diff mbox series

[v3,1/3] usb: misc: onboard-hub: support multiple power supplies

Message ID 20230620-hx3-v3-1-2acbc03ca949@skidata.com (mailing list archive)
State Superseded
Headers show
Series usb: misc: onboard_usb_hub: add support for Cypress HX3 USB 3.0 family | expand

Commit Message

Benjamin Bara June 21, 2023, 3:58 p.m. UTC 
From: Benjamin Bara <benjamin.bara@skidata.com>

As some of the onboard hubs require multiple power supplies, provide the
environment to support them.

Signed-off-by: Benjamin Bara <benjamin.bara@skidata.com>
---
v3:
- fix nits mentioned in v2

v2:
- replace (err != 0) with (err)
---
 drivers/usb/misc/onboard_usb_hub.c | 39 ++++++++++++++++++++++++++++++--------
 drivers/usb/misc/onboard_usb_hub.h |  1 +
 2 files changed, 32 insertions(+), 8 deletions(-)

Comments

gregkh@linuxfoundation.org June 21, 2023, 4:06 p.m. UTC | #1 
On Wed, Jun 21, 2023 at 05:58:30PM +0200, Benjamin Bara wrote:
> From: Benjamin Bara <benjamin.bara@skidata.com>
> 
> As some of the onboard hubs require multiple power supplies, provide the
> environment to support them.
> 
> Signed-off-by: Benjamin Bara <benjamin.bara@skidata.com>
> ---
> v3:
> - fix nits mentioned in v2
> 
> v2:
> - replace (err != 0) with (err)
> ---
>  drivers/usb/misc/onboard_usb_hub.c | 39 ++++++++++++++++++++++++++++++--------
>  drivers/usb/misc/onboard_usb_hub.h |  1 +
>  2 files changed, 32 insertions(+), 8 deletions(-)
> 
> diff --git a/drivers/usb/misc/onboard_usb_hub.c b/drivers/usb/misc/onboard_usb_hub.c
> index 12fc6eb67c3b..a56e712d3a45 100644
> --- a/drivers/usb/misc/onboard_usb_hub.c
> +++ b/drivers/usb/misc/onboard_usb_hub.c
> @@ -27,6 +27,13 @@
>  
>  #include "onboard_usb_hub.h"
>  
> +#define MAX_SUPPLIES 2

Why 2?

> +
> +static const char * const supply_names[] = {
> +	"vdd",
> +	"vdd2",
> +};

Do those names have anything to do with the number above?  If so, please
document it!

>  struct onboard_hub_pdata {
>  	unsigned long reset_us;		/* reset pulse width in us */
> +	unsigned int num_supplies;	/* number of supplies: 0 considered as 1 */

I can not understand that comment at all :(
Benjamin Bara June 21, 2023, 4:22 p.m. UTC | #2 
On Wed, 21 Jun 2023 at 18:07, Greg Kroah-Hartman
<gregkh@linuxfoundation.org> wrote:
> On Wed, Jun 21, 2023 at 05:58:30PM +0200, Benjamin Bara wrote:
> > From: Benjamin Bara <benjamin.bara@skidata.com>
> >
> > As some of the onboard hubs require multiple power supplies, provide the
> > environment to support them.
> >
> > Signed-off-by: Benjamin Bara <benjamin.bara@skidata.com>
> > ---
> > v3:
> > - fix nits mentioned in v2
> >
> > v2:
> > - replace (err != 0) with (err)
> > ---
> >  drivers/usb/misc/onboard_usb_hub.c | 39 ++++++++++++++++++++++++++++++--------
> >  drivers/usb/misc/onboard_usb_hub.h |  1 +
> >  2 files changed, 32 insertions(+), 8 deletions(-)
> >
> > diff --git a/drivers/usb/misc/onboard_usb_hub.c b/drivers/usb/misc/onboard_usb_hub.c
> > index 12fc6eb67c3b..a56e712d3a45 100644
> > --- a/drivers/usb/misc/onboard_usb_hub.c
> > +++ b/drivers/usb/misc/onboard_usb_hub.c
> > @@ -27,6 +27,13 @@
> >
> >  #include "onboard_usb_hub.h"
> >
> > +#define MAX_SUPPLIES 2
>
> Why 2?

I picked 2 because with 3/3, this is the maximum of "required" supplies. The
currently implemented ones require only one (up to now just named "vdd"). The
new one added in 3/3 requires 2, therefore I tried to be generic if some future
hub might require 3 or more.

> > +
> > +static const char * const supply_names[] = {
> > +     "vdd",
> > +     "vdd2",
> > +};
>
> Do those names have anything to do with the number above?  If so, please
> document it!

I picked "vdd" for the first to be compatible with the existing device-trees. As
the actual names differ between hubs, I thought it might be generic to just use
"vdd2" here. If I should add some comment like "if you increase MAX_SUPPLIES,
please also add a supply_name below", I can do that. I could also implement
"vdd${i+1}" for i>0 instead.

>
> >  struct onboard_hub_pdata {
> >       unsigned long reset_us;         /* reset pulse width in us */
> > +     unsigned int num_supplies;      /* number of supplies: 0 considered as 1 */
>
> I can not understand that comment at all :(

This should just indicate that leaving the field empty means one supply is
required. Maybe "defaults to 1" is better?
Matthias Kaehlcke June 21, 2023, 4:32 p.m. UTC | #3 
On Wed, Jun 21, 2023 at 06:22:41PM +0200, Benjamin Bara wrote:
> On Wed, 21 Jun 2023 at 18:07, Greg Kroah-Hartman
> <gregkh@linuxfoundation.org> wrote:
> > On Wed, Jun 21, 2023 at 05:58:30PM +0200, Benjamin Bara wrote:
> > > From: Benjamin Bara <benjamin.bara@skidata.com>
> > >
> > > As some of the onboard hubs require multiple power supplies, provide the
> > > environment to support them.
> > >
> > > Signed-off-by: Benjamin Bara <benjamin.bara@skidata.com>
> > > ---
> > > v3:
> > > - fix nits mentioned in v2
> > >
> > > v2:
> > > - replace (err != 0) with (err)
> > > ---
> > >  drivers/usb/misc/onboard_usb_hub.c | 39 ++++++++++++++++++++++++++++++--------
> > >  drivers/usb/misc/onboard_usb_hub.h |  1 +
> > >  2 files changed, 32 insertions(+), 8 deletions(-)
> > >
> > > diff --git a/drivers/usb/misc/onboard_usb_hub.c b/drivers/usb/misc/onboard_usb_hub.c
> > > index 12fc6eb67c3b..a56e712d3a45 100644
> > > --- a/drivers/usb/misc/onboard_usb_hub.c
> > > +++ b/drivers/usb/misc/onboard_usb_hub.c
> > > @@ -27,6 +27,13 @@
> > >
> > >  #include "onboard_usb_hub.h"
> > >
> > > +#define MAX_SUPPLIES 2
> >
> > Why 2?
> 
> I picked 2 because with 3/3, this is the maximum of "required" supplies. The
> currently implemented ones require only one (up to now just named "vdd"). The
> new one added in 3/3 requires 2, therefore I tried to be generic if some future
> hub might require 3 or more.
> 
> > > +
> > > +static const char * const supply_names[] = {
> > > +     "vdd",
> > > +     "vdd2",
> > > +};
> >
> > Do those names have anything to do with the number above?  If so, please
> > document it!
> 
> I picked "vdd" for the first to be compatible with the existing device-trees. As
> the actual names differ between hubs, I thought it might be generic to just use
> "vdd2" here. If I should add some comment like "if you increase MAX_SUPPLIES,
> please also add a supply_name below", I can do that. I could also implement
> "vdd${i+1}" for i>0 instead.
> 
> >
> > >  struct onboard_hub_pdata {
> > >       unsigned long reset_us;         /* reset pulse width in us */
> > > +     unsigned int num_supplies;      /* number of supplies: 0 considered as 1 */
> >
> > I can not understand that comment at all :(
> 
> This should just indicate that leaving the field empty means one supply is
> required. Maybe "defaults to 1" is better?

Whatever the comment, I suggest to put it in parentheses, rather than after a
colon. If you keep the current style maybe s/considered/is interpreted/.

Another option would be to initialize all existing hubs to num_supplies = 1,
then there is no need for a comment and the check in _probe(). That might
be the clearest thing to do, even though it adds a few extra lines.
diff mbox series

Patch

diff --git a/drivers/usb/misc/onboard_usb_hub.c b/drivers/usb/misc/onboard_usb_hub.c
index 12fc6eb67c3b..a56e712d3a45 100644
--- a/drivers/usb/misc/onboard_usb_hub.c
+++ b/drivers/usb/misc/onboard_usb_hub.c
@@ -27,6 +27,13 @@ 
 
 #include "onboard_usb_hub.h"
 
+#define MAX_SUPPLIES 2
+
+static const char * const supply_names[] = {
+	"vdd",
+	"vdd2",
+};
+
 static void onboard_hub_attach_usb_driver(struct work_struct *work);
 
 static struct usb_device_driver onboard_hub_usbdev_driver;
@@ -40,7 +47,8 @@  struct usbdev_node {
 };
 
 struct onboard_hub {
-	struct regulator *vdd;
+	struct regulator_bulk_data supplies[MAX_SUPPLIES];
+	unsigned int num_supplies;
 	struct device *dev;
 	const struct onboard_hub_pdata *pdata;
 	struct gpio_desc *reset_gpio;
@@ -55,9 +63,9 @@  static int onboard_hub_power_on(struct onboard_hub *hub)
 {
 	int err;
 
-	err = regulator_enable(hub->vdd);
+	err = regulator_bulk_enable(hub->num_supplies, hub->supplies);
 	if (err) {
-		dev_err(hub->dev, "failed to enable regulator: %d\n", err);
+		dev_err(hub->dev, "failed to enable supplies: %d\n", err);
 		return err;
 	}
 
@@ -75,9 +83,9 @@  static int onboard_hub_power_off(struct onboard_hub *hub)
 
 	gpiod_set_value_cansleep(hub->reset_gpio, 1);
 
-	err = regulator_disable(hub->vdd);
+	err = regulator_bulk_disable(hub->num_supplies, hub->supplies);
 	if (err) {
-		dev_err(hub->dev, "failed to disable regulator: %d\n", err);
+		dev_err(hub->dev, "failed to disable supplies: %d\n", err);
 		return err;
 	}
 
@@ -232,6 +240,7 @@  static int onboard_hub_probe(struct platform_device *pdev)
 	const struct of_device_id *of_id;
 	struct device *dev = &pdev->dev;
 	struct onboard_hub *hub;
+	unsigned int i;
 	int err;
 
 	hub = devm_kzalloc(dev, sizeof(*hub), GFP_KERNEL);
@@ -246,9 +255,23 @@  static int onboard_hub_probe(struct platform_device *pdev)
 	if (!hub->pdata)
 		return -EINVAL;
 
-	hub->vdd = devm_regulator_get(dev, "vdd");
-	if (IS_ERR(hub->vdd))
-		return PTR_ERR(hub->vdd);
+	if (hub->pdata->num_supplies > MAX_SUPPLIES)
+		return dev_err_probe(dev, -EINVAL, "max %d supplies supported!\n",
+				     MAX_SUPPLIES);
+
+	if (hub->pdata->num_supplies)
+		hub->num_supplies = hub->pdata->num_supplies;
+	else
+		hub->num_supplies = 1;
+
+	for (i = 0; i < MAX_SUPPLIES; i++)
+		hub->supplies[i].supply = supply_names[i];
+
+	err = devm_regulator_bulk_get(dev, hub->num_supplies, hub->supplies);
+	if (err) {
+		dev_err(dev, "Failed to get regulator supplies: %d\n", err);
+		return err;
+	}
 
 	hub->reset_gpio = devm_gpiod_get_optional(dev, "reset",
 						  GPIOD_OUT_HIGH);
diff --git a/drivers/usb/misc/onboard_usb_hub.h b/drivers/usb/misc/onboard_usb_hub.h
index aca5f50eb0da..7a6d497ce849 100644
--- a/drivers/usb/misc/onboard_usb_hub.h
+++ b/drivers/usb/misc/onboard_usb_hub.h
@@ -8,6 +8,7 @@ 
 
 struct onboard_hub_pdata {
 	unsigned long reset_us;		/* reset pulse width in us */
+	unsigned int num_supplies;	/* number of supplies: 0 considered as 1 */
 };
 
 static const struct onboard_hub_pdata microchip_usb424_data = {