[PATCH v2 0/2] mmc: host: renesas_sdhi_core: support configuring an optional sdio mux

[PATCH v2 0/2] mmc: host: renesas_sdhi_core: support configuring an optional sdio mux

[Josua Mayer]

Some Renesas SoC based boards mux SD and eMMC on a single sdio
controller, exposing user control by dip switch and software control by
gpio.

Purpose is to simplify development and provisioning by selecting boot
media at power-on, and again before starting linux.

Add binding and driver support for linking a (gpio) mux to renesas sdio
controller.

Signed-off-by: Josua Mayer <josua@solid-run.com>
---
Changes in v2:
- dropped mux-controller node from dt binding example
  (Reported-by: Conor Dooley <conor@kernel.org>
   Reported-by: Krzysztof Kozlowski <krzk@kernel.org>)
- Link to v1: https://lore.kernel.org/r/20251128-rz-sdio-mux-v1-0-1ede318d160f@solid-run.com

---
Josua Mayer (2):
      dt-bindings: mmc: renesas,sdhi: Add mux-states property
      mmc: host: renesas_sdhi_core: support configuring an optional sdio mux

 .../devicetree/bindings/mmc/renesas,sdhi.yaml      |  6 +++++
 drivers/mmc/host/Kconfig                           |  1 +
 drivers/mmc/host/renesas_sdhi.h                    |  1 +
 drivers/mmc/host/renesas_sdhi_core.c               | 27 +++++++++++++++++++---
 4 files changed, 32 insertions(+), 3 deletions(-)
---
base-commit: 3a8660878839faadb4f1a6dd72c3179c1df56787
change-id: 20251128-rz-sdio-mux-acc5137f1618

Best regards,
-- 
Josua Mayer <josua@solid-run.com>

[PATCH v2 1/2] dt-bindings: mmc: renesas,sdhi: Add mux-states property

[Josua Mayer]

Add mux controller support for when sdio lines are muxed between a host
and multiple cards.

There are several devices supporting a choice of eMMC or SD on a single
board by both dip switch and gpio, e.g. Renesas RZ/G2L SMARC SoM and
SolidRun RZ/G2L SoM.

In-tree dts for the Renesas boards currently rely on preprocessor macros
to hog gpios and define the card.

By adding mux-states property to sdio controller description, boards can
correctly describe the mux that already exists in hardware - and drivers
can coordinate between mux selection and probing for cards.

Signed-off-by: Josua Mayer <josua@solid-run.com>
---
 Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
index c754ea71f51f7..754ccb1c30efb 100644
--- a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
+++ b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
@@ -106,6 +106,11 @@ properties:
   iommus:
     maxItems: 1
 
+  mux-states:
+    description:
+      mux controller node to route the SDIO signals from SoC to cards.
+    maxItems: 1
+
   power-domains:
     maxItems: 1
 
@@ -275,6 +280,7 @@ examples:
         max-frequency = <195000000>;
         power-domains = <&sysc R8A7790_PD_ALWAYS_ON>;
         resets = <&cpg 314>;
+        mux-states = <&mux 0>;
     };
 
     sdhi1: mmc@ee120000 {

-- 
2.51.0

RE: [PATCH v2 1/2] dt-bindings: mmc: renesas,sdhi: Add mux-states property

[Biju Das]

Hi Josua Mayer,


Thanks for the patch.

> -----Original Message-----
> From: Josua Mayer <josua@solid-run.com>
> Sent: 01 December 2025 12:31
> Subject: [PATCH v2 1/2] dt-bindings: mmc: renesas,sdhi: Add mux-states property
> 
> Add mux controller support for when sdio lines are muxed between a host and multiple cards.
> 
> There are several devices supporting a choice of eMMC or SD on a single board by both dip switch and
> gpio, e.g. Renesas RZ/G2L SMARC SoM and SolidRun RZ/G2L SoM.
> 
> In-tree dts for the Renesas boards currently rely on preprocessor macros to hog gpios and define the
> card.
> 
> By adding mux-states property to sdio controller description, boards can correctly describe the mux
> that already exists in hardware - and drivers can coordinate between mux selection and probing for
> cards.

> 
> Signed-off-by: Josua Mayer <josua@solid-run.com>
> ---
>  Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> index c754ea71f51f7..754ccb1c30efb 100644
> --- a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> +++ b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> @@ -106,6 +106,11 @@ properties:
>    iommus:
>      maxItems: 1
> 
> +  mux-states:
> +    description:
> +      mux controller node to route the SDIO signals from SoC to cards.

Maybe describe 0 - state for SD and 1 - state for eMMC ??


> +    maxItems: 1

> +
>    power-domains:
>      maxItems: 1
> 
> @@ -275,6 +280,7 @@ examples:
>          max-frequency = <195000000>;
>          power-domains = <&sysc R8A7790_PD_ALWAYS_ON>;
>          resets = <&cpg 314>;
> +        mux-states = <&mux 0>;

On R-Car mmc/sd mux available only on SD2/SD3, so I guess you picked wrong node SD0??

or 

Add separate example with RZ/G2L, as the boards(RZ/G2L SMARC EVK)
use gpio/switch (XOR function) to select the eMMC or SD signals??

Cheers,
Biju

Re: [PATCH v2 1/2] dt-bindings: mmc: renesas,sdhi: Add mux-states property

[Josua Mayer]

Hi Biju,

Am 01.12.25 um 14:02 schrieb Biju Das:
> Hi Josua Mayer,
>
>
> Thanks for the patch.
>
>> -----Original Message-----
>> From: Josua Mayer <josua@solid-run.com>
>> Sent: 01 December 2025 12:31
>> Subject: [PATCH v2 1/2] dt-bindings: mmc: renesas,sdhi: Add mux-states property
>>
>> Add mux controller support for when sdio lines are muxed between a host and multiple cards.
>>
>> There are several devices supporting a choice of eMMC or SD on a single board by both dip switch and
>> gpio, e.g. Renesas RZ/G2L SMARC SoM and SolidRun RZ/G2L SoM.
>>
>> In-tree dts for the Renesas boards currently rely on preprocessor macros to hog gpios and define the
>> card.
>>
>> By adding mux-states property to sdio controller description, boards can correctly describe the mux
>> that already exists in hardware - and drivers can coordinate between mux selection and probing for
>> cards.
>> Signed-off-by: Josua Mayer <josua@solid-run.com>
>> ---
>>  Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml | 6 ++++++
>>  1 file changed, 6 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
>> b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
>> index c754ea71f51f7..754ccb1c30efb 100644
>> --- a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
>> +++ b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
>> @@ -106,6 +106,11 @@ properties:
>>    iommus:
>>      maxItems: 1
>>
>> +  mux-states:
>> +    description:
>> +      mux controller node to route the SDIO signals from SoC to cards.
> Maybe describe 0 - state for SD and 1 - state for eMMC ??
The state-mmc-sd mapping depends on the wiring and description
of the mux, so we cannot differentiate in the mux-states property.

The idea is that whatever numeric argument you add to the phandle
will select a particular state for the mux.
The rest of dt properties must still be specific to SD or eMMC, e.g.:

&sdhi0 {
    pinctrl-0 = <&sdhi0_pins>, <&sdhi0_cd_pins>;
    pinctrl-1 = <&sdhi0_uhs_pins>, <&sdhi0_cd_pins>;
    pinctrl-names = "default", "state_uhs";
    vmmc-supply = <&vmmc>;
    vqmmc-supply = <&reg_pmic_ldo1>;
    bus-width = <4>;
    sd-uhs-sdr50;
    sd-uhs-sdr104;
    full-pwr-cycle;
    mux-states = <&sdhi0_mux 1>;
    status = "okay";
};

>
>
>> +    maxItems: 1
>> +
>>    power-domains:
>>      maxItems: 1
>>
>> @@ -275,6 +280,7 @@ examples:
>>          max-frequency = <195000000>;
>>          power-domains = <&sysc R8A7790_PD_ALWAYS_ON>;
>>          resets = <&cpg 314>;
>> +        mux-states = <&mux 0>;
> On R-Car mmc/sd mux available only on SD2/SD3, so I guess you picked wrong node SD0??
As they were examples I did not give it much thought,
We could even omit mux-states from examples completely since it is optional.
>
> or 
>
> Add separate example with RZ/G2L, as the boards(RZ/G2L SMARC EVK)
> use gpio/switch (XOR function) to select the eMMC or SD signals??

I can if so desired add verbose example from our own G2L based board, e.g.:

sdhi0: mmc@11c00000 {
    compatible = "renesas,sdhi-r9a07g044", "renesas,rzg2l-sdhi";
    reg = <0x0 0x11c00000 0 0x10000>;
    interrupts = <GIC_SPI 104 IRQ_TYPE_LEVEL_HIGH>,
             <GIC_SPI 105 IRQ_TYPE_LEVEL_HIGH>;
    clocks = <&cpg CPG_MOD R9A07G044_SDHI0_IMCLK>,
         <&cpg CPG_MOD R9A07G044_SDHI0_CLK_HS>,
         <&cpg CPG_MOD R9A07G044_SDHI0_IMCLK2>,
         <&cpg CPG_MOD R9A07G044_SDHI0_ACLK>;
    clock-names = "core", "clkh", "cd", "aclk";
    resets = <&cpg R9A07G044_SDHI0_IXRST>;
    power-domains = <&cpg>;
    pinctrl-0 = <&sdhi0_pins>, <&sdhi0_cd_pins>;
    pinctrl-1 = <&sdhi0_uhs_pins>, <&sdhi0_cd_pins>;
    pinctrl-names = "default", "state_uhs";
    vmmc-supply = <&vmmc>;
    vqmmc-supply = <&reg_pmic_ldo1>;
    bus-width = <4>;
    sd-uhs-sdr50;
    sd-uhs-sdr104;
    full-pwr-cycle;
    mux-states = <&sdhi0_mux 1>;
};

I do not believe however that this is adding any value.

RE: [PATCH v2 1/2] dt-bindings: mmc: renesas,sdhi: Add mux-states property

[Biju Das]

Hi Josua Mayer,

Thanks for the feedback.

> -----Original Message-----
> From: Josua Mayer <josua@solid-run.com>
> Sent: 01 December 2025 14:01
> Subject: Re: [PATCH v2 1/2] dt-bindings: mmc: renesas,sdhi: Add mux-states property
> 
> Hi Biju,
> 
> Am 01.12.25 um 14:02 schrieb Biju Das:
> > Hi Josua Mayer,
> >
> >
> > Thanks for the patch.
> >
> >> -----Original Message-----
> >> From: Josua Mayer <josua@solid-run.com>
> >> Sent: 01 December 2025 12:31
> >> Subject: [PATCH v2 1/2] dt-bindings: mmc: renesas,sdhi: Add
> >> mux-states property
> >>
> >> Add mux controller support for when sdio lines are muxed between a host and multiple cards.
> >>
> >> There are several devices supporting a choice of eMMC or SD on a
> >> single board by both dip switch and gpio, e.g. Renesas RZ/G2L SMARC SoM and SolidRun RZ/G2L SoM.
> >>
> >> In-tree dts for the Renesas boards currently rely on preprocessor
> >> macros to hog gpios and define the card.
> >>
> >> By adding mux-states property to sdio controller description, boards
> >> can correctly describe the mux that already exists in hardware - and
> >> drivers can coordinate between mux selection and probing for cards.
> >> Signed-off-by: Josua Mayer <josua@solid-run.com>
> >> ---
> >>  Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml | 6 ++++++
> >>  1 file changed, 6 insertions(+)
> >>
> >> diff --git a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> >> b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> >> index c754ea71f51f7..754ccb1c30efb 100644
> >> --- a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> >> +++ b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> >> @@ -106,6 +106,11 @@ properties:
> >>    iommus:
> >>      maxItems: 1
> >>
> >> +  mux-states:
> >> +    description:
> >> +      mux controller node to route the SDIO signals from SoC to cards.
> > Maybe describe 0 - state for SD and 1 - state for eMMC ??
> The state-mmc-sd mapping depends on the wiring and description of the mux, so we cannot differentiate
> in the mux-states property.

OK. Agreed.

> 
> The idea is that whatever numeric argument you add to the phandle will select a particular state for
> the mux.
> The rest of dt properties must still be specific to SD or eMMC, e.g.:
> 
> &sdhi0 {
>     pinctrl-0 = <&sdhi0_pins>, <&sdhi0_cd_pins>;
>     pinctrl-1 = <&sdhi0_uhs_pins>, <&sdhi0_cd_pins>;
>     pinctrl-names = "default", "state_uhs";
>     vmmc-supply = <&vmmc>;
>     vqmmc-supply = <&reg_pmic_ldo1>;
>     bus-width = <4>;
>     sd-uhs-sdr50;
>     sd-uhs-sdr104;
>     full-pwr-cycle;
>     mux-states = <&sdhi0_mux 1>;
>     status = "okay";
> };
> 
> >
> >
> >> +    maxItems: 1
> >> +
> >>    power-domains:
> >>      maxItems: 1
> >>
> >> @@ -275,6 +280,7 @@ examples:
> >>          max-frequency = <195000000>;
> >>          power-domains = <&sysc R8A7790_PD_ALWAYS_ON>;
> >>          resets = <&cpg 314>;
> >> +        mux-states = <&mux 0>;
> > On R-Car mmc/sd mux available only on SD2/SD3, so I guess you picked wrong node SD0??
> As they were examples I did not give it much thought, We could even omit mux-states from examples
> completely since it is optional.

OK, I guess that is also fine, instead of the wrong usage in sd0 device node.

> >
> > or
> >
> > Add separate example with RZ/G2L, as the boards(RZ/G2L SMARC EVK) use
> > gpio/switch (XOR function) to select the eMMC or SD signals??
> 
> I can if so desired add verbose example from our own G2L based board, e.g.:
> 
> sdhi0: mmc@11c00000 {
>     compatible = "renesas,sdhi-r9a07g044", "renesas,rzg2l-sdhi";
>     reg = <0x0 0x11c00000 0 0x10000>;
>     interrupts = <GIC_SPI 104 IRQ_TYPE_LEVEL_HIGH>,
>              <GIC_SPI 105 IRQ_TYPE_LEVEL_HIGH>;
>     clocks = <&cpg CPG_MOD R9A07G044_SDHI0_IMCLK>,
>          <&cpg CPG_MOD R9A07G044_SDHI0_CLK_HS>,
>          <&cpg CPG_MOD R9A07G044_SDHI0_IMCLK2>,
>          <&cpg CPG_MOD R9A07G044_SDHI0_ACLK>;
>     clock-names = "core", "clkh", "cd", "aclk";
>     resets = <&cpg R9A07G044_SDHI0_IXRST>;
>     power-domains = <&cpg>;
>     pinctrl-0 = <&sdhi0_pins>, <&sdhi0_cd_pins>;
>     pinctrl-1 = <&sdhi0_uhs_pins>, <&sdhi0_cd_pins>;
>     pinctrl-names = "default", "state_uhs";
>     vmmc-supply = <&vmmc>;
>     vqmmc-supply = <&reg_pmic_ldo1>;
>     bus-width = <4>;
>     sd-uhs-sdr50;
>     sd-uhs-sdr104;
>     full-pwr-cycle;
>     mux-states = <&sdhi0_mux 1>;
> };
> 
> I do not believe however that this is adding any value.

Maybe Wolfram/Geert can comment on this. Looks like SD mux is an important feature of the SoC.

Previously I got feedback to provide real example. This rule might have changed??

Cheers,
Biju

Re: [PATCH v2 1/2] dt-bindings: mmc: renesas,sdhi: Add mux-states property

[Rob Herring (Arm)]

On Mon, 01 Dec 2025 13:31:10 +0100, Josua Mayer wrote:
> Add mux controller support for when sdio lines are muxed between a host
> and multiple cards.
> 
> There are several devices supporting a choice of eMMC or SD on a single
> board by both dip switch and gpio, e.g. Renesas RZ/G2L SMARC SoM and
> SolidRun RZ/G2L SoM.
> 
> In-tree dts for the Renesas boards currently rely on preprocessor macros
> to hog gpios and define the card.
> 
> By adding mux-states property to sdio controller description, boards can
> correctly describe the mux that already exists in hardware - and drivers
> can coordinate between mux selection and probing for cards.
> 
> Signed-off-by: Josua Mayer <josua@solid-run.com>
> ---
>  Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml | 6 ++++++
>  1 file changed, 6 insertions(+)
> 

Acked-by: Rob Herring (Arm) <robh@kernel.org>

Re: [PATCH v2 1/2] dt-bindings: mmc: renesas,sdhi: Add mux-states property

[Ulf Hansson]

On Mon, 1 Dec 2025 at 13:31, Josua Mayer <josua@solid-run.com> wrote:
>
> Add mux controller support for when sdio lines are muxed between a host
> and multiple cards.

We have the SD, SDIO, (e)MMC interfaces with their corresponding pins.
Using "sdio lines" becomes confusing as it kind of indicates this is
for "SDIO" only.

>
> There are several devices supporting a choice of eMMC or SD on a single
> board by both dip switch and gpio, e.g. Renesas RZ/G2L SMARC SoM and
> SolidRun RZ/G2L SoM.
>
> In-tree dts for the Renesas boards currently rely on preprocessor macros
> to hog gpios and define the card.
>
> By adding mux-states property to sdio controller description, boards can

Again, please don't use "sdio" here.

> correctly describe the mux that already exists in hardware - and drivers
> can coordinate between mux selection and probing for cards.
>
> Signed-off-by: Josua Mayer <josua@solid-run.com>
> ---
>  Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml | 6 ++++++
>  1 file changed, 6 insertions(+)
>
> diff --git a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> index c754ea71f51f7..754ccb1c30efb 100644
> --- a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> +++ b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> @@ -106,6 +106,11 @@ properties:
>    iommus:
>      maxItems: 1
>
> +  mux-states:
> +    description:
> +      mux controller node to route the SDIO signals from SoC to cards.

I suggest we change from "SDIO" to "SD/SDIO/eMMC".

> +    maxItems: 1
> +
>    power-domains:
>      maxItems: 1
>
> @@ -275,6 +280,7 @@ examples:
>          max-frequency = <195000000>;
>          power-domains = <&sysc R8A7790_PD_ALWAYS_ON>;
>          resets = <&cpg 314>;
> +        mux-states = <&mux 0>;
>      };
>
>      sdhi1: mmc@ee120000 {
>
> --
> 2.51.0
>
>

Kind regards
Uffe

Re: [PATCH v2 1/2] dt-bindings: mmc: renesas,sdhi: Add mux-states property

[Josua Mayer]

Am 09.12.25 um 13:08 schrieb Ulf Hansson:

> On Mon, 1 Dec 2025 at 13:31, Josua Mayer <josua@solid-run.com> wrote:
>> Add mux controller support for when sdio lines are muxed between a host
>> and multiple cards.
> We have the SD, SDIO, (e)MMC interfaces with their corresponding pins.
> Using "sdio lines" becomes confusing as it kind of indicates this is
> for "SDIO" only.
Agreed.
>
>> There are several devices supporting a choice of eMMC or SD on a single
>> board by both dip switch and gpio, e.g. Renesas RZ/G2L SMARC SoM and
>> SolidRun RZ/G2L SoM.
>>
>> In-tree dts for the Renesas boards currently rely on preprocessor macros
>> to hog gpios and define the card.
>>
>> By adding mux-states property to sdio controller description, boards can
> Again, please don't use "sdio" here.
Ack.
>
>> correctly describe the mux that already exists in hardware - and drivers
>> can coordinate between mux selection and probing for cards.
>>
>> Signed-off-by: Josua Mayer <josua@solid-run.com>
>> ---
>>  Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml | 6 ++++++
>>  1 file changed, 6 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
>> index c754ea71f51f7..754ccb1c30efb 100644
>> --- a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
>> +++ b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
>> @@ -106,6 +106,11 @@ properties:
>>    iommus:
>>      maxItems: 1
>>
>> +  mux-states:
>> +    description:
>> +      mux controller node to route the SDIO signals from SoC to cards.
> I suggest we change from "SDIO" to "SD/SDIO/eMMC".

Agreed.

I will replace here, and reword each commit description for v3.

>
>> +    maxItems: 1
>> +
>>    power-domains:
>>      maxItems: 1
>>
>> @@ -275,6 +280,7 @@ examples:
>>          max-frequency = <195000000>;
>>          power-domains = <&sysc R8A7790_PD_ALWAYS_ON>;
>>          resets = <&cpg 314>;
>> +        mux-states = <&mux 0>;
>>      };
>>
>>      sdhi1: mmc@ee120000 {
>>
>> --
>> 2.51.0
>>
>>
> Kind regards
> Uffe
>
>

[PATCH v2 2/2] mmc: host: renesas_sdhi_core: support configuring an optional sdio mux

[Josua Mayer]

Some hardware designs route sdio signals through a mux to support
multiple devices on a single sdio controller.
In particular SolidRun RZ/G2L/G2LC/V2L System on Module use a mux for
switching between soldered eMMC and an optional microSD on a carrier
board, e.g. for development or provisioning.

SDIO is not well suited for runtime switching between different cards,
however boot-time selection is possible and useful in particular with dt
overlays.

Add support for an optional sdio mux defined in dt and select it during
probe.

Similar functionality already exists in other places, e.g. i2c-omap.

Signed-off-by: Josua Mayer <josua@solid-run.com>
---
 drivers/mmc/host/Kconfig             |  1 +
 drivers/mmc/host/renesas_sdhi.h      |  1 +
 drivers/mmc/host/renesas_sdhi_core.c | 27 ++++++++++++++++++++++++---
 3 files changed, 26 insertions(+), 3 deletions(-)

diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index 2c963cb6724b9..c01ab7d81a5af 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -707,6 +707,7 @@ config MMC_SDHI
 	depends on SUPERH || ARCH_RENESAS || COMPILE_TEST
 	depends on (RESET_CONTROLLER && REGULATOR) || !OF
 	select MMC_TMIO_CORE
+	select MULTIPLEXER
 	help
 	  This provides support for the SDHI SD/SDIO controller found in
 	  Renesas SuperH, ARM and ARM64 based SoCs
diff --git a/drivers/mmc/host/renesas_sdhi.h b/drivers/mmc/host/renesas_sdhi.h
index 084964cecf9d8..9508908d8179f 100644
--- a/drivers/mmc/host/renesas_sdhi.h
+++ b/drivers/mmc/host/renesas_sdhi.h
@@ -97,6 +97,7 @@ struct renesas_sdhi {
 	struct reset_control *rstc;
 	struct tmio_mmc_host *host;
 	struct regulator_dev *rdev;
+	struct mux_state *mux_state;
 };
 
 #define host_to_priv(host) \
diff --git a/drivers/mmc/host/renesas_sdhi_core.c b/drivers/mmc/host/renesas_sdhi_core.c
index f56fa2cd208dd..c58f412ea2028 100644
--- a/drivers/mmc/host/renesas_sdhi_core.c
+++ b/drivers/mmc/host/renesas_sdhi_core.c
@@ -26,6 +26,7 @@
 #include <linux/mmc/mmc.h>
 #include <linux/mmc/slot-gpio.h>
 #include <linux/module.h>
+#include <linux/mux/consumer.h>
 #include <linux/pinctrl/consumer.h>
 #include <linux/pinctrl/pinctrl-state.h>
 #include <linux/platform_data/tmio.h>
@@ -1061,6 +1062,7 @@ int renesas_sdhi_probe(struct platform_device *pdev,
 	struct regulator_dev *rdev;
 	struct renesas_sdhi_dma *dma_priv;
 	struct device *dev = &pdev->dev;
+	struct device_node *node = pdev->dev.of_node;
 	struct tmio_mmc_host *host;
 	struct renesas_sdhi *priv;
 	int num_irqs, irq, ret, i;
@@ -1115,9 +1117,25 @@ int renesas_sdhi_probe(struct platform_device *pdev,
 						"state_uhs");
 	}
 
+	if (of_property_present(node, "mux-states")) {
+		priv->mux_state = devm_mux_state_get(&pdev->dev, NULL);
+		if (IS_ERR(priv->mux_state)) {
+			ret = PTR_ERR(priv->mux_state);
+			dev_dbg(&pdev->dev, "failed to get SDIO mux: %d\n", ret);
+			return ret;
+		}
+		ret = mux_state_select(priv->mux_state);
+		if (ret) {
+			dev_err(&pdev->dev, "failed to select SDIO mux: %d\n", ret);
+			return ret;
+		}
+	}
+
 	host = tmio_mmc_host_alloc(pdev, mmc_data);
-	if (IS_ERR(host))
-		return PTR_ERR(host);
+	if (IS_ERR(host)) {
+		ret = PTR_ERR(host);
+		goto edselmux;
+	}
 
 	priv->host = host;
 
@@ -1200,7 +1218,7 @@ int renesas_sdhi_probe(struct platform_device *pdev,
 
 	ret = renesas_sdhi_clk_enable(host);
 	if (ret)
-		return ret;
+		goto edselmux;
 
 	rcfg.of_node = of_get_available_child_by_name(dev->of_node, "vqmmc-regulator");
 	if (rcfg.of_node) {
@@ -1304,6 +1322,9 @@ int renesas_sdhi_probe(struct platform_device *pdev,
 
 edisclk:
 	renesas_sdhi_clk_disable(host);
+edselmux:
+	if (priv->mux_state)
+		mux_state_deselect(priv->mux_state);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(renesas_sdhi_probe);

-- 
2.51.0

Re: [PATCH v2 2/2] mmc: host: renesas_sdhi_core: support configuring an optional sdio mux

[Rob Herring]

On Mon, Dec 01, 2025 at 01:31:11PM +0100, Josua Mayer wrote:
> Some hardware designs route sdio signals through a mux to support
> multiple devices on a single sdio controller.
> In particular SolidRun RZ/G2L/G2LC/V2L System on Module use a mux for
> switching between soldered eMMC and an optional microSD on a carrier
> board, e.g. for development or provisioning.
> 
> SDIO is not well suited for runtime switching between different cards,
> however boot-time selection is possible and useful in particular with dt
> overlays.
> 
> Add support for an optional sdio mux defined in dt and select it during
> probe.
> 
> Similar functionality already exists in other places, e.g. i2c-omap.
> 
> Signed-off-by: Josua Mayer <josua@solid-run.com>
> ---
>  drivers/mmc/host/Kconfig             |  1 +
>  drivers/mmc/host/renesas_sdhi.h      |  1 +
>  drivers/mmc/host/renesas_sdhi_core.c | 27 ++++++++++++++++++++++++---
>  3 files changed, 26 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
> index 2c963cb6724b9..c01ab7d81a5af 100644
> --- a/drivers/mmc/host/Kconfig
> +++ b/drivers/mmc/host/Kconfig
> @@ -707,6 +707,7 @@ config MMC_SDHI
>  	depends on SUPERH || ARCH_RENESAS || COMPILE_TEST
>  	depends on (RESET_CONTROLLER && REGULATOR) || !OF
>  	select MMC_TMIO_CORE
> +	select MULTIPLEXER
>  	help
>  	  This provides support for the SDHI SD/SDIO controller found in
>  	  Renesas SuperH, ARM and ARM64 based SoCs
> diff --git a/drivers/mmc/host/renesas_sdhi.h b/drivers/mmc/host/renesas_sdhi.h
> index 084964cecf9d8..9508908d8179f 100644
> --- a/drivers/mmc/host/renesas_sdhi.h
> +++ b/drivers/mmc/host/renesas_sdhi.h
> @@ -97,6 +97,7 @@ struct renesas_sdhi {
>  	struct reset_control *rstc;
>  	struct tmio_mmc_host *host;
>  	struct regulator_dev *rdev;
> +	struct mux_state *mux_state;
>  };
>  
>  #define host_to_priv(host) \
> diff --git a/drivers/mmc/host/renesas_sdhi_core.c b/drivers/mmc/host/renesas_sdhi_core.c
> index f56fa2cd208dd..c58f412ea2028 100644
> --- a/drivers/mmc/host/renesas_sdhi_core.c
> +++ b/drivers/mmc/host/renesas_sdhi_core.c
> @@ -26,6 +26,7 @@
>  #include <linux/mmc/mmc.h>
>  #include <linux/mmc/slot-gpio.h>
>  #include <linux/module.h>
> +#include <linux/mux/consumer.h>
>  #include <linux/pinctrl/consumer.h>
>  #include <linux/pinctrl/pinctrl-state.h>
>  #include <linux/platform_data/tmio.h>
> @@ -1061,6 +1062,7 @@ int renesas_sdhi_probe(struct platform_device *pdev,
>  	struct regulator_dev *rdev;
>  	struct renesas_sdhi_dma *dma_priv;
>  	struct device *dev = &pdev->dev;
> +	struct device_node *node = pdev->dev.of_node;
>  	struct tmio_mmc_host *host;
>  	struct renesas_sdhi *priv;
>  	int num_irqs, irq, ret, i;
> @@ -1115,9 +1117,25 @@ int renesas_sdhi_probe(struct platform_device *pdev,
>  						"state_uhs");
>  	}
>  
> +	if (of_property_present(node, "mux-states")) {
> +		priv->mux_state = devm_mux_state_get(&pdev->dev, NULL);

These 2 lines are really just devm_mux_state_get_optional(). Add it if 
it doesn't exist.

> +		if (IS_ERR(priv->mux_state)) {
> +			ret = PTR_ERR(priv->mux_state);
> +			dev_dbg(&pdev->dev, "failed to get SDIO mux: %d\n", ret);
> +			return ret;
> +		}
> +		ret = mux_state_select(priv->mux_state);
> +		if (ret) {
> +			dev_err(&pdev->dev, "failed to select SDIO mux: %d\n", ret);

Is there a case where mux_state_select() failing is not an error? If 
not, mux_state_select() should do the printing.

> +			return ret;
> +		}
> +	}

Re: [PATCH v2 2/2] mmc: host: renesas_sdhi_core: support configuring an optional sdio mux

[Josua Mayer]

Am 08.12.25 um 02:24 schrieb Rob Herring:
> On Mon, Dec 01, 2025 at 01:31:11PM +0100, Josua Mayer wrote:
>> Some hardware designs route sdio signals through a mux to support
>> multiple devices on a single sdio controller.
>> In particular SolidRun RZ/G2L/G2LC/V2L System on Module use a mux for
>> switching between soldered eMMC and an optional microSD on a carrier
>> board, e.g. for development or provisioning.
>>
>> SDIO is not well suited for runtime switching between different cards,
>> however boot-time selection is possible and useful in particular with dt
>> overlays.
>>
>> Add support for an optional sdio mux defined in dt and select it during
>> probe.
>>
>> Similar functionality already exists in other places, e.g. i2c-omap.
>>
>> Signed-off-by: Josua Mayer <josua@solid-run.com>
>> ---
>>  drivers/mmc/host/Kconfig             |  1 +
>>  drivers/mmc/host/renesas_sdhi.h      |  1 +
>>  drivers/mmc/host/renesas_sdhi_core.c | 27 ++++++++++++++++++++++++---
>>  3 files changed, 26 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
>> index 2c963cb6724b9..c01ab7d81a5af 100644
>> --- a/drivers/mmc/host/Kconfig
>> +++ b/drivers/mmc/host/Kconfig
>> @@ -707,6 +707,7 @@ config MMC_SDHI
>>  	depends on SUPERH || ARCH_RENESAS || COMPILE_TEST
>>  	depends on (RESET_CONTROLLER && REGULATOR) || !OF
>>  	select MMC_TMIO_CORE
>> +	select MULTIPLEXER
>>  	help
>>  	  This provides support for the SDHI SD/SDIO controller found in
>>  	  Renesas SuperH, ARM and ARM64 based SoCs
>> diff --git a/drivers/mmc/host/renesas_sdhi.h b/drivers/mmc/host/renesas_sdhi.h
>> index 084964cecf9d8..9508908d8179f 100644
>> --- a/drivers/mmc/host/renesas_sdhi.h
>> +++ b/drivers/mmc/host/renesas_sdhi.h
>> @@ -97,6 +97,7 @@ struct renesas_sdhi {
>>  	struct reset_control *rstc;
>>  	struct tmio_mmc_host *host;
>>  	struct regulator_dev *rdev;
>> +	struct mux_state *mux_state;
>>  };
>>  
>>  #define host_to_priv(host) \
>> diff --git a/drivers/mmc/host/renesas_sdhi_core.c b/drivers/mmc/host/renesas_sdhi_core.c
>> index f56fa2cd208dd..c58f412ea2028 100644
>> --- a/drivers/mmc/host/renesas_sdhi_core.c
>> +++ b/drivers/mmc/host/renesas_sdhi_core.c
>> @@ -26,6 +26,7 @@
>>  #include <linux/mmc/mmc.h>
>>  #include <linux/mmc/slot-gpio.h>
>>  #include <linux/module.h>
>> +#include <linux/mux/consumer.h>
>>  #include <linux/pinctrl/consumer.h>
>>  #include <linux/pinctrl/pinctrl-state.h>
>>  #include <linux/platform_data/tmio.h>
>> @@ -1061,6 +1062,7 @@ int renesas_sdhi_probe(struct platform_device *pdev,
>>  	struct regulator_dev *rdev;
>>  	struct renesas_sdhi_dma *dma_priv;
>>  	struct device *dev = &pdev->dev;
>> +	struct device_node *node = pdev->dev.of_node;
>>  	struct tmio_mmc_host *host;
>>  	struct renesas_sdhi *priv;
>>  	int num_irqs, irq, ret, i;
>> @@ -1115,9 +1117,25 @@ int renesas_sdhi_probe(struct platform_device *pdev,
>>  						"state_uhs");
>>  	}
>>  
>> +	if (of_property_present(node, "mux-states")) {
>> +		priv->mux_state = devm_mux_state_get(&pdev->dev, NULL);
> These 2 lines are really just devm_mux_state_get_optional(). Add it if 
> it doesn't exist.
Other drivers have so far avoided introducing a general helper for this.
Notably drivers/phy/phy-can-transceiver.c has already implemented
its own version of devm_mux_state_get_optional.
>
>> +		if (IS_ERR(priv->mux_state)) {
>> +			ret = PTR_ERR(priv->mux_state);
>> +			dev_dbg(&pdev->dev, "failed to get SDIO mux: %d\n", ret);
>> +			return ret;
>> +		}
>> +		ret = mux_state_select(priv->mux_state);
>> +		if (ret) {
>> +			dev_err(&pdev->dev, "failed to select SDIO mux: %d\n", ret);
> Is there a case where mux_state_select() failing is not an error?
I suppose it is always an error.
> If 
> not, mux_state_select() should do the printing.

Possibly. Except when it is EDEFER.

Inside driver we would use dev_err_probe in this case.
Should helpers do the same?

>
>> +			return ret;
>> +		}
>> +	}

RE: [PATCH v2 0/2] mmc: host: renesas_sdhi_core: support configuring an optional sdio mux

[Biju Das]

> -----Original Message-----
> From: Josua Mayer <josua@solid-run.com>
> Sent: 01 December 2025 12:31
> Subject: [PATCH v2 0/2] mmc: host: renesas_sdhi_core: support configuring an optional sdio mux
> 
> Some Renesas SoC based boards mux SD and eMMC on a single sdio controller, exposing user control by
> dip switch and software control by gpio.
> 
> Purpose is to simplify development and provisioning by selecting boot media at power-on, and again
> before starting linux.

Does it mean, you boot with eSD mode, boot the linux with SD mode and flash the image to eMMC??

and then, switch to eMMC boot mode and boot everything from eMMC??


Or

boot with eSD mode, flash emmc images at initramfs level/u-boot level

and then, switch to eMMC boot mode and boot everything from eMMC??


Cheers,
Biju

Re: [PATCH v2 0/2] mmc: host: renesas_sdhi_core: support configuring an optional sdio mux

[Josua Mayer]

Am 01.12.25 um 14:17 schrieb Biju Das:
>
>> -----Original Message-----
>> From: Josua Mayer <josua@solid-run.com>
>> Sent: 01 December 2025 12:31
>> Subject: [PATCH v2 0/2] mmc: host: renesas_sdhi_core: support configuring an optional sdio mux
>>
>> Some Renesas SoC based boards mux SD and eMMC on a single sdio controller, exposing user control by
>> dip switch and software control by gpio.
>>
>> Purpose is to simplify development and provisioning by selecting boot media at power-on, and again
>> before starting linux.
> Does it mean, you boot with eSD mode, boot the linux with SD mode and flash the image to eMMC??
>
> and then, switch to eMMC boot mode and boot everything from eMMC??
>
>
> Or
>
> boot with eSD mode, flash emmc images at initramfs level/u-boot level
>
> and then, switch to eMMC boot mode and boot everything from eMMC??

The latter.

DIP switch decides bootloader starting from eMMC or SD.
Then u-boot has complete freedom.

Finally U-Boot shall load (and patch) DTB for either eMMC or SD.
Regardless of whether rootfs is initramfs or physical media.

RE: [PATCH v2 0/2] mmc: host: renesas_sdhi_core: support configuring an optional sdio mux

[Biju Das]

Hi Josua Mayer,

> -----Original Message-----
> From: Josua Mayer <josua@solid-run.com>
> Sent: 01 December 2025 14:21
> Subject: Re: [PATCH v2 0/2] mmc: host: renesas_sdhi_core: support configuring an optional sdio mux
> 
> Am 01.12.25 um 14:17 schrieb Biju Das:
> >
> >> -----Original Message-----
> >> From: Josua Mayer <josua@solid-run.com>
> >> Sent: 01 December 2025 12:31
> >> Subject: [PATCH v2 0/2] mmc: host: renesas_sdhi_core: support
> >> configuring an optional sdio mux
> >>
> >> Some Renesas SoC based boards mux SD and eMMC on a single sdio
> >> controller, exposing user control by dip switch and software control by gpio.
> >>
> >> Purpose is to simplify development and provisioning by selecting boot
> >> media at power-on, and again before starting linux.
> > Does it mean, you boot with eSD mode, boot the linux with SD mode and flash the image to eMMC??
> >
> > and then, switch to eMMC boot mode and boot everything from eMMC??
> >
> >
> > Or
> >
> > boot with eSD mode, flash emmc images at initramfs level/u-boot level
> >
> > and then, switch to eMMC boot mode and boot everything from eMMC??
> 
> The latter.
> 
> DIP switch decides bootloader starting from eMMC or SD.
> Then u-boot has complete freedom.
> 
> Finally U-Boot shall load (and patch) DTB for either eMMC or SD.
> Regardless of whether rootfs is initramfs or physical media.

OK, it is clear now.

Cheers,
Biju