: acpi-cpufreq: Use cpufreq_for_each_entry macro for iteration

On 15/04/2014 08:32 πμ, Viresh Kumar wrote:
> On 15 April 2014 02:38, Stratos Karafotis <[email protected]> wrote:
>> The cpufreq core supports the cpufreq_for_each_entry macro helper
>> for iteration over the cpufreq_frequency_table, so use it.
>>
>> It should have no functional changes.
>>
>> Signed-off-by: Stratos Karafotis <[email protected]>
>> ---
>> drivers/cpufreq/acpi-cpufreq.c | 10 +++++-----
>> 1 file changed, 5 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c
>> index 000e4e0..12a5750 100644
>> --- a/drivers/cpufreq/acpi-cpufreq.c
>> +++ b/drivers/cpufreq/acpi-cpufreq.c
>> @@ -213,7 +213,7 @@ static unsigned extract_io(u32 value, struct acpi_cpufreq_data *data)
>>
>> static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data)
>> {
>> - int i;
>> + struct cpufreq_frequency_table *pos;
>> struct acpi_processor_performance *perf;
>>
>> if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
>> @@ -223,11 +223,11 @@ static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data)
>>
>> perf = data->acpi_data;
>>
>> - for (i = 0; data->freq_table[i].frequency != CPUFREQ_TABLE_END; i++) {
>> - if (msr == perf->states[data->freq_table[i].driver_data].status)
>> - return data->freq_table[i].frequency;
>> + cpufreq_for_each_entry(pos, data->freq_table) {
>> + if (msr == perf->states[pos->driver_data].status)
>> + return pos->frequency;
>> }
>> - return data->freq_table[0].frequency;
>> + return data->freq_table->frequency;
>
> This isn't a related change, isn't it? Also it make it less readable.
> So probably
> just leave it as is.
>

I thought that since we use pointer notation it would be more clear.
But, I will change it according to your suggestion.


Stratos