[GIT PULL] Microchip clock updates for v6.12

[GIT PULL] Microchip clock updates for v6.12

[Claudiu Beznea]

The following changes since commit 8400291e289ee6b2bf9779ff1c83a291501f017b:

  Linux 6.11-rc1 (2024-07-28 14:19:55 -0700)

are available in the Git repository at:

  https://git.kernel.org/pub/scm/linux/kernel/git/at91/linux.git tags/clk-microchip-6.12

for you to fetch changes up to 2d6e9ee7cb3e79b1713783c633b13af9aeffc90c:

  clk: at91: sama7g5: Allocate only the needed amount of memory for PLLs (2024-08-24 17:44:11 +0300)

----------------------------------------------------------------
Microchip clock updates for v6.12

It contains:
- support for the Microchip SAM9X7 SoC as follows:
-- updates on the PLL drivers
-- a new clock driver was added for SAM9X7
-- dt-binding documentation updates (for the new clock driver and for
   the slow clock controller that SAM9X7 is using)
- a fix for the Microchip SAMA7G5 clock driver to avoid allocating mode
  than necessary memory

----------------------------------------------------------------
Claudiu Beznea (1):
      clk: at91: sama7g5: Allocate only the needed amount of memory for PLLs

Varshini Rajendran (7):
      dt-bindings: clocks: atmel,at91sam9x5-sckc: add sam9x7
      dt-bindings: clocks: atmel,at91rm9200-pmc: add sam9x7 clock controller
      clk: at91: clk-sam9x60-pll: re-factor to support individual core freq outputs
      clk: at91: sam9x7: add support for HW PLL freq dividers
      clk: at91: sama7g5: move mux table macros to header file
      dt-bindings: clock: at91: Allow PLLs to be exported and referenced in DT
      clk: at91: sam9x7: add sam9x7 pmc driver

 .../bindings/clock/atmel,at91rm9200-pmc.yaml       |   2 +
 .../bindings/clock/atmel,at91sam9x5-sckc.yaml      |   4 +-
 drivers/clk/at91/Makefile                          |   1 +
 drivers/clk/at91/clk-sam9x60-pll.c                 |  42 +-
 drivers/clk/at91/pmc.h                             |  18 +
 drivers/clk/at91/sam9x60.c                         |   7 +
 drivers/clk/at91/sam9x7.c                          | 946 +++++++++++++++++++++
 drivers/clk/at91/sama7g5.c                         |  47 +-
 include/dt-bindings/clock/at91.h                   |   4 +
 9 files changed, 1035 insertions(+), 36 deletions(-)
 create mode 100644 drivers/clk/at91/sam9x7.c

Re: [GIT PULL] Microchip clock updates for v6.12

[Stephen Boyd]

Quoting Claudiu Beznea (2024-09-01 06:34:25)
> The following changes since commit 8400291e289ee6b2bf9779ff1c83a291501f017b:
> 
>   Linux 6.11-rc1 (2024-07-28 14:19:55 -0700)
> 
> are available in the Git repository at:
> 
>   https://git.kernel.org/pub/scm/linux/kernel/git/at91/linux.git tags/clk-microchip-6.12
> 
> for you to fetch changes up to 2d6e9ee7cb3e79b1713783c633b13af9aeffc90c:
> 
>   clk: at91: sama7g5: Allocate only the needed amount of memory for PLLs (2024-08-24 17:44:11 +0300)
> 
> ----------------------------------------------------------------

Thanks. Pulled into clk-next

Please get rid of the clk.h include in sam9x7.c though. Looks like you
can use struct clk_parent_data and use fw_name for that. Also, could it
be a simple platform driver instead? That would be ideal.

Re: [GIT PULL] Microchip clock updates for v6.12

[claudiu beznea]

On 03.09.2024 23:55, Stephen Boyd wrote:
> Quoting Claudiu Beznea (2024-09-01 06:34:25)
>> The following changes since commit 8400291e289ee6b2bf9779ff1c83a291501f017b:
>>
>>   Linux 6.11-rc1 (2024-07-28 14:19:55 -0700)
>>
>> are available in the Git repository at:
>>
>>   https://git.kernel.org/pub/scm/linux/kernel/git/at91/linux.git tags/clk-microchip-6.12
>>
>> for you to fetch changes up to 2d6e9ee7cb3e79b1713783c633b13af9aeffc90c:
>>
>>   clk: at91: sama7g5: Allocate only the needed amount of memory for PLLs (2024-08-24 17:44:11 +0300)
>>
>> ----------------------------------------------------------------
> 
> Thanks. Pulled into clk-next

Thanks!

> 
> Please get rid of the clk.h include in sam9x7.c though. Looks like you
> can use struct clk_parent_data and use fw_name for that.

It's on my todo list. I have something prepared but didn't managed to
publish something.

> Also, could it
> be a simple platform driver instead? That would be ideal.

I'm not sure at the moment as there are clocks that feed the timer. I'll
have to explore this.

Thank you,
Claudiu Beznea