[PATCH v1] phy: fsl-imx8mq-usb: fix typec orientation switch when built as module

[PATCH v1] phy: fsl-imx8mq-usb: fix typec orientation switch when built as module

[Franz Schnyder]

From: Franz Schnyder <franz.schnyder@toradex.com>

Currently, the PHY only registers the typec orientation switch when it
is built in. If the typec driver is built as a module, the switch
registration is skipped due to the preprocessor condition, causing
orientation detection to fail.

This patch replaces the preprocessor condition so that the orientation
switch is correctly registered for both built-in and module builds.

Fixes: b58f0f86fd61 ("phy: fsl-imx8mq-usb: add tca function driver for imx95")
Cc: stable@vger.kernel.org
Signed-off-by: Franz Schnyder <franz.schnyder@toradex.com>
---
 drivers/phy/freescale/phy-fsl-imx8mq-usb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/phy/freescale/phy-fsl-imx8mq-usb.c b/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
index b94f242420fc..d498a6b7234b 100644
--- a/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
+++ b/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
@@ -124,7 +124,7 @@ struct imx8mq_usb_phy {
 static void tca_blk_orientation_set(struct tca_blk *tca,
 				enum typec_orientation orientation);
 
-#ifdef CONFIG_TYPEC
+#if IS_ENABLED(CONFIG_TYPEC)
 
 static int tca_blk_typec_switch_set(struct typec_switch_dev *sw,
 				enum typec_orientation orientation)
-- 
2.43.0

Re: [PATCH v1] phy: fsl-imx8mq-usb: fix typec orientation switch when built as module

[Neil Armstrong]

On 11/24/25 10:50, Franz Schnyder wrote:
> From: Franz Schnyder <franz.schnyder@toradex.com>
> 
> Currently, the PHY only registers the typec orientation switch when it
> is built in. If the typec driver is built as a module, the switch
> registration is skipped due to the preprocessor condition, causing
> orientation detection to fail.
> 
> This patch replaces the preprocessor condition so that the orientation
> switch is correctly registered for both built-in and module builds.
> 
> Fixes: b58f0f86fd61 ("phy: fsl-imx8mq-usb: add tca function driver for imx95")
> Cc: stable@vger.kernel.org
> Signed-off-by: Franz Schnyder <franz.schnyder@toradex.com>
> ---
>   drivers/phy/freescale/phy-fsl-imx8mq-usb.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/phy/freescale/phy-fsl-imx8mq-usb.c b/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> index b94f242420fc..d498a6b7234b 100644
> --- a/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> +++ b/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> @@ -124,7 +124,7 @@ struct imx8mq_usb_phy {
>   static void tca_blk_orientation_set(struct tca_blk *tca,
>   				enum typec_orientation orientation);
>   
> -#ifdef CONFIG_TYPEC
> +#if IS_ENABLED(CONFIG_TYPEC)
>   
>   static int tca_blk_typec_switch_set(struct typec_switch_dev *sw,
>   				enum typec_orientation orientation)

Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>

Re: [PATCH v1] phy: fsl-imx8mq-usb: fix typec orientation switch when built as module

[Xu Yang]

Hi Franz,

On Mon, Nov 24, 2025 at 10:50:04AM +0100, Franz Schnyder wrote:
> From: Franz Schnyder <franz.schnyder@toradex.com>
> 
> Currently, the PHY only registers the typec orientation switch when it
> is built in. If the typec driver is built as a module, the switch
> registration is skipped due to the preprocessor condition, causing
> orientation detection to fail.
> 
> This patch replaces the preprocessor condition so that the orientation
> switch is correctly registered for both built-in and module builds.
> 
> Fixes: b58f0f86fd61 ("phy: fsl-imx8mq-usb: add tca function driver for imx95")
> Cc: stable@vger.kernel.org
> Signed-off-by: Franz Schnyder <franz.schnyder@toradex.com>
> ---
>  drivers/phy/freescale/phy-fsl-imx8mq-usb.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/phy/freescale/phy-fsl-imx8mq-usb.c b/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> index b94f242420fc..d498a6b7234b 100644
> --- a/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> +++ b/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> @@ -124,7 +124,7 @@ struct imx8mq_usb_phy {
>  static void tca_blk_orientation_set(struct tca_blk *tca,
>  				enum typec_orientation orientation);
>  
> -#ifdef CONFIG_TYPEC
> +#if IS_ENABLED(CONFIG_TYPEC)

With below commit:

45fe729be9a6 usb: typec: Stub out typec_switch APIs when CONFIG_TYPEC=n

I think this #if/else/endif condition can be removed.

Thanks,
Xu Yang

>  
>  static int tca_blk_typec_switch_set(struct typec_switch_dev *sw,
>  				enum typec_orientation orientation)
> -- 
> 2.43.0
> 

Re: [PATCH v1] phy: fsl-imx8mq-usb: fix typec orientation switch when built as module

[Francesco Dolcini]

On Tue, Nov 25, 2025 at 03:04:43PM +0800, Xu Yang wrote:
> On Mon, Nov 24, 2025 at 10:50:04AM +0100, Franz Schnyder wrote:
> > From: Franz Schnyder <franz.schnyder@toradex.com>
> > 
> > Currently, the PHY only registers the typec orientation switch when it
> > is built in. If the typec driver is built as a module, the switch
> > registration is skipped due to the preprocessor condition, causing
> > orientation detection to fail.
> > 
> > This patch replaces the preprocessor condition so that the orientation
> > switch is correctly registered for both built-in and module builds.
> > 
> > Fixes: b58f0f86fd61 ("phy: fsl-imx8mq-usb: add tca function driver for imx95")
> > Cc: stable@vger.kernel.org
> > Signed-off-by: Franz Schnyder <franz.schnyder@toradex.com>
> > ---
> >  drivers/phy/freescale/phy-fsl-imx8mq-usb.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/drivers/phy/freescale/phy-fsl-imx8mq-usb.c b/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> > index b94f242420fc..d498a6b7234b 100644
> > --- a/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> > +++ b/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> > @@ -124,7 +124,7 @@ struct imx8mq_usb_phy {
> >  static void tca_blk_orientation_set(struct tca_blk *tca,
> >  				enum typec_orientation orientation);
> >  
> > -#ifdef CONFIG_TYPEC
> > +#if IS_ENABLED(CONFIG_TYPEC)
> 
> With below commit:
> 
> 45fe729be9a6 usb: typec: Stub out typec_switch APIs when CONFIG_TYPEC=n
> 
> I think this #if/else/endif condition can be removed.

This patch should go to stable, and that commit is not present in any
such previous kernel.

Should we have 2 patches or "force" 45fe729be9a6 to be also backported?

What's the general advise in these situations?

Francesco

Re: [PATCH v1] phy: fsl-imx8mq-usb: fix typec orientation switch when built as module

[Xu Yang]

Hi Francesco,

On Tue, Nov 25, 2025 at 08:20:11AM +0100, Francesco Dolcini wrote:
> On Tue, Nov 25, 2025 at 03:04:43PM +0800, Xu Yang wrote:
> > On Mon, Nov 24, 2025 at 10:50:04AM +0100, Franz Schnyder wrote:
> > > From: Franz Schnyder <franz.schnyder@toradex.com>
> > > 
> > > Currently, the PHY only registers the typec orientation switch when it
> > > is built in. If the typec driver is built as a module, the switch
> > > registration is skipped due to the preprocessor condition, causing
> > > orientation detection to fail.
> > > 
> > > This patch replaces the preprocessor condition so that the orientation
> > > switch is correctly registered for both built-in and module builds.
> > > 
> > > Fixes: b58f0f86fd61 ("phy: fsl-imx8mq-usb: add tca function driver for imx95")
> > > Cc: stable@vger.kernel.org
> > > Signed-off-by: Franz Schnyder <franz.schnyder@toradex.com>
> > > ---
> > >  drivers/phy/freescale/phy-fsl-imx8mq-usb.c | 2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/phy/freescale/phy-fsl-imx8mq-usb.c b/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> > > index b94f242420fc..d498a6b7234b 100644
> > > --- a/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> > > +++ b/drivers/phy/freescale/phy-fsl-imx8mq-usb.c
> > > @@ -124,7 +124,7 @@ struct imx8mq_usb_phy {
> > >  static void tca_blk_orientation_set(struct tca_blk *tca,
> > >  				enum typec_orientation orientation);
> > >  
> > > -#ifdef CONFIG_TYPEC
> > > +#if IS_ENABLED(CONFIG_TYPEC)
> > 
> > With below commit:
> > 
> > 45fe729be9a6 usb: typec: Stub out typec_switch APIs when CONFIG_TYPEC=n
> > 
> > I think this #if/else/endif condition can be removed.
> 
> This patch should go to stable, and that commit is not present in any
> such previous kernel.

Well, commit 45fe729be9a6 was merged in v6.18-rc2. Previous stable kernel
doesn't have it yet.

> 
> Should we have 2 patches or "force" 45fe729be9a6 to be also backported?
> 
> What's the general advise in these situations?

I am not quite sure either.
But commit b58f0f86fd61 was merged in v6.15-rc2. So the stable version to be fixed
will be v6.18, right? Then commit 45fe729be9a6 will be included already. If so,
I think only 1 patch will be enough.

Thanks,
Xu Yang

> 
> Francesco
>