Re: [PATCH v5 12/21] clk: tegra: pll: Add specialized logic for T210

[Rhyland Klein <rklein@nvidia.com>]

On 5/13/2015 4:59 PM, Benson Leung wrote:
> On Tue, May 12, 2015 at 10:23 AM, Rhyland Klein <rklein@nvidia.com> wrote:
>> On Tegra210 SoC's, the logic to enable several of the plls is different
>> from previous generations. Therefore, add registeration functions specific
> 
> s/registeration/registration
> 
>> to Tegra210 which will handle them appropriately.
>>
>> Signed-off-by: Rhyland Klein <rklein@nvidia.com>
>> ---
>> v5:
>>   - Removed unused variables.
>>
>> v4:
>>   - Fixed plle_tegra210_is_enabled callback typos
>>
>> v2:
>>   - Fixed plle logic. PLLE on Tegra210 has had its enable bit moved, so
>>     we can't use the default _clk_pll_enable/disable routines. Instead,
>>     manually set/clear the correct bit for PLLE within its functions.
>>
>>  drivers/clk/tegra/clk-pll.c |  347 ++++++++++++++++++++++++++++++++++++++++++-
>>  drivers/clk/tegra/clk.h     |   24 +++
>>  2 files changed, 369 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/clk/tegra/clk-pll.c b/drivers/clk/tegra/clk-pll.c
>> index 54ebab2b85e5..2b1f672e3bdb 100644
>> --- a/drivers/clk/tegra/clk-pll.c
>> +++ b/drivers/clk/tegra/clk-pll.c
>> @@ -65,6 +65,7 @@
>>  #define PLLE_BASE_DIVN_WIDTH 8
>>  #define PLLE_BASE_DIVM_SHIFT 0
>>  #define PLLE_BASE_DIVM_WIDTH 8
>> +#define PLLE_BASE_ENABLE BIT(31)
>>
>>  #define PLLE_MISC_SETUP_BASE_SHIFT 16
>>  #define PLLE_MISC_SETUP_BASE_MASK (0xffff << PLLE_MISC_SETUP_BASE_SHIFT)
>> @@ -910,7 +911,8 @@ const struct clk_ops tegra_clk_plle_ops = {
>>
>>  #if defined(CONFIG_ARCH_TEGRA_114_SOC) || \
>>         defined(CONFIG_ARCH_TEGRA_124_SOC) || \
>> -       defined(CONFIG_ARCH_TEGRA_132_SOC)
>> +       defined(CONFIG_ARCH_TEGRA_132_SOC) || \
>> +       defined(CONFIG_ARCH_TEGRA_210_SOC)
>>
>>  static int _pll_fixed_mdiv(struct tegra_clk_pll_params *pll_params,
>>                            unsigned long parent_rate)
>> @@ -1579,7 +1581,8 @@ struct clk *tegra_clk_register_plle(const char *name, const char *parent_name,
>>
>>  #if defined(CONFIG_ARCH_TEGRA_114_SOC) || \
>>         defined(CONFIG_ARCH_TEGRA_124_SOC) || \
>> -       defined(CONFIG_ARCH_TEGRA_132_SOC)
>> +       defined(CONFIG_ARCH_TEGRA_132_SOC) || \
>> +       defined(CONFIG_ARCH_TEGRA_210_SOC)
>>  static const struct clk_ops tegra_clk_pllxc_ops = {
>>         .is_enabled = clk_pll_is_enabled,
>>         .enable = clk_pll_enable,
> 
> 
> It looks like you might have missed changing one set of #if defined
> further down below to include CONFIG_ARCH_TEGRA_210_SOC :
> 
> #if defined(CONFIG_ARCH_TEGRA_124_SOC) || defined(CONFIG_ARCH_TEGRA_132_SOC)
...
> struct clk *tegra_clk_register_pllss(const char *name, const char *parent_name,
...
> #endif
> 

No, this was intentional. tegra_clk_register_pllss isn't used on
Tegra210, it adds its own version "register_clk_register_pllss_tegra210"
which it uses and it doesn't need the tegra_clk_pllss_ops either. So it
should be fine without this.

-rhyland

-- 
nvpublic