[PATCH 0/1] power_supply: twl4030_charger modification

[PATCH 0/1] power_supply: twl4030_charger modification

[Heikki Krogerus]

Hi guys,

Grazvydas, I don't have hardware to test this with. If this is OK to
you, could you test it?

Felipe, was the idea that you take this with the atomic notifier
support for otg patches? I guess this is how it needs to be done in
order to keep things bisectable, right? Or does this simply go to
Anton?


Heikki Krogerus (1):
  twl4030_charger: Make the driver atomic notifier safe

 drivers/power/twl4030_charger.c |   25 +++++++++++++++++++------
 1 files changed, 19 insertions(+), 6 deletions(-)

[PATCH 1/1] twl4030_charger: Make the driver atomic notifier safe

[Heikki Krogerus]

This queues work from the otg notification where the
i2c operations can be safely made. Needed for atomic otg
notifiers.

Signed-off-by: Heikki Krogerus <heikki.krogerus@nokia.com>
Cc: Grazvydas Ignotas <notasas@gmail.com>
Cc: Anton Vorontsov <cbouatmailru@gmail.com>
---
 drivers/power/twl4030_charger.c |   25 +++++++++++++++++++------
 1 files changed, 19 insertions(+), 6 deletions(-)

diff --git a/drivers/power/twl4030_charger.c b/drivers/power/twl4030_charger.c
index ff1f423..92c16e1 100644
--- a/drivers/power/twl4030_charger.c
+++ b/drivers/power/twl4030_charger.c
@@ -71,8 +71,11 @@ struct twl4030_bci {
 	struct power_supply	usb;
 	struct otg_transceiver	*transceiver;
 	struct notifier_block	otg_nb;
+	struct work_struct	work;
 	int			irq_chg;
 	int			irq_bci;
+
+	unsigned long		event;
 };
 
 /*
@@ -258,14 +261,11 @@ static irqreturn_t twl4030_bci_interrupt(int irq, void *arg)
 	return IRQ_HANDLED;
 }
 
-static int twl4030_bci_usb_ncb(struct notifier_block *nb, unsigned long val,
-			       void *priv)
+static void twl4030_bci_usb_work(struct work_struct *data)
 {
-	struct twl4030_bci *bci = container_of(nb, struct twl4030_bci, otg_nb);
+	struct twl4030_bci *bci = container_of(data, struct twl4030_bci, work);
 
-	dev_dbg(bci->dev, "OTG notify %lu\n", val);
-
-	switch (val) {
+	switch (bci->event) {
 	case USB_EVENT_VBUS:
 	case USB_EVENT_CHARGER:
 		twl4030_charger_enable_usb(bci, true);
@@ -274,6 +274,17 @@ static int twl4030_bci_usb_ncb(struct notifier_block *nb, unsigned long val,
 		twl4030_charger_enable_usb(bci, false);
 		break;
 	}
+}
+
+static int twl4030_bci_usb_ncb(struct notifier_block *nb, unsigned long val,
+			       void *priv)
+{
+	struct twl4030_bci *bci = container_of(nb, struct twl4030_bci, otg_nb);
+
+	dev_dbg(bci->dev, "OTG notify %lu\n", val);
+
+	bci->event = val;
+	schedule_work(&bci->work);
 
 	return NOTIFY_OK;
 }
@@ -466,6 +477,8 @@ static int __init twl4030_bci_probe(struct platform_device *pdev)
 		goto fail_bci_irq;
 	}
 
+	INIT_WORK(&bci->work, twl4030_bci_usb_work);
+
 	bci->transceiver = otg_get_transceiver();
 	if (bci->transceiver != NULL) {
 		bci->otg_nb.notifier_call = twl4030_bci_usb_ncb;
-- 
1.7.1

Re: [PATCH 0/1] power_supply: twl4030_charger modification

[Felipe Balbi]

On Fri, Feb 25, 2011 at 01:56:35PM +0200, Heikki Krogerus wrote:
> Hi guys,
> 
> Grazvydas, I don't have hardware to test this with. If this is OK to
> you, could you test it?
> 
> Felipe, was the idea that you take this with the atomic notifier
> support for otg patches? I guess this is how it needs to be done in
> order to keep things bisectable, right? Or does this simply go to
> Anton?

We can't change patches anymore :-( So as long as it goes on the merge
window, we should be ok, hopefully.

-- 
balbi

Re: [PATCH 1/1] twl4030_charger: Make the driver atomic notifier safe

[Grazvydas Ignotas]

On Fri, Feb 25, 2011 at 1:56 PM, Heikki Krogerus
<heikki.krogerus@nokia.com> wrote:
> This queues work from the otg notification where the
> i2c operations can be safely made. Needed for atomic otg
> notifiers.
>
> Signed-off-by: Heikki Krogerus <heikki.krogerus@nokia.com>
> Cc: Grazvydas Ignotas <notasas@gmail.com>
> Cc: Anton Vorontsov <cbouatmailru@gmail.com>

seems to work fine.
Tested-by: Grazvydas Ignotas <notasas@gmail.com>

> ---
>  drivers/power/twl4030_charger.c |   25 +++++++++++++++++++------
>  1 files changed, 19 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/power/twl4030_charger.c b/drivers/power/twl4030_charger.c
> index ff1f423..92c16e1 100644
> --- a/drivers/power/twl4030_charger.c
> +++ b/drivers/power/twl4030_charger.c
> @@ -71,8 +71,11 @@ struct twl4030_bci {
>        struct power_supply     usb;
>        struct otg_transceiver  *transceiver;
>        struct notifier_block   otg_nb;
> +       struct work_struct      work;
>        int                     irq_chg;
>        int                     irq_bci;
> +
> +       unsigned long           event;
>  };
>
>  /*
> @@ -258,14 +261,11 @@ static irqreturn_t twl4030_bci_interrupt(int irq, void *arg)
>        return IRQ_HANDLED;
>  }
>
> -static int twl4030_bci_usb_ncb(struct notifier_block *nb, unsigned long val,
> -                              void *priv)
> +static void twl4030_bci_usb_work(struct work_struct *data)
>  {
> -       struct twl4030_bci *bci = container_of(nb, struct twl4030_bci, otg_nb);
> +       struct twl4030_bci *bci = container_of(data, struct twl4030_bci, work);
>
> -       dev_dbg(bci->dev, "OTG notify %lu\n", val);
> -
> -       switch (val) {
> +       switch (bci->event) {
>        case USB_EVENT_VBUS:
>        case USB_EVENT_CHARGER:
>                twl4030_charger_enable_usb(bci, true);
> @@ -274,6 +274,17 @@ static int twl4030_bci_usb_ncb(struct notifier_block *nb, unsigned long val,
>                twl4030_charger_enable_usb(bci, false);
>                break;
>        }
> +}
> +
> +static int twl4030_bci_usb_ncb(struct notifier_block *nb, unsigned long val,
> +                              void *priv)
> +{
> +       struct twl4030_bci *bci = container_of(nb, struct twl4030_bci, otg_nb);
> +
> +       dev_dbg(bci->dev, "OTG notify %lu\n", val);
> +
> +       bci->event = val;
> +       schedule_work(&bci->work);
>
>        return NOTIFY_OK;
>  }
> @@ -466,6 +477,8 @@ static int __init twl4030_bci_probe(struct platform_device *pdev)
>                goto fail_bci_irq;
>        }
>
> +       INIT_WORK(&bci->work, twl4030_bci_usb_work);
> +
>        bci->transceiver = otg_get_transceiver();
>        if (bci->transceiver != NULL) {
>                bci->otg_nb.notifier_call = twl4030_bci_usb_ncb;
> --
> 1.7.1
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/1] power_supply: twl4030_charger modification

[Grazvydas Ignotas]

On Fri, Feb 25, 2011 at 2:12 PM, Felipe Balbi <balbi@ti.com> wrote:
> On Fri, Feb 25, 2011 at 01:56:35PM +0200, Heikki Krogerus wrote:
>> Hi guys,
>>
>> Grazvydas, I don't have hardware to test this with. If this is OK to
>> you, could you test it?

Sure, replied to your patch.

>> Felipe, was the idea that you take this with the atomic notifier
>> support for otg patches? I guess this is how it needs to be done in
>> order to keep things bisectable, right? Or does this simply go to
>> Anton?
>
> We can't change patches anymore :-( So as long as it goes on the merge
> window, we should be ok, hopefully.

Talking about notifiers, do we have away to find the amount of current
host can provide yet? I haven't followed USB development closely
lately.

Re: [PATCH 0/1] power_supply: twl4030_charger modification

[Heikki Krogerus]

On Sat, Feb 26, 2011 at 08:10:17PM +0200, ext Grazvydas Ignotas wrote:
> On Fri, Feb 25, 2011 at 2:12 PM, Felipe Balbi <balbi@ti.com> wrote:
> > On Fri, Feb 25, 2011 at 01:56:35PM +0200, Heikki Krogerus wrote:
> >> Hi guys,
> >>
> >> Grazvydas, I don't have hardware to test this with. If this is OK to
> >> you, could you test it?
> 
> Sure, replied to your patch.

Thanks!

> >> Felipe, was the idea that you take this with the atomic notifier
> >> support for otg patches? I guess this is how it needs to be done in
> >> order to keep things bisectable, right? Or does this simply go to
> >> Anton?
> >
> > We can't change patches anymore :-( So as long as it goes on the merge
> > window, we should be ok, hopefully.
> 
> Talking about notifiers, do we have away to find the amount of current
> host can provide yet? I haven't followed USB development closely
> lately.

There is nothing to support this yet. I'm only assuming the ENUMERATED
event will be used for this. Now with the atomic notifiers it should
be OK for the controller drivers to send this notification always when
usb_gadget_vbus_draw() is called.

But Felipe is working with the udc class. Maybe he has some other
method in mind? For now I think we could send the ENUMERATED
event from all the controller drivers. Felipe?

-- 
heikki

Re: [PATCH 1/1] twl4030_charger: Make the driver atomic notifier safe

[Heikki Krogerus]

On Sat, Feb 26, 2011 at 07:53:03PM +0200, ext Grazvydas Ignotas wrote:
> On Fri, Feb 25, 2011 at 1:56 PM, Heikki Krogerus
> <heikki.krogerus@nokia.com> wrote:
> > This queues work from the otg notification where the
> > i2c operations can be safely made. Needed for atomic otg
> > notifiers.
> >
> > Signed-off-by: Heikki Krogerus <heikki.krogerus@nokia.com>
> > Cc: Grazvydas Ignotas <notasas@gmail.com>
> > Cc: Anton Vorontsov <cbouatmailru@gmail.com>
> 
> seems to work fine.
> Tested-by: Grazvydas Ignotas <notasas@gmail.com>

Anton, can you take this?

-- 
heikki

Re: [PATCH 1/1] twl4030_charger: Make the driver atomic notifier safe

[Anton Vorontsov]

On Mon, Feb 28, 2011 at 08:29:53AM +0200, Heikki Krogerus wrote:
> On Sat, Feb 26, 2011 at 07:53:03PM +0200, ext Grazvydas Ignotas wrote:
> > On Fri, Feb 25, 2011 at 1:56 PM, Heikki Krogerus
> > <heikki.krogerus@nokia.com> wrote:
> > > This queues work from the otg notification where the
> > > i2c operations can be safely made. Needed for atomic otg
> > > notifiers.
> > >
> > > Signed-off-by: Heikki Krogerus <heikki.krogerus@nokia.com>
> > > Cc: Grazvydas Ignotas <notasas@gmail.com>
> > > Cc: Anton Vorontsov <cbouatmailru@gmail.com>
> > 
> > seems to work fine.
> > Tested-by: Grazvydas Ignotas <notasas@gmail.com>
> 
> Anton, can you take this?

Applied, thanks!

-- 
Anton Vorontsov
Email: cbouatmailru@gmail.com