diff mbox series

usb: core: skip interfaces disabled in devicetree

Message ID 20190216170452.10884-1-mans@mansr.com (mailing list archive)
State Superseded
Headers show
Series usb: core: skip interfaces disabled in devicetree | expand

Commit Message

Måns Rullgård Feb. 16, 2019, 5:04 p.m. UTC
If an interface has an associated devicetree node with status disabled,
do not register the device.  This is useful for boards with a built-in
multifunction USB device where some functions are broken or otherwise
undesired.

Signed-off-by: Mans Rullgard <mans@mansr.com>
---
 drivers/usb/core/message.c | 4 ++++
 1 file changed, 4 insertions(+)

Comments

Greg KH Feb. 19, 2019, 1:11 p.m. UTC | #1
On Sat, Feb 16, 2019 at 05:04:52PM +0000, Mans Rullgard wrote:
> If an interface has an associated devicetree node with status disabled,
> do not register the device.  This is useful for boards with a built-in
> multifunction USB device where some functions are broken or otherwise
> undesired.
> 
> Signed-off-by: Mans Rullgard <mans@mansr.com>
> ---
>  drivers/usb/core/message.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
> index bfa5eda0cc26..6b45d4835e41 100644
> --- a/drivers/usb/core/message.c
> +++ b/drivers/usb/core/message.c
> @@ -2007,6 +2007,10 @@ int usb_set_configuration(struct usb_device *dev, int configuration)
>  	for (i = 0; i < nintf; ++i) {
>  		struct usb_interface *intf = cp->interface[i];
>  
> +		if (intf->dev.of_node &&
> +		    !of_device_is_available(intf->dev.of_node))
> +			continue;

Shouldn't you at least print some message out saying you are skipping
this?  Odds are this is going to cause regressions in devices that were
not expecting this, right?  So pointing them at why their devices now no
longer work would be good :)

thanks,

greg k-h
Måns Rullgård Feb. 19, 2019, 1:24 p.m. UTC | #2
Greg Kroah-Hartman <gregkh@linuxfoundation.org> writes:

> On Sat, Feb 16, 2019 at 05:04:52PM +0000, Mans Rullgard wrote:
>> If an interface has an associated devicetree node with status disabled,
>> do not register the device.  This is useful for boards with a built-in
>> multifunction USB device where some functions are broken or otherwise
>> undesired.
>> 
>> Signed-off-by: Mans Rullgard <mans@mansr.com>
>> ---
>>  drivers/usb/core/message.c | 4 ++++
>>  1 file changed, 4 insertions(+)
>> 
>> diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
>> index bfa5eda0cc26..6b45d4835e41 100644
>> --- a/drivers/usb/core/message.c
>> +++ b/drivers/usb/core/message.c
>> @@ -2007,6 +2007,10 @@ int usb_set_configuration(struct usb_device *dev, int configuration)
>>  	for (i = 0; i < nintf; ++i) {
>>  		struct usb_interface *intf = cp->interface[i];
>>  
>> +		if (intf->dev.of_node &&
>> +		    !of_device_is_available(intf->dev.of_node))
>> +			continue;
>
> Shouldn't you at least print some message out saying you are skipping
> this?  Odds are this is going to cause regressions in devices that were
> not expecting this, right?  So pointing them at why their devices now no
> longer work would be good :)

They will only be skipped if there is a device tree node for the
interface _and_ it has and explicit status = "disabled" property.
The default is still to create devices for all interfaces.

That said, printing a message is probably a good idea anyway.  Would
"info" level be appropriate for this?
Greg KH Feb. 19, 2019, 1:42 p.m. UTC | #3
On Tue, Feb 19, 2019 at 01:24:51PM +0000, Måns Rullgård wrote:
> Greg Kroah-Hartman <gregkh@linuxfoundation.org> writes:
> 
> > On Sat, Feb 16, 2019 at 05:04:52PM +0000, Mans Rullgard wrote:
> >> If an interface has an associated devicetree node with status disabled,
> >> do not register the device.  This is useful for boards with a built-in
> >> multifunction USB device where some functions are broken or otherwise
> >> undesired.
> >> 
> >> Signed-off-by: Mans Rullgard <mans@mansr.com>
> >> ---
> >>  drivers/usb/core/message.c | 4 ++++
> >>  1 file changed, 4 insertions(+)
> >> 
> >> diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
> >> index bfa5eda0cc26..6b45d4835e41 100644
> >> --- a/drivers/usb/core/message.c
> >> +++ b/drivers/usb/core/message.c
> >> @@ -2007,6 +2007,10 @@ int usb_set_configuration(struct usb_device *dev, int configuration)
> >>  	for (i = 0; i < nintf; ++i) {
> >>  		struct usb_interface *intf = cp->interface[i];
> >>  
> >> +		if (intf->dev.of_node &&
> >> +		    !of_device_is_available(intf->dev.of_node))
> >> +			continue;
> >
> > Shouldn't you at least print some message out saying you are skipping
> > this?  Odds are this is going to cause regressions in devices that were
> > not expecting this, right?  So pointing them at why their devices now no
> > longer work would be good :)
> 
> They will only be skipped if there is a device tree node for the
> interface _and_ it has and explicit status = "disabled" property.
> The default is still to create devices for all interfaces.

Yes, but today if you have such a status field set, it will not skip
them, so who knows who has their DT messed up :)

> That said, printing a message is probably a good idea anyway.  Would
> "info" level be appropriate for this?

That is a good idea.

thanks,

greg k-h
diff mbox series

Patch

diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
index bfa5eda0cc26..6b45d4835e41 100644
--- a/drivers/usb/core/message.c
+++ b/drivers/usb/core/message.c
@@ -2007,6 +2007,10 @@  int usb_set_configuration(struct usb_device *dev, int configuration)
 	for (i = 0; i < nintf; ++i) {
 		struct usb_interface *intf = cp->interface[i];
 
+		if (intf->dev.of_node &&
+		    !of_device_is_available(intf->dev.of_node))
+			continue;
+
 		dev_dbg(&dev->dev,
 			"adding %s (config #%d, interface %d)\n",
 			dev_name(&intf->dev), configuration,