Re: [PATCH v2 1/2] usb: gadget: udc-core: independent registration of gadgets and gadget drivers

2015-02-17 Thread Ruslan Bilovol 
On Mon, Feb 16, 2015 at 1:23 AM, Alan Stern  wrote:
> On Mon, 16 Feb 2015, Ruslan Bilovol wrote:
>
>> Change behavior during registration of gadgets and
>> gadget drivers in udc-core. Instead of previous
>> approach when for successful probe of usb gadget driver
>> at least one usb gadget should be already registered
>> use another one where gadget drivers and gadgets
>> can be registered in udc-core independently.
>>
>> Independent registration of gadgets and gadget drivers
>> is useful for built-in into kernel gadget and gadget
>> driver case - because it's possible that gadget is
>> really probed only on late_init stage (due to deferred
>> probe) whereas gadget driver's probe is silently failed
>> on module_init stage due to no any UDC added.
>>
>> Also it is useful for modules case - now there is no
>> difference what module to insert first: gadget module
>> or gadget driver one.
>>
>> Signed-off-by: Ruslan Bilovol 
>> ---
>>  drivers/usb/gadget/udc/udc-core.c | 77 
>> +--
>>  1 file changed, 74 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/usb/gadget/udc/udc-core.c 
>> b/drivers/usb/gadget/udc/udc-core.c
>> index 5a81cb0..71b6942 100644
>> --- a/drivers/usb/gadget/udc/udc-core.c
>> +++ b/drivers/usb/gadget/udc/udc-core.c
>> @@ -46,10 +46,18 @@ struct usb_udc {
>>   struct list_headlist;
>>  };
>>
>> +struct pending_gadget_driver {
>> + struct usb_gadget_driver*driver;
>> + char*udc_name;
>> + struct list_headlist;
>> +};
>
> Don't make this a separate structure.  It should be embedded in the
> usb_gadget_driver structure.  This will make a lot of the changes below
> much simpler.

Yes, but require more changes across gadget sources. Will do it in next version

>
>> @@ -288,6 +297,24 @@ int usb_add_gadget_udc_release(struct device *parent, 
>> struct usb_gadget *gadget,
>>
>>   usb_gadget_set_state(gadget, USB_STATE_NOTATTACHED);
>>
>> + if (!list_empty(_driver_pending_list)) {
>> + pending = list_first_entry(_driver_pending_list,
>> + struct pending_gadget_driver, list);
>> +
>> + if (pending->udc_name) {
>> + if (!strcmp(pending->udc_name, dev_name(>dev))) {
>> + udc_bind_to_driver(udc, pending->driver);
>> + list_del(>list);
>> + kfree(pending->udc_name);
>> + kfree(pending);
>> + }
>> + } else {
>> + udc_bind_to_driver(udc, pending->driver);
>> + list_del(>list);
>> + kfree(pending);
>> + }
>
> This code can be simplified too:
>
> if (!pending->udc_name || strcmp(pending->udc_name,
> dev_name(>dev) == 0) {
> ...

Good catch!

>> + }
>> +
>>   mutex_unlock(_lock);
>>
>>   return 0;
>
> Don't you want to add a driver back to the pending list
> usb_del_gadget_udc() after the call to usb_gadget_remove_driver?

Will be added in next version, should be few lines change

>
>> @@ -423,7 +450,27 @@ int usb_udc_attach_driver(const char *name, struct 
>> usb_gadget_driver *driver)
>>   break;
>>   }
>>   if (ret) {
>> - ret = -ENODEV;
>> + struct pending_gadget_driver *pending;
>> +
>> + pending = kzalloc(sizeof(*pending), GFP_KERNEL);
>> + if (!pending) {
>> + ret = -ENOMEM;
>> + goto out;
>> + }
>> + pending->driver = driver;
>> + pending->udc_name = kstrdup(name, GFP_KERNEL);
>> + if (!pending->udc_name) {
>> + kfree(pending);
>> + ret = -ENOMEM;
>> + goto out;
>> + }
>
> Why do you need to copy the name?  Just use the original.

Because it's only place we keep it. Will not be needed in next version of patch

Best regards,
Ruslan

>
> Alan Stern
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH v2 1/2] usb: gadget: udc-core: independent registration of gadgets and gadget drivers

2015-02-17 Thread Ruslan Bilovol 
On Mon, Feb 16, 2015 at 1:23 AM, Alan Stern st...@rowland.harvard.edu wrote:
 On Mon, 16 Feb 2015, Ruslan Bilovol wrote:

 Change behavior during registration of gadgets and
 gadget drivers in udc-core. Instead of previous
 approach when for successful probe of usb gadget driver
 at least one usb gadget should be already registered
 use another one where gadget drivers and gadgets
 can be registered in udc-core independently.

 Independent registration of gadgets and gadget drivers
 is useful for built-in into kernel gadget and gadget
 driver case - because it's possible that gadget is
 really probed only on late_init stage (due to deferred
 probe) whereas gadget driver's probe is silently failed
 on module_init stage due to no any UDC added.

 Also it is useful for modules case - now there is no
 difference what module to insert first: gadget module
 or gadget driver one.

 Signed-off-by: Ruslan Bilovol ruslan.bilo...@gmail.com
 ---
  drivers/usb/gadget/udc/udc-core.c | 77 
 +--
  1 file changed, 74 insertions(+), 3 deletions(-)

 diff --git a/drivers/usb/gadget/udc/udc-core.c 
 b/drivers/usb/gadget/udc/udc-core.c
 index 5a81cb0..71b6942 100644
 --- a/drivers/usb/gadget/udc/udc-core.c
 +++ b/drivers/usb/gadget/udc/udc-core.c
 @@ -46,10 +46,18 @@ struct usb_udc {
   struct list_headlist;
  };

 +struct pending_gadget_driver {
 + struct usb_gadget_driver*driver;
 + char*udc_name;
 + struct list_headlist;
 +};

 Don't make this a separate structure.  It should be embedded in the
 usb_gadget_driver structure.  This will make a lot of the changes below
 much simpler.

Yes, but require more changes across gadget sources. Will do it in next version


 @@ -288,6 +297,24 @@ int usb_add_gadget_udc_release(struct device *parent, 
 struct usb_gadget *gadget,

   usb_gadget_set_state(gadget, USB_STATE_NOTATTACHED);

 + if (!list_empty(gadget_driver_pending_list)) {
 + pending = list_first_entry(gadget_driver_pending_list,
 + struct pending_gadget_driver, list);
 +
 + if (pending-udc_name) {
 + if (!strcmp(pending-udc_name, dev_name(udc-dev))) {
 + udc_bind_to_driver(udc, pending-driver);
 + list_del(pending-list);
 + kfree(pending-udc_name);
 + kfree(pending);
 + }
 + } else {
 + udc_bind_to_driver(udc, pending-driver);
 + list_del(pending-list);
 + kfree(pending);
 + }

 This code can be simplified too:

 if (!pending-udc_name || strcmp(pending-udc_name,
 dev_name(udc-dev) == 0) {
 ...

Good catch!

 + }
 +
   mutex_unlock(udc_lock);

   return 0;

 Don't you want to add a driver back to the pending list
 usb_del_gadget_udc() after the call to usb_gadget_remove_driver?

Will be added in next version, should be few lines change


 @@ -423,7 +450,27 @@ int usb_udc_attach_driver(const char *name, struct 
 usb_gadget_driver *driver)
   break;
   }
   if (ret) {
 - ret = -ENODEV;
 + struct pending_gadget_driver *pending;
 +
 + pending = kzalloc(sizeof(*pending), GFP_KERNEL);
 + if (!pending) {
 + ret = -ENOMEM;
 + goto out;
 + }
 + pending-driver = driver;
 + pending-udc_name = kstrdup(name, GFP_KERNEL);
 + if (!pending-udc_name) {
 + kfree(pending);
 + ret = -ENOMEM;
 + goto out;
 + }

 Why do you need to copy the name?  Just use the original.

Because it's only place we keep it. Will not be needed in next version of patch

Best regards,
Ruslan


 Alan Stern

--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH v2 1/2] usb: gadget: udc-core: independent registration of gadgets and gadget drivers

2015-02-16 Thread Felipe Balbi 
On Sun, Feb 15, 2015 at 06:23:02PM -0500, Alan Stern wrote:
> On Mon, 16 Feb 2015, Ruslan Bilovol wrote:
> 
> > Change behavior during registration of gadgets and
> > gadget drivers in udc-core. Instead of previous
> > approach when for successful probe of usb gadget driver
> > at least one usb gadget should be already registered
> > use another one where gadget drivers and gadgets
> > can be registered in udc-core independently.
> > 
> > Independent registration of gadgets and gadget drivers
> > is useful for built-in into kernel gadget and gadget
> > driver case - because it's possible that gadget is
> > really probed only on late_init stage (due to deferred
> > probe) whereas gadget driver's probe is silently failed
> > on module_init stage due to no any UDC added.
> > 
> > Also it is useful for modules case - now there is no
> > difference what module to insert first: gadget module
> > or gadget driver one.
> > 
> > Signed-off-by: Ruslan Bilovol 
> > ---
> >  drivers/usb/gadget/udc/udc-core.c | 77 
> > +--
> >  1 file changed, 74 insertions(+), 3 deletions(-)
> > 
> > diff --git a/drivers/usb/gadget/udc/udc-core.c 
> > b/drivers/usb/gadget/udc/udc-core.c
> > index 5a81cb0..71b6942 100644
> > --- a/drivers/usb/gadget/udc/udc-core.c
> > +++ b/drivers/usb/gadget/udc/udc-core.c
> > @@ -46,10 +46,18 @@ struct usb_udc {
> > struct list_headlist;
> >  };
> >  
> > +struct pending_gadget_driver {
> > +   struct usb_gadget_driver*driver;
> > +   char*udc_name;
> > +   struct list_headlist;
> > +};
> 
> Don't make this a separate structure.  It should be embedded in the
> usb_gadget_driver structure.  This will make a lot of the changes
> below much simpler.

right, also when resending, make sure to Cc linux-usb as well.

-- 
balbi


signature.asc
Description: Digital signature

Re: [PATCH v2 1/2] usb: gadget: udc-core: independent registration of gadgets and gadget drivers

2015-02-16 Thread Felipe Balbi 
On Sun, Feb 15, 2015 at 06:23:02PM -0500, Alan Stern wrote:
 On Mon, 16 Feb 2015, Ruslan Bilovol wrote:
 
  Change behavior during registration of gadgets and
  gadget drivers in udc-core. Instead of previous
  approach when for successful probe of usb gadget driver
  at least one usb gadget should be already registered
  use another one where gadget drivers and gadgets
  can be registered in udc-core independently.
  
  Independent registration of gadgets and gadget drivers
  is useful for built-in into kernel gadget and gadget
  driver case - because it's possible that gadget is
  really probed only on late_init stage (due to deferred
  probe) whereas gadget driver's probe is silently failed
  on module_init stage due to no any UDC added.
  
  Also it is useful for modules case - now there is no
  difference what module to insert first: gadget module
  or gadget driver one.
  
  Signed-off-by: Ruslan Bilovol ruslan.bilo...@gmail.com
  ---
   drivers/usb/gadget/udc/udc-core.c | 77 
  +--
   1 file changed, 74 insertions(+), 3 deletions(-)
  
  diff --git a/drivers/usb/gadget/udc/udc-core.c 
  b/drivers/usb/gadget/udc/udc-core.c
  index 5a81cb0..71b6942 100644
  --- a/drivers/usb/gadget/udc/udc-core.c
  +++ b/drivers/usb/gadget/udc/udc-core.c
  @@ -46,10 +46,18 @@ struct usb_udc {
  struct list_headlist;
   };
   
  +struct pending_gadget_driver {
  +   struct usb_gadget_driver*driver;
  +   char*udc_name;
  +   struct list_headlist;
  +};
 
 Don't make this a separate structure.  It should be embedded in the
 usb_gadget_driver structure.  This will make a lot of the changes
 below much simpler.

right, also when resending, make sure to Cc linux-usb as well.

-- 
balbi


signature.asc
Description: Digital signature

Re: [PATCH v2 1/2] usb: gadget: udc-core: independent registration of gadgets and gadget drivers

2015-02-15 Thread Alan Stern 
On Mon, 16 Feb 2015, Ruslan Bilovol wrote:

> Change behavior during registration of gadgets and
> gadget drivers in udc-core. Instead of previous
> approach when for successful probe of usb gadget driver
> at least one usb gadget should be already registered
> use another one where gadget drivers and gadgets
> can be registered in udc-core independently.
> 
> Independent registration of gadgets and gadget drivers
> is useful for built-in into kernel gadget and gadget
> driver case - because it's possible that gadget is
> really probed only on late_init stage (due to deferred
> probe) whereas gadget driver's probe is silently failed
> on module_init stage due to no any UDC added.
> 
> Also it is useful for modules case - now there is no
> difference what module to insert first: gadget module
> or gadget driver one.
> 
> Signed-off-by: Ruslan Bilovol 
> ---
>  drivers/usb/gadget/udc/udc-core.c | 77 
> +--
>  1 file changed, 74 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/usb/gadget/udc/udc-core.c 
> b/drivers/usb/gadget/udc/udc-core.c
> index 5a81cb0..71b6942 100644
> --- a/drivers/usb/gadget/udc/udc-core.c
> +++ b/drivers/usb/gadget/udc/udc-core.c
> @@ -46,10 +46,18 @@ struct usb_udc {
>   struct list_headlist;
>  };
>  
> +struct pending_gadget_driver {
> + struct usb_gadget_driver*driver;
> + char*udc_name;
> + struct list_headlist;
> +};

Don't make this a separate structure.  It should be embedded in the 
usb_gadget_driver structure.  This will make a lot of the changes below 
much simpler.

> @@ -288,6 +297,24 @@ int usb_add_gadget_udc_release(struct device *parent, 
> struct usb_gadget *gadget,
>  
>   usb_gadget_set_state(gadget, USB_STATE_NOTATTACHED);
>  
> + if (!list_empty(_driver_pending_list)) {
> + pending = list_first_entry(_driver_pending_list,
> + struct pending_gadget_driver, list);
> +
> + if (pending->udc_name) {
> + if (!strcmp(pending->udc_name, dev_name(>dev))) {
> + udc_bind_to_driver(udc, pending->driver);
> + list_del(>list);
> + kfree(pending->udc_name);
> + kfree(pending);
> + }
> + } else {
> + udc_bind_to_driver(udc, pending->driver);
> + list_del(>list);
> + kfree(pending);
> + }

This code can be simplified too:

if (!pending->udc_name || strcmp(pending->udc_name, 
dev_name(>dev) == 0) {
...
> + }
> +
>   mutex_unlock(_lock);
>  
>   return 0;

Don't you want to add a driver back to the pending list 
usb_del_gadget_udc() after the call to usb_gadget_remove_driver?

> @@ -423,7 +450,27 @@ int usb_udc_attach_driver(const char *name, struct 
> usb_gadget_driver *driver)
>   break;
>   }
>   if (ret) {
> - ret = -ENODEV;
> + struct pending_gadget_driver *pending;
> +
> + pending = kzalloc(sizeof(*pending), GFP_KERNEL);
> + if (!pending) {
> + ret = -ENOMEM;
> + goto out;
> + }
> + pending->driver = driver;
> + pending->udc_name = kstrdup(name, GFP_KERNEL);
> + if (!pending->udc_name) {
> + kfree(pending);
> + ret = -ENOMEM;
> + goto out;
> + }

Why do you need to copy the name?  Just use the original.

Alan Stern

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[PATCH v2 1/2] usb: gadget: udc-core: independent registration of gadgets and gadget drivers

2015-02-15 Thread Ruslan Bilovol 
Change behavior during registration of gadgets and
gadget drivers in udc-core. Instead of previous
approach when for successful probe of usb gadget driver
at least one usb gadget should be already registered
use another one where gadget drivers and gadgets
can be registered in udc-core independently.

Independent registration of gadgets and gadget drivers
is useful for built-in into kernel gadget and gadget
driver case - because it's possible that gadget is
really probed only on late_init stage (due to deferred
probe) whereas gadget driver's probe is silently failed
on module_init stage due to no any UDC added.

Also it is useful for modules case - now there is no
difference what module to insert first: gadget module
or gadget driver one.

Signed-off-by: Ruslan Bilovol 
---
 drivers/usb/gadget/udc/udc-core.c | 77 +--
 1 file changed, 74 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/gadget/udc/udc-core.c 
b/drivers/usb/gadget/udc/udc-core.c
index 5a81cb0..71b6942 100644
--- a/drivers/usb/gadget/udc/udc-core.c
+++ b/drivers/usb/gadget/udc/udc-core.c
@@ -46,10 +46,18 @@ struct usb_udc {
struct list_headlist;
 };
 
+struct pending_gadget_driver {
+   struct usb_gadget_driver*driver;
+   char*udc_name;
+   struct list_headlist;
+};
+
 static struct class *udc_class;
 static LIST_HEAD(udc_list);
+static LIST_HEAD(gadget_driver_pending_list);
 static DEFINE_MUTEX(udc_lock);
 
+static int udc_bind_to_driver(struct usb_udc *udc, struct usb_gadget_driver 
*driver);
 /* - */
 
 #ifdef CONFIG_HAS_DMA
@@ -244,6 +252,7 @@ int usb_add_gadget_udc_release(struct device *parent, 
struct usb_gadget *gadget,
 {
struct usb_udc  *udc;
int ret = -ENOMEM;
+   struct pending_gadget_driver *pending;
 
udc = kzalloc(sizeof(*udc), GFP_KERNEL);
if (!udc)
@@ -288,6 +297,24 @@ int usb_add_gadget_udc_release(struct device *parent, 
struct usb_gadget *gadget,
 
usb_gadget_set_state(gadget, USB_STATE_NOTATTACHED);
 
+   if (!list_empty(_driver_pending_list)) {
+   pending = list_first_entry(_driver_pending_list,
+   struct pending_gadget_driver, list);
+
+   if (pending->udc_name) {
+   if (!strcmp(pending->udc_name, dev_name(>dev))) {
+   udc_bind_to_driver(udc, pending->driver);
+   list_del(>list);
+   kfree(pending->udc_name);
+   kfree(pending);
+   }
+   } else {
+   udc_bind_to_driver(udc, pending->driver);
+   list_del(>list);
+   kfree(pending);
+   }
+   }
+
mutex_unlock(_lock);
 
return 0;
@@ -423,7 +450,27 @@ int usb_udc_attach_driver(const char *name, struct 
usb_gadget_driver *driver)
break;
}
if (ret) {
-   ret = -ENODEV;
+   struct pending_gadget_driver *pending;
+
+   pending = kzalloc(sizeof(*pending), GFP_KERNEL);
+   if (!pending) {
+   ret = -ENOMEM;
+   goto out;
+   }
+   pending->driver = driver;
+   pending->udc_name = kstrdup(name, GFP_KERNEL);
+   if (!pending->udc_name) {
+   kfree(pending);
+   ret = -ENOMEM;
+   goto out;
+   }
+
+   list_add_tail(>list, _driver_pending_list);
+
+   pr_info("udc-core: couldn't find the [%s] UDC "
+   "- added [%s] to list of pending drivers\n",
+   name, driver->function);
+   ret = 0;
goto out;
}
if (udc->driver) {
@@ -440,6 +487,7 @@ EXPORT_SYMBOL_GPL(usb_udc_attach_driver);
 int usb_gadget_probe_driver(struct usb_gadget_driver *driver)
 {
struct usb_udc  *udc = NULL;
+   struct pending_gadget_driver *pending;
int ret;
 
if (!driver || !driver->bind || !driver->setup)
@@ -452,9 +500,20 @@ int usb_gadget_probe_driver(struct usb_gadget_driver 
*driver)
goto found;
}
 
-   pr_debug("couldn't find an available UDC\n");
+   pending = kzalloc(sizeof(*pending), GFP_KERNEL);
+   if (!pending) {
+   mutex_unlock(_lock);
+   return -ENOMEM;
+   }
+   pending->driver = driver;
+   list_add_tail(>list, _driver_pending_list);
+
+   pr_info("udc-core: couldn't find an available UDC "
+   "- add [%s] to list of pending drivers\n",
+   driver->function);
+

[PATCH v2 1/2] usb: gadget: udc-core: independent registration of gadgets and gadget drivers

2015-02-15 Thread Ruslan Bilovol 
Change behavior during registration of gadgets and
gadget drivers in udc-core. Instead of previous
approach when for successful probe of usb gadget driver
at least one usb gadget should be already registered
use another one where gadget drivers and gadgets
can be registered in udc-core independently.

Independent registration of gadgets and gadget drivers
is useful for built-in into kernel gadget and gadget
driver case - because it's possible that gadget is
really probed only on late_init stage (due to deferred
probe) whereas gadget driver's probe is silently failed
on module_init stage due to no any UDC added.

Also it is useful for modules case - now there is no
difference what module to insert first: gadget module
or gadget driver one.

Signed-off-by: Ruslan Bilovol ruslan.bilo...@gmail.com
---
 drivers/usb/gadget/udc/udc-core.c | 77 +--
 1 file changed, 74 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/gadget/udc/udc-core.c 
b/drivers/usb/gadget/udc/udc-core.c
index 5a81cb0..71b6942 100644
--- a/drivers/usb/gadget/udc/udc-core.c
+++ b/drivers/usb/gadget/udc/udc-core.c
@@ -46,10 +46,18 @@ struct usb_udc {
struct list_headlist;
 };
 
+struct pending_gadget_driver {
+   struct usb_gadget_driver*driver;
+   char*udc_name;
+   struct list_headlist;
+};
+
 static struct class *udc_class;
 static LIST_HEAD(udc_list);
+static LIST_HEAD(gadget_driver_pending_list);
 static DEFINE_MUTEX(udc_lock);
 
+static int udc_bind_to_driver(struct usb_udc *udc, struct usb_gadget_driver 
*driver);
 /* - */
 
 #ifdef CONFIG_HAS_DMA
@@ -244,6 +252,7 @@ int usb_add_gadget_udc_release(struct device *parent, 
struct usb_gadget *gadget,
 {
struct usb_udc  *udc;
int ret = -ENOMEM;
+   struct pending_gadget_driver *pending;
 
udc = kzalloc(sizeof(*udc), GFP_KERNEL);
if (!udc)
@@ -288,6 +297,24 @@ int usb_add_gadget_udc_release(struct device *parent, 
struct usb_gadget *gadget,
 
usb_gadget_set_state(gadget, USB_STATE_NOTATTACHED);
 
+   if (!list_empty(gadget_driver_pending_list)) {
+   pending = list_first_entry(gadget_driver_pending_list,
+   struct pending_gadget_driver, list);
+
+   if (pending-udc_name) {
+   if (!strcmp(pending-udc_name, dev_name(udc-dev))) {
+   udc_bind_to_driver(udc, pending-driver);
+   list_del(pending-list);
+   kfree(pending-udc_name);
+   kfree(pending);
+   }
+   } else {
+   udc_bind_to_driver(udc, pending-driver);
+   list_del(pending-list);
+   kfree(pending);
+   }
+   }
+
mutex_unlock(udc_lock);
 
return 0;
@@ -423,7 +450,27 @@ int usb_udc_attach_driver(const char *name, struct 
usb_gadget_driver *driver)
break;
}
if (ret) {
-   ret = -ENODEV;
+   struct pending_gadget_driver *pending;
+
+   pending = kzalloc(sizeof(*pending), GFP_KERNEL);
+   if (!pending) {
+   ret = -ENOMEM;
+   goto out;
+   }
+   pending-driver = driver;
+   pending-udc_name = kstrdup(name, GFP_KERNEL);
+   if (!pending-udc_name) {
+   kfree(pending);
+   ret = -ENOMEM;
+   goto out;
+   }
+
+   list_add_tail(pending-list, gadget_driver_pending_list);
+
+   pr_info(udc-core: couldn't find the [%s] UDC 
+   - added [%s] to list of pending drivers\n,
+   name, driver-function);
+   ret = 0;
goto out;
}
if (udc-driver) {
@@ -440,6 +487,7 @@ EXPORT_SYMBOL_GPL(usb_udc_attach_driver);
 int usb_gadget_probe_driver(struct usb_gadget_driver *driver)
 {
struct usb_udc  *udc = NULL;
+   struct pending_gadget_driver *pending;
int ret;
 
if (!driver || !driver-bind || !driver-setup)
@@ -452,9 +500,20 @@ int usb_gadget_probe_driver(struct usb_gadget_driver 
*driver)
goto found;
}
 
-   pr_debug(couldn't find an available UDC\n);
+   pending = kzalloc(sizeof(*pending), GFP_KERNEL);
+   if (!pending) {
+   mutex_unlock(udc_lock);
+   return -ENOMEM;
+   }
+   pending-driver = driver;
+   list_add_tail(pending-list, gadget_driver_pending_list);
+
+   pr_info(udc-core: couldn't find an available UDC 
+   - add [%s] to list of pending

Re: [PATCH v2 1/2] usb: gadget: udc-core: independent registration of gadgets and gadget drivers

2015-02-15 Thread Alan Stern 
On Mon, 16 Feb 2015, Ruslan Bilovol wrote:

 Change behavior during registration of gadgets and
 gadget drivers in udc-core. Instead of previous
 approach when for successful probe of usb gadget driver
 at least one usb gadget should be already registered
 use another one where gadget drivers and gadgets
 can be registered in udc-core independently.
 
 Independent registration of gadgets and gadget drivers
 is useful for built-in into kernel gadget and gadget
 driver case - because it's possible that gadget is
 really probed only on late_init stage (due to deferred
 probe) whereas gadget driver's probe is silently failed
 on module_init stage due to no any UDC added.
 
 Also it is useful for modules case - now there is no
 difference what module to insert first: gadget module
 or gadget driver one.
 
 Signed-off-by: Ruslan Bilovol ruslan.bilo...@gmail.com
 ---
  drivers/usb/gadget/udc/udc-core.c | 77 
 +--
  1 file changed, 74 insertions(+), 3 deletions(-)
 
 diff --git a/drivers/usb/gadget/udc/udc-core.c 
 b/drivers/usb/gadget/udc/udc-core.c
 index 5a81cb0..71b6942 100644
 --- a/drivers/usb/gadget/udc/udc-core.c
 +++ b/drivers/usb/gadget/udc/udc-core.c
 @@ -46,10 +46,18 @@ struct usb_udc {
   struct list_headlist;
  };
  
 +struct pending_gadget_driver {
 + struct usb_gadget_driver*driver;
 + char*udc_name;
 + struct list_headlist;
 +};

Don't make this a separate structure.  It should be embedded in the 
usb_gadget_driver structure.  This will make a lot of the changes below 
much simpler.

 @@ -288,6 +297,24 @@ int usb_add_gadget_udc_release(struct device *parent, 
 struct usb_gadget *gadget,
  
   usb_gadget_set_state(gadget, USB_STATE_NOTATTACHED);
  
 + if (!list_empty(gadget_driver_pending_list)) {
 + pending = list_first_entry(gadget_driver_pending_list,
 + struct pending_gadget_driver, list);
 +
 + if (pending-udc_name) {
 + if (!strcmp(pending-udc_name, dev_name(udc-dev))) {
 + udc_bind_to_driver(udc, pending-driver);
 + list_del(pending-list);
 + kfree(pending-udc_name);
 + kfree(pending);
 + }
 + } else {
 + udc_bind_to_driver(udc, pending-driver);
 + list_del(pending-list);
 + kfree(pending);
 + }

This code can be simplified too:

if (!pending-udc_name || strcmp(pending-udc_name, 
dev_name(udc-dev) == 0) {
...
 + }
 +
   mutex_unlock(udc_lock);
  
   return 0;

Don't you want to add a driver back to the pending list 
usb_del_gadget_udc() after the call to usb_gadget_remove_driver?

 @@ -423,7 +450,27 @@ int usb_udc_attach_driver(const char *name, struct 
 usb_gadget_driver *driver)
   break;
   }
   if (ret) {
 - ret = -ENODEV;
 + struct pending_gadget_driver *pending;
 +
 + pending = kzalloc(sizeof(*pending), GFP_KERNEL);
 + if (!pending) {
 + ret = -ENOMEM;
 + goto out;
 + }
 + pending-driver = driver;
 + pending-udc_name = kstrdup(name, GFP_KERNEL);
 + if (!pending-udc_name) {
 + kfree(pending);
 + ret = -ENOMEM;
 + goto out;
 + }

Why do you need to copy the name?  Just use the original.

Alan Stern

--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/