[PATCH] at91: at91-ohci: fix set/get power

[PATCH] at91: at91-ohci: fix set/get power

[Jean-Christophe PLAGNIOL-VILLARD]

in commit aa6e52a35 we introduce the support of overcurrent notification
but the set and get of the power without checking if the gpio is valid or not

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Nicolas Ferre <nicolas.ferre@atmel.com>
---
 drivers/usb/host/ohci-at91.c |    6 ++++++
 1 files changed, 6 insertions(+), 0 deletions(-)

diff --git a/drivers/usb/host/ohci-at91.c b/drivers/usb/host/ohci-at91.c
index ba3a46b..95a9fec 100644
--- a/drivers/usb/host/ohci-at91.c
+++ b/drivers/usb/host/ohci-at91.c
@@ -223,6 +223,9 @@ static void ohci_at91_usb_set_power(struct at91_usbh_data *pdata, int port, int
 	if (port < 0 || port >= 2)
 		return;
 
+	if (pdata->vbus_pin[port] <= 0)
+		return;
+
 	gpio_set_value(pdata->vbus_pin[port], !pdata->vbus_pin_inverted ^ enable);
 }
 
@@ -231,6 +234,9 @@ static int ohci_at91_usb_get_power(struct at91_usbh_data *pdata, int port)
 	if (port < 0 || port >= 2)
 		return -EINVAL;
 
+	if (pdata->vbus_pin[port] <= 0)
+		return -EINVAL;
+
 	return gpio_get_value(pdata->vbus_pin[port]) ^ !pdata->vbus_pin_inverted;
 }
 
-- 
1.7.7

[PATCH] at91: at91-ohci: fix set/get power

[Nicolas Ferre]

On 11/12/2011 04:46 PM, Jean-Christophe PLAGNIOL-VILLARD :
> in commit aa6e52a35 we introduce the support of overcurrent notification
> but the set and get of the power without checking if the gpio is valid or not
> 
> Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
> Cc: Nicolas Ferre <nicolas.ferre@atmel.com>

Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>

> ---
>  drivers/usb/host/ohci-at91.c |    6 ++++++
>  1 files changed, 6 insertions(+), 0 deletions(-)
> 
> diff --git a/drivers/usb/host/ohci-at91.c b/drivers/usb/host/ohci-at91.c
> index ba3a46b..95a9fec 100644
> --- a/drivers/usb/host/ohci-at91.c
> +++ b/drivers/usb/host/ohci-at91.c
> @@ -223,6 +223,9 @@ static void ohci_at91_usb_set_power(struct at91_usbh_data *pdata, int port, int
>  	if (port < 0 || port >= 2)
>  		return;
>  
> +	if (pdata->vbus_pin[port] <= 0)
> +		return;
> +
>  	gpio_set_value(pdata->vbus_pin[port], !pdata->vbus_pin_inverted ^ enable);
>  }
>  
> @@ -231,6 +234,9 @@ static int ohci_at91_usb_get_power(struct at91_usbh_data *pdata, int port)
>  	if (port < 0 || port >= 2)
>  		return -EINVAL;
>  
> +	if (pdata->vbus_pin[port] <= 0)
> +		return -EINVAL;
> +
>  	return gpio_get_value(pdata->vbus_pin[port]) ^ !pdata->vbus_pin_inverted;
>  }
>  


-- 
Nicolas Ferre

[PATCH] at91: at91-ohci: fix set/get power

[Thomas Petazzoni]

Le Sat, 12 Nov 2011 16:46:13 +0100,
Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com> a ?crit :

> in commit aa6e52a35 we introduce the support of overcurrent notification
> but the set and get of the power without checking if the gpio is valid or not
> 
> Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
> Cc: Nicolas Ferre <nicolas.ferre@atmel.com>

Acked-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

Thomas
-- 
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com

[PATCH] at91: at91-ohci: fix set/get power

[Andrew Victor]

hi,

> in commit aa6e52a35 we introduce the support of overcurrent notification
> but the set and get of the power without checking if the gpio is valid or not

> + ? ? ? if (pdata->vbus_pin[port] <= 0)
> + ? ? ? ? ? ? ? return;

Shouldn't that rather be:
   if (!gpio_is_valid(pdata->vbus_pin[port]))
          return;


Regards,
  Andrew Victor

[PATCH] at91: at91-ohci: fix set/get power

[Thomas Petazzoni]

Le Mon, 14 Nov 2011 22:39:21 +0200,
Andrew Victor <avictor.za@gmail.com> a ?crit :

> Shouldn't that rather be:
>    if (!gpio_is_valid(pdata->vbus_pin[port]))
>           return;

No, because gpio_is_valid() accept 0 as a valid GPIO. Therefore, when
you don't set anything for the .vbus_pin field, the values are zero,
and gpio_is_valid() will tell you that it's a valid GPIO. If we wanted
to use gpio_is_valid(), we would have to do something like .vbus_pin =
{ -EINVAL, -EINVAL } in every board for which the vbus pin isn't used.
It's a quite unfortunate behaviour of the current gpio subsystem.

Regards,

Thomas
-- 
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com

[PATCH] at91: at91-ohci: fix set/get power

[Jean-Christophe PLAGNIOL-VILLARD]

On 22:39 Mon 14 Nov     , Andrew Victor wrote:
> hi,
> 
> > in commit aa6e52a35 we introduce the support of overcurrent notification
> > but the set and get of the power without checking if the gpio is valid or not
> 
> > + ? ? ? if (pdata->vbus_pin[port] <= 0)
> > + ? ? ? ? ? ? ? return;
> 
> Shouldn't that rather be:
>    if (!gpio_is_valid(pdata->vbus_pin[port]))
>           return;
Yes should be but 0 is not a valid gpio on at91

I've a patch serie to fix the gpio support for at91 so we can use
gpio_is_valid for 3.3

Best Regards,
J.