[PATCH v5 0/2] Samsung S2MPG10 PMIC MFD-based drivers

[PATCH v5 0/2] Samsung S2MPG10 PMIC MFD-based drivers

[André Draszik]

Original cover letter further down.

This is a resend of two patches from the original series that haven't
been merged yet. That series was merged except for the attached two
patches here. Other than rebasing against next-20250729 there are no
changes to them.

Lee, I think Stephen's intention was to get these two merged via the
MFD tree please.

Original cover letter:
----------------------

This series adds initial support for the Samsung S2MPG10 PMIC using the
MFD framework. This is a PMIC for mobile applications and is used on
the Google Pixel 6 and 6 Pro (oriole / raven).

*** dependency note ***

To compile, this depends on the Samsung ACPM driver in Linux next with
the following additional patches:
https://lore.kernel.org/all/20250324-acpm-atomic-v2-0-7d87746e1765@linaro.org/
https://lore.kernel.org/all/20250319-acpm-fixes-v2-0-ac2c1bcf322b@linaro.org/
https://lore.kernel.org/all/20250327-acpm-children-v1-0-0afe15ee2ff7@linaro.org/

*** dependency note end ***

+++ Kconfig update +++

There is a Kconfig symbol update in this series, because the existing
Samsung S2M driver has been split into core and transport (I2C & ACPM)
parts. CONFIG_MFD_SEC_CORE is now truly a core driver, and
the I2C code that was part of it is now enabled via CONFIG_MFD_SEC_I2C.

This was necessary because unlike the other S2M PMICs, S2MPG10 doesn't
talk via I2C, but via the Samsung ACPM firmware.

+++ Kconfig update end +++

This series must be applied in-order, due to interdependencies of some
of the patches. There are also various cleanup patches to the S2M
drivers. I've kept them ordered as:
  * DT bindings (patches 1 ... 3)
  * s2m mfd prep for adding S2MPG10 support (patches 4 ... 7)
  * split S2M mfd driver into s2m-core and s2m-i2c, including the
    kconfig symbol update (patch 8)
  * S2MPG10 core driver (patch 9)
  * s2m mfd driver cleanup patches (patches 10 ... 23)
  * S2MPG10 clock driver (patch 24)
  * s2m RTC prep for adding S2MPG10 (patch 25 ... 26)
  * S2MPG10 RTC driver (patch 27)
  * s2m RTC cleanup patches (patches 28 ... 31)

I realise these are many, but since some prep-work was required to be
able to add S2MPG anyway, I wanted to get the cleanup patches in as
well :-) Let me know if I should postpone them to a later date instead.

The S2MPG10 includes buck converters, various LDOs, power meters, RTC,
clock outputs, and additional GPIOs interfaces.

This series adds support in the top-level device driver, and for the
RTC and clock. Importantly, having the RTC driver allows to do a proper
reset of the system. Drivers or driver updates for the other components
will be added in future patches.

This will need a DT update for Oriole / Raven to enable this device. I
will send that out separately.

Cheers,
Andre'

Signed-off-by: André Draszik <andre.draszik@linaro.org>
---
Changes in v5:
- just a rebase & resend of the last two remaining patches
- no other changes
- Link to v4: https://lore.kernel.org/r/20250409-s2mpg10-v4-0-d66d5f39b6bf@linaro.org

Changes in v4:
- various updates to sec-acpm (patch 9, Lee)
- cache enum type in patch 25 (Krzysztof)
- collect tags
- Link to v3: https://lore.kernel.org/r/20250403-s2mpg10-v3-0-b542b3505e68@linaro.org

Changes in v3:
- Krzysztof:
  - keep 'regulators' subnode required even for s2mpg10
  - drop '$ref' and 'unevaluatedProperties' from pmic subnode, use
    'additionalProperties' instead
  - add some regulators to examples since s2mpg10 requires them as of
    v3
- sec-acpm:
  - use an enum for struct sec_acpm_bus_context::type
  - consistent name space for all functions sec_pmic_acpm_... to be
    similar to i2c and consistent in this file
- Link to v2: https://lore.kernel.org/r/20250328-s2mpg10-v2-0-b54dee33fb6b@linaro.org

Changes in v2:
- Rob:
  - make PMIC node a child of ACPM, and all related changes (binding,
    driver)
- Krzysztof:
  - merge defconfig updates into patch changing the symbols (patch 8)
  - split MODULE_AUTHOR update into a separate patch
  - better alignment fix (patch 11)
  - merge two s2dos05/s2mpu05 related patches into one (patch 14)
- myself:
  - keep PMIC DT parsing in core, not in transport driver
  - several updates in sec-acpm.c, see separate entries in patch 9
  - fix typo in patch 17
  - collect tags
- Link to v1: https://lore.kernel.org/r/20250323-s2mpg10-v1-0-d08943702707@linaro.org

---
André Draszik (2):
      dt-bindings: clock: samsung,s2mps11: add s2mpg10
      clk: s2mps11: add support for S2MPG10 PMIC clock

 Documentation/devicetree/bindings/clock/samsung,s2mps11.yaml | 1 +
 drivers/clk/clk-s2mps11.c                                    | 8 ++++++++
 2 files changed, 9 insertions(+)
---
base-commit: 54efec8782214652b331c50646013f8526570e8d
change-id: 20250321-s2mpg10-ef5d1ebd3043

Best regards,
-- 
André Draszik <andre.draszik@linaro.org>

[PATCH v5 1/2] dt-bindings: clock: samsung,s2mps11: add s2mpg10

[André Draszik]

The Samsung S2MPG10 clock controller is similar to the existing clock
controllers supported by this binding. Register offsets / layout are
slightly different, so it needs its own compatible.

Acked-by: Stephen Boyd <sboyd@kernel.org>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: André Draszik <andre.draszik@linaro.org>
---
 Documentation/devicetree/bindings/clock/samsung,s2mps11.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/clock/samsung,s2mps11.yaml b/Documentation/devicetree/bindings/clock/samsung,s2mps11.yaml
index d5296e6053a1881650b8e8ff2524ea01689b7395..91d455155a606a60ed2006e57709466ae8d72664 100644
--- a/Documentation/devicetree/bindings/clock/samsung,s2mps11.yaml
+++ b/Documentation/devicetree/bindings/clock/samsung,s2mps11.yaml
@@ -25,6 +25,7 @@ description: |
 properties:
   compatible:
     enum:
+      - samsung,s2mpg10-clk
       - samsung,s2mps11-clk
       - samsung,s2mps13-clk # S2MPS13 and S2MPS15
       - samsung,s2mps14-clk

-- 
2.50.1.552.g942d659e1b-goog

[PATCH v5 2/2] clk: s2mps11: add support for S2MPG10 PMIC clock

[André Draszik]

Add support for Samsung's S2MPG10 PMIC clock, which is similar to the
existing PMIC clocks supported by this driver.

S2MPG10 has three clock outputs @ 32kHz: AP, peri1 and peri2.

Acked-by: Stephen Boyd <sboyd@kernel.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: André Draszik <andre.draszik@linaro.org>
---
 drivers/clk/clk-s2mps11.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/clk/clk-s2mps11.c b/drivers/clk/clk-s2mps11.c
index d4e9c3577b35dec8d9ec67c489b7b5ae27211f55..ff7ce12a5da6b437b5c92b9a32dcaf9423647cde 100644
--- a/drivers/clk/clk-s2mps11.c
+++ b/drivers/clk/clk-s2mps11.c
@@ -11,6 +11,7 @@
 #include <linux/regmap.h>
 #include <linux/clk-provider.h>
 #include <linux/platform_device.h>
+#include <linux/mfd/samsung/s2mpg10.h>
 #include <linux/mfd/samsung/s2mps11.h>
 #include <linux/mfd/samsung/s2mps13.h>
 #include <linux/mfd/samsung/s2mps14.h>
@@ -140,6 +141,9 @@ static int s2mps11_clk_probe(struct platform_device *pdev)
 	clk_data->num = S2MPS11_CLKS_NUM;
 
 	switch (hwid) {
+	case S2MPG10:
+		s2mps11_reg = S2MPG10_PMIC_RTCBUF;
+		break;
 	case S2MPS11X:
 		s2mps11_reg = S2MPS11_REG_RTC_CTRL;
 		break;
@@ -221,6 +225,7 @@ static void s2mps11_clk_remove(struct platform_device *pdev)
 }
 
 static const struct platform_device_id s2mps11_clk_id[] = {
+	{ "s2mpg10-clk", S2MPG10},
 	{ "s2mps11-clk", S2MPS11X},
 	{ "s2mps13-clk", S2MPS13X},
 	{ "s2mps14-clk", S2MPS14X},
@@ -241,6 +246,9 @@ MODULE_DEVICE_TABLE(platform, s2mps11_clk_id);
  */
 static const struct of_device_id s2mps11_dt_match[] __used = {
 	{
+		.compatible = "samsung,s2mpg10-clk",
+		.data = (void *)S2MPG10,
+	}, {
 		.compatible = "samsung,s2mps11-clk",
 		.data = (void *)S2MPS11X,
 	}, {

-- 
2.50.1.552.g942d659e1b-goog

Re: [PATCH v5 0/2] Samsung S2MPG10 PMIC MFD-based drivers

[Lee Jones]

On Wed, 30 Jul 2025, André Draszik wrote:

> Original cover letter further down.
> 
> This is a resend of two patches from the original series that haven't
> been merged yet. That series was merged except for the attached two
> patches here. Other than rebasing against next-20250729 there are no
> changes to them.
> 
> Lee, I think Stephen's intention was to get these two merged via the
> MFD tree please.

Although I have no issue with this, it does seem a little odd now that
the set consists of only Clk patches.  Let me know what you / Stephen
decide.

-- 
Lee Jones [李琼斯]

Re: [PATCH v5 0/2] Samsung S2MPG10 PMIC MFD-based drivers

[André Draszik]

On Wed, 2025-07-30 at 15:51 +0100, Lee Jones wrote:
> On Wed, 30 Jul 2025, André Draszik wrote:
> 
> > Original cover letter further down.
> > 
> > This is a resend of two patches from the original series that haven't
> > been merged yet. That series was merged except for the attached two
> > patches here. Other than rebasing against next-20250729 there are no
> > changes to them.
> > 
> > Lee, I think Stephen's intention was to get these two merged via the
> > MFD tree please.
> 
> Although I have no issue with this, it does seem a little odd now that
> the set consists of only Clk patches.  Let me know what you / Stephen
> decide.

Thanks Lee.

I simply went by Stephen's ACK, which to me implies he wanted it merged
via a different tree (mfd). I guess at this stage it doesn't matter anymore,
since all the core changes are in already.

I'll defer to Stephen :-)


Cheers,
Andre'

Re: [PATCH v5 0/2] Samsung S2MPG10 PMIC MFD-based drivers

[Stephen Boyd]

Quoting André Draszik (2025-07-31 03:20:56)
> On Wed, 2025-07-30 at 15:51 +0100, Lee Jones wrote:
> > On Wed, 30 Jul 2025, André Draszik wrote:
> > 
> > > Original cover letter further down.
> > > 
> > > This is a resend of two patches from the original series that haven't
> > > been merged yet. That series was merged except for the attached two
> > > patches here. Other than rebasing against next-20250729 there are no
> > > changes to them.
> > > 
> > > Lee, I think Stephen's intention was to get these two merged via the
> > > MFD tree please.
> > 
> > Although I have no issue with this, it does seem a little odd now that
> > the set consists of only Clk patches.  Let me know what you / Stephen
> > decide.
> 
> Thanks Lee.
> 
> I simply went by Stephen's ACK, which to me implies he wanted it merged
> via a different tree (mfd). I guess at this stage it doesn't matter anymore,
> since all the core changes are in already.
> 

I'll pick it up after the merge window closes.