usb: phy: twl4030-usb fixes

PATCH Superseded Grazvydas Ignotas 3 participants 13 messages Mar 17, 2013 – Mar 21, 2013 5 messages not in the archive

Subsystems (3)

USB MUSB multipoint high speed dual-role controller OMAP USB support

lore

Share & tools

lore

Grazvydas Ignotas

author

2013-03-17 18:23 UTC #0 lore raw

v2 cover usb: phy: twl4030-usb fixes

I have a pandora board which has similar musb setup to beagleboard (OMAP3530 + TWL4030) and musb never worked well on it for me in mainline. Well it usually works if you plug the cable once, but as soon as you start replugging cables and mixing host adapter into the game it totally breaks and reboot is then needed. Host mode is especially broken, any replugs after musb has been in host mode result in dead port that needs reboot to recover.

With this series I can switch host/peripheral cables any way I like and even suspend works with cable plugged with musb in peripheral mode! ("ARM: OMAP3: hwmod data: keep MIDLEMODE in force-standby for musb" is needed that was sent separately). This also fixes power drain when cable is plugged an no gadget driver is loaded.

Changed since v1:
- rebased on Felipe's testing branch
- added locking for patch 4 to take care of possible races between work item and IRQ
- changed patch 6 to only disable VBUS if not runtime suspended, otherwise we get data abort on OMAP3

Grazvydas Ignotas (7):
usb: phy: twl4030-usb: don't enable PHY during init
usb: phy: twl4030-usb: ignore duplicate events
usb: phy: twl4030-usb: don't switch the phy on/off needlessly
usb: phy: twl4030-usb: poll for ID disconnect
usb: phy: twl4030-usb: check if vbus is driven by twl itself
usb: musb: omap2430: turn off vbus on cable disconnect
usb: musb: gadget: use platform callback to enable vbus

 drivers/usb/musb/musb_gadget.c    |    5 +-
 drivers/usb/musb/omap2430.c       |    1 +
 drivers/usb/phy/phy-twl4030-usb.c |  123 +++++++++++++++++++++++++++++--------
 3 files changed, 99 insertions(+), 30 deletions(-)

Signature · 1 line

1.7.9.5

A message that is not in the archive (sent off-list or not yet received): 1363544607-17634-1-git-send-email-notasas-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org

Grazvydas Ignotas

2013-03-17 18:23 UTC #2 lore raw

PATCH v2 1/7 usb: phy: twl4030-usb: don't enable PHY during init

There is no need to do it, otg.set_suspend(false) (which itself comes from runtime_pm OMAP glue calls) will enable it later anyway. This used to be the place where things were enabled if booted with cable connected before runtime_pm conversion, but now can be dropped.

Signed-off-by Grazvydas Ignotas <[email protected]>

---
 drivers/usb/phy/phy-twl4030-usb.c |   20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)

drivers/usb/phy/phy-twl4030-usb.c +9 ~~-11~~

index a994715..9e47118 100644
@@ -522,19 +522,17 @@ static void twl4030_usb_phy_init(struct twl4030_usb *twl)
 {
 	enum omap_musb_vbus_id_status status;
 
-	status = twl4030_usb_linkstat(twl);
-	if (status > 0) {
-		if (status == OMAP_MUSB_VBUS_OFF ||
-				status == OMAP_MUSB_ID_FLOAT) {
-			__twl4030_phy_power(twl, 0);
-			twl->asleep = 1;
-		} else {
-			__twl4030_phy_resume(twl);
-			twl->asleep = 0;
-		}
+	/*
+	 * Start in sleep state, we'll get called through set_suspend()
+	 * callback when musb is runtime resumed and it's time to start.
+	 */
+	__twl4030_phy_power(twl, 0);
+	twl->asleep = 1;
 
+	status = twl4030_usb_linkstat(twl);
+	if (status > 0)
 		omap_musb_mailbox(twl->linkstat);
-	}
+
 	sysfs_notify(&twl->dev->kobj, NULL, "vbus");
 }

Signature · 6 lines

1.7.9.5

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Grazvydas Ignotas

2013-03-17 18:23 UTC #3 lore raw

PATCH v2 2/7 usb: phy: twl4030-usb: ignore duplicate events

In reply to a message not in the archive

In some rare cases we may get multiple interrupts that will generate duplicate omap_musb_mailbox() calls. This is a problem because each VBUS/ID event generates runtime_pm call in OMAP glue code, causing unbalanced gets or puts and breaking PM.

The same goes for initial state, glue already defaults to "no cable" state, so only bother it if we have VBUS or ID.

Signed-off-by Grazvydas Ignotas <[email protected]>

Reviewed-by Kishon Vijay Abraham I <[email protected]>

---
 drivers/usb/phy/phy-twl4030-usb.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

drivers/usb/phy/phy-twl4030-usb.c +3 -2

index 9e47118..305463b 100644
@@ -491,9 +491,10 @@ static irqreturn_t twl4030_usb_irq(int irq, void *_twl)
 {
 	struct twl4030_usb *twl = _twl;
 	enum omap_musb_vbus_id_status status;
+	enum omap_musb_vbus_id_status status_prev = twl->linkstat;
 
 	status = twl4030_usb_linkstat(twl);
-	if (status > 0) {
+	if (status > 0 && status != status_prev) {
 		/* FIXME add a set_power() method so that B-devices can
 		 * configure the charger appropriately.  It's not always
 		 * correct to consume VBUS power, and how much current to

@@ -530,7 +531,7 @@ static void twl4030_usb_phy_init(struct twl4030_usb *twl)
 	twl->asleep = 1;
 
 	status = twl4030_usb_linkstat(twl);
-	if (status > 0)
+	if (status == OMAP_MUSB_ID_GROUND || status == OMAP_MUSB_VBUS_VALID)
 		omap_musb_mailbox(twl->linkstat);
 
 	sysfs_notify(&twl->dev->kobj, NULL, "vbus");

Signature · 6 lines

1.7.9.5

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Grazvydas Ignotas

2013-03-17 18:23 UTC #4 lore raw

PATCH v2 4/7 usb: phy: twl4030-usb: poll for ID disconnect

In reply to a message not in the archive

On pandora, STS_USB interrupt doesn't arrive on USB host cable disconnect for some reason while VBUS is driven by twl itself, but STS_HW_CONDITIONS is updated correctly. It does work fine when PHY is powered down though. To work around that we have to poll.

This patch also moves twl->linkstat update code to callers so that changes can be handled in thread safe way (as polling work can trigger at the same time as real irq now).

TI PSP kernels have similar workarounds, so (many?) more boards are likely affected.

Signed-off-by Grazvydas Ignotas <[email protected]>

---
 drivers/usb/phy/phy-twl4030-usb.c |   64 +++++++++++++++++++++++++++++++++----
 1 file changed, 57 insertions(+), 7 deletions(-)

drivers/usb/phy/phy-twl4030-usb.c +57 -7

index b53a2a2..425c18a 100644
@@ -163,6 +163,8 @@ struct twl4030_usb {
 	bool			vbus_supplied;
 	u8			asleep;
 	bool			irq_enabled;
+
+	struct delayed_work	id_workaround_work;
 };
 
 /* internal define on top of container_of */

@@ -287,10 +289,6 @@ static enum omap_musb_vbus_id_status
 	 * are registered, and that both are active...
 	 */
 
-	spin_lock_irq(&twl->lock);
-	twl->linkstat = linkstat;
-	spin_unlock_irq(&twl->lock);
-
 	return linkstat;
 }

@@ -412,6 +410,16 @@ static void twl4030_phy_resume(struct twl4030_usb *twl)
 	__twl4030_phy_resume(twl);
 	twl->asleep = 0;
 	dev_dbg(twl->dev, "%s\n", __func__);
+
+	/*
+	 * XXX When VBUS gets driven after musb goes to A mode,
+	 * ID_PRES related interrupts no longer arrive, why?
+	 * Register itself is updated fine though, so we must poll.
+	 */
+	if (twl->linkstat == OMAP_MUSB_ID_GROUND) {
+		cancel_delayed_work(&twl->id_workaround_work);
+		schedule_delayed_work(&twl->id_workaround_work, HZ);
+	}
 }
 
 static int twl4030_usb_ldo_init(struct twl4030_usb *twl)

@@ -491,10 +499,18 @@ static irqreturn_t twl4030_usb_irq(int irq, void *_twl)
 {
 	struct twl4030_usb *twl = _twl;
 	enum omap_musb_vbus_id_status status;
-	enum omap_musb_vbus_id_status status_prev = twl->linkstat;
+	bool status_changed = false;
 
 	status = twl4030_usb_linkstat(twl);
-	if (status > 0 && status != status_prev) {
+
+	spin_lock_irq(&twl->lock);
+	if (status >= 0 && status != twl->linkstat) {
+		twl->linkstat = status;
+		status_changed = true;
+	}
+	spin_unlock_irq(&twl->lock);
+
+	if (status_changed) {
 		/* FIXME add a set_power() method so that B-devices can
 		 * configure the charger appropriately.  It's not always
 		 * correct to consume VBUS power, and how much current to

@@ -506,13 +522,42 @@ static irqreturn_t twl4030_usb_irq(int irq, void *_twl)
 		 * USB_LINK_VBUS state.  musb_hdrc won't care until it
 		 * starts to handle softconnect right.
 		 */
-		omap_musb_mailbox(twl->linkstat);
+		omap_musb_mailbox(status);
 	}
 	sysfs_notify(&twl->dev->kobj, NULL, "vbus");
 
 	return IRQ_HANDLED;
 }
 
+static void twl4030_id_workaround_work(struct work_struct *work)
+{
+	struct twl4030_usb *twl = container_of(work, struct twl4030_usb,
+		id_workaround_work.work);
+	enum omap_musb_vbus_id_status status;
+	bool status_changed = false;
+
+	status = twl4030_usb_linkstat(twl);
+
+	spin_lock_irq(&twl->lock);
+	if (status >= 0 && status != twl->linkstat) {
+		twl->linkstat = status;
+		status_changed = true;
+	}
+	spin_unlock_irq(&twl->lock);
+
+	if (status_changed) {
+		dev_dbg(twl->dev, "handle missing status change to %d\n",
+			status);
+		omap_musb_mailbox(status);
+	}
+
+	/* don't schedule during sleep - irq works right then */
+	if (status == OMAP_MUSB_ID_GROUND && !twl->asleep) {
+		cancel_delayed_work(&twl->id_workaround_work);
+		schedule_delayed_work(&twl->id_workaround_work, HZ);
+	}
+}
+
 static void twl4030_usb_phy_init(struct twl4030_usb *twl)
 {
 	enum omap_musb_vbus_id_status status;

@@ -525,6 +570,8 @@ static void twl4030_usb_phy_init(struct twl4030_usb *twl)
 	twl->asleep = 1;
 
 	status = twl4030_usb_linkstat(twl);
+	twl->linkstat = status;
+
 	if (status == OMAP_MUSB_ID_GROUND || status == OMAP_MUSB_VBUS_VALID)
 		omap_musb_mailbox(twl->linkstat);

@@ -613,6 +660,8 @@ static int twl4030_usb_probe(struct platform_device *pdev)
 	/* init spinlock for workqueue */
 	spin_lock_init(&twl->lock);
 
+	INIT_DELAYED_WORK(&twl->id_workaround_work, twl4030_id_workaround_work);
+
 	err = twl4030_usb_ldo_init(twl);
 	if (err) {
 		dev_err(&pdev->dev, "ldo init failed\n");

@@ -656,6 +705,7 @@ static int __exit twl4030_usb_remove(struct platform_device *pdev)
 	struct twl4030_usb *twl = platform_get_drvdata(pdev);
 	int val;
 
+	cancel_delayed_work(&twl->id_workaround_work);
 	free_irq(twl->irq, twl);
 	device_remove_file(twl->dev, &dev_attr_vbus);

Signature · 6 lines

1.7.9.5

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Grazvydas Ignotas

2013-03-17 18:23 UTC #5 lore raw

PATCH v2 5/7 usb: phy: twl4030-usb: check if vbus is driven by twl itself

In reply to a message not in the archive

At least on pandora, STS_VBUS gets set even when VBUS is driven by twl itself. Reporting VBUS in this case confuses OMAP musb glue and charger driver, so check if OTG VBUS charge pump is on before reporting VBUS event to avoid this problem.

Signed-off-by Grazvydas Ignotas <[email protected]>

---
 drivers/usb/phy/phy-twl4030-usb.c |   36 +++++++++++++++++++++++++++++++-----
 1 file changed, 31 insertions(+), 5 deletions(-)

drivers/usb/phy/phy-twl4030-usb.c +31 -5

index 425c18a..87bf11d 100644
@@ -248,11 +248,31 @@ twl4030_usb_clear_bits(struct twl4030_usb *twl, u8 reg, u8 bits)
 
 /*-------------------------------------------------------------------------*/
 
+static bool twl4030_is_driving_vbus(struct twl4030_usb *twl)
+{
+	int ret;
+
+	ret = twl4030_usb_read(twl, PHY_CLK_CTRL_STS);
+	if (ret < 0 || !(ret & PHY_DPLL_CLK))
+		/*
+		 * if clocks are off, registers are not updated,
+		 * but we can assume we don't drive VBUS in this case
+		 */
+		return false;
+
+	ret = twl4030_usb_read(twl, ULPI_OTG_CTRL);
+	if (ret < 0)
+		return false;
+
+	return (ret & (ULPI_OTG_DRVVBUS | ULPI_OTG_CHRGVBUS)) ? true : false;
+}
+
 static enum omap_musb_vbus_id_status
 	twl4030_usb_linkstat(struct twl4030_usb *twl)
 {
 	int	status;
 	enum omap_musb_vbus_id_status linkstat = OMAP_MUSB_UNKNOWN;
+	bool	driving_vbus = false;
 
 	twl->vbus_supplied = false;

@@ -270,20 +290,26 @@ static enum omap_musb_vbus_id_status
 	if (status < 0)
 		dev_err(twl->dev, "USB link status err %d\n", status);
 	else if (status & (BIT(7) | BIT(2))) {
-		if (status & (BIT(7)))
-                        twl->vbus_supplied = true;
+		if (status & BIT(7)) {
+			driving_vbus = twl4030_is_driving_vbus(twl);
+			if (driving_vbus)
+				status &= ~BIT(7);
+		}
 
 		if (status & BIT(2))
 			linkstat = OMAP_MUSB_ID_GROUND;
-		else
+		else if (status & BIT(7)) {
 			linkstat = OMAP_MUSB_VBUS_VALID;
+			twl->vbus_supplied = true;
+		} else
+			linkstat = OMAP_MUSB_VBUS_OFF;
 	} else {
 		if (twl->linkstat != OMAP_MUSB_UNKNOWN)
 			linkstat = OMAP_MUSB_VBUS_OFF;
 	}
 
-	dev_dbg(twl->dev, "HW_CONDITIONS 0x%02x/%d; link %d\n",
-			status, status, linkstat);
+	dev_dbg(twl->dev, "HW_CONDITIONS 0x%02x; link %d, driving_vbus %d\n",
+		status, linkstat, driving_vbus);
 
 	/* REVISIT this assumes host and peripheral controllers
 	 * are registered, and that both are active...

Signature · 6 lines

1.7.9.5

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): 1363544607-17634-6-git-send-email-notasas-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org

Felipe Balbi

2013-03-20 13:07 UTC #7 lore raw

Re: [PATCHv2 5/7] usb: phy: twl4030-usb: check if vbus is driven by twl itself

Quoted 47 lines · Grazvydas Ignotas

Signed-off-by Grazvydas Ignotas <[email protected]>

---
 drivers/usb/phy/phy-twl4030-usb.c |   36 +++++++++++++++++++++++++++++++-----
 1 file changed, 31 insertions(+), 5 deletions(-)

drivers/usb/phy/phy-twl4030-usb.c +20 -0

index 425c18a..87bf11d 100644
@@ -248,11 +248,31 @@ twl4030_usb_clear_bits(struct twl4030_usb *twl, u8 reg, u8 bits)
 
 /*-------------------------------------------------------------------------*/
 
+static bool twl4030_is_driving_vbus(struct twl4030_usb *twl)
+{
+	int ret;
+
+	ret = twl4030_usb_read(twl, PHY_CLK_CTRL_STS);
+	if (ret < 0 || !(ret & PHY_DPLL_CLK))
+		/*
+		 * if clocks are off, registers are not updated,
+		 * but we can assume we don't drive VBUS in this case
+		 */
+		return false;
+
+	ret = twl4030_usb_read(twl, ULPI_OTG_CTRL);
+	if (ret < 0)
+		return false;
+
+	return (ret & (ULPI_OTG_DRVVBUS | ULPI_OTG_CHRGVBUS)) ? true : false;
+}
+
 static enum omap_musb_vbus_id_status
 	twl4030_usb_linkstat(struct twl4030_usb *twl)
 {
 	int	status;
 	enum omap_musb_vbus_id_status linkstat = OMAP_MUSB_UNKNOWN;
+	bool	driving_vbus = false;
 
 	twl->vbus_supplied = false;

drivers/usb/phy/phy-twl4030-usb.c static enum omap_musb_vbus_id_status · line 295

Quoted 8 lines

drivers/usb/phy/phy-twl4030-usb.c +3 -2

@@ -270,20 +290,26 @@ static enum omap_musb_vbus_id_status
 	if (status < 0)
 		dev_err(twl->dev, "USB link status err %d\n", status);
 	else if (status & (BIT(7) | BIT(2))) {
-		if (status & (BIT(7)))
-                        twl->vbus_supplied = true;
+		if (status & BIT(7)) {
+			driving_vbus = twl4030_is_driving_vbus(twl);
+			if (driving_vbus)

how about just:

if (twl4030_is_driving_vbus(twl))

	status &= ~BIT(7);

????

Signature · 1 line

balbi

A message that is not in the archive (sent off-list or not yet received): 20130320130750.GR29659-S8G//mZuvNWo5Im9Ml3/[email protected]

Grazvydas Ignotas

2013-03-21 13:42 UTC #9 lore raw

Re: [PATCHv2 5/7] usb: phy: twl4030-usb: check if vbus is driven by twl itself

Quoted 59 lines · Felipe Balbi

Quoted 47 lines · Grazvydas Ignotas

Signed-off-by Grazvydas Ignotas <[email protected]>

---
 drivers/usb/phy/phy-twl4030-usb.c |   36 +++++++++++++++++++++++++++++++-----
 1 file changed, 31 insertions(+), 5 deletions(-)

drivers/usb/phy/phy-twl4030-usb.c +20 -0

index 425c18a..87bf11d 100644
@@ -248,11 +248,31 @@ twl4030_usb_clear_bits(struct twl4030_usb *twl, u8 reg, u8 bits)
 
 /*-------------------------------------------------------------------------*/
 
+static bool twl4030_is_driving_vbus(struct twl4030_usb *twl)
+{
+     int ret;
+
+     ret = twl4030_usb_read(twl, PHY_CLK_CTRL_STS);
+     if (ret < 0 || !(ret & PHY_DPLL_CLK))
+             /*
+              * if clocks are off, registers are not updated,
+              * but we can assume we don't drive VBUS in this case
+              */
+             return false;
+
+     ret = twl4030_usb_read(twl, ULPI_OTG_CTRL);
+     if (ret < 0)
+             return false;
+
+     return (ret & (ULPI_OTG_DRVVBUS | ULPI_OTG_CHRGVBUS)) ? true : false;
+}
+
 static enum omap_musb_vbus_id_status
      twl4030_usb_linkstat(struct twl4030_usb *twl)
 {
      int     status;
      enum omap_musb_vbus_id_status linkstat = OMAP_MUSB_UNKNOWN;
+     bool    driving_vbus = false;
 
      twl->vbus_supplied = false;

drivers/usb/phy/phy-twl4030-usb.c static enum omap_musb_vbus_id_status · line 295

Quoted 8 lines

drivers/usb/phy/phy-twl4030-usb.c +3 -2

@@ -270,20 +290,26 @@ static enum omap_musb_vbus_id_status
      if (status < 0)
              dev_err(twl->dev, "USB link status err %d\n", status);
      else if (status & (BIT(7) | BIT(2))) {
-             if (status & (BIT(7)))
-                        twl->vbus_supplied = true;
+             if (status & BIT(7)) {
+                     driving_vbus = twl4030_is_driving_vbus(twl);
+                     if (driving_vbus)

how about just:

Quoted 4 lines

if (twl4030_is_driving_vbus(twl))
        status &= ~BIT(7);
????

I'm logging driving_vbus below with dev_dbg(), so that it's easier to see what going on..

Quoted 3 lines

--
balbi

Signature · 5 lines

Gražvydas
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): CANOLnOOS6X95ijodL903zqLn3emNZ-rDX0xrxYQUXFJ1PR-PZw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org

Felipe Balbi

2013-03-21 16:43 UTC #11 lore raw

Re: [PATCHv2 5/7] usb: phy: twl4030-usb: check if vbus is driven by twl itself

Hi,

Quoted 58 lines · Grazvydas Ignotas

Quoted 58 lines

Quoted 47 lines

Signed-off-by Grazvydas Ignotas <[email protected]>

---
 drivers/usb/phy/phy-twl4030-usb.c |   36 +++++++++++++++++++++++++++++++-----
 1 file changed, 31 insertions(+), 5 deletions(-)

drivers/usb/phy/phy-twl4030-usb.c +20 -0

index 425c18a..87bf11d 100644
@@ -248,11 +248,31 @@ twl4030_usb_clear_bits(struct twl4030_usb *twl, u8 reg, u8 bits)
 
 /*-------------------------------------------------------------------------*/
 
+static bool twl4030_is_driving_vbus(struct twl4030_usb *twl)
+{
+     int ret;
+
+     ret = twl4030_usb_read(twl, PHY_CLK_CTRL_STS);
+     if (ret < 0 || !(ret & PHY_DPLL_CLK))
+             /*
+              * if clocks are off, registers are not updated,
+              * but we can assume we don't drive VBUS in this case
+              */
+             return false;
+
+     ret = twl4030_usb_read(twl, ULPI_OTG_CTRL);
+     if (ret < 0)
+             return false;
+
+     return (ret & (ULPI_OTG_DRVVBUS | ULPI_OTG_CHRGVBUS)) ? true : false;
+}
+
 static enum omap_musb_vbus_id_status
      twl4030_usb_linkstat(struct twl4030_usb *twl)
 {
      int     status;
      enum omap_musb_vbus_id_status linkstat = OMAP_MUSB_UNKNOWN;
+     bool    driving_vbus = false;
 
      twl->vbus_supplied = false;

drivers/usb/phy/phy-twl4030-usb.c static enum omap_musb_vbus_id_status · line 295

Quoted 8 lines

drivers/usb/phy/phy-twl4030-usb.c +3 -2

@@ -270,20 +290,26 @@ static enum omap_musb_vbus_id_status
      if (status < 0)
              dev_err(twl->dev, "USB link status err %d\n", status);
      else if (status & (BIT(7) | BIT(2))) {
-             if (status & (BIT(7)))
-                        twl->vbus_supplied = true;
+             if (status & BIT(7)) {
+                     driving_vbus = twl4030_is_driving_vbus(twl);
+                     if (driving_vbus)

how about just:

Quoted 7 lines

Quoted 4 lines
if (twl4030_is_driving_vbus(twl))
        status &= ~BIT(7);
????
I'm logging driving_vbus below with dev_dbg(), so that it's easier to see what going on..

is that really necessary ? Don't we expose it through sysfs already ?

Signature · 1 line

balbi

Felipe Balbi

2013-03-20 12:54 UTC #12 lore raw

In reply to a message not in the archive

Quoted 21 lines · Grazvydas Ignotas

I have a pandora board which has similar musb setup to beagleboard (OMAP3530 + TWL4030) and musb never worked well on it for me in mainline. Well it usually works if you plug the cable once, but as soon as you start replugging cables and mixing host adapter into the game it totally breaks and reboot is then needed. Host mode is especially broken, any replugs after musb has been in host mode result in dead port that needs reboot to recover.
With this series I can switch host/peripheral cables any way I like and even suspend works with cable plugged with musb in peripheral mode! ("ARM: OMAP3: hwmod data: keep MIDLEMODE in force-standby for musb" is needed that was sent separately). This also fixes power drain when cable is plugged an no gadget driver is loaded.
Changed since v1:
- rebased on Felipe's testing branch
- added locking for patch 4 to take care of possible races between work item and IRQ
- changed patch 6 to only disable VBUS if not runtime suspended, otherwise we get data abort on OMAP3

Quoted 8 lines

Grazvydas Ignotas (7):
usb: phy: twl4030-usb: don't enable PHY during init
usb: phy: twl4030-usb: ignore duplicate events
usb: phy: twl4030-usb: don't switch the phy on/off needlessly
usb: phy: twl4030-usb: poll for ID disconnect
usb: phy: twl4030-usb: check if vbus is driven by twl itself
usb: musb: omap2430: turn off vbus on cable disconnect
usb: musb: gadget: use platform callback to enable vbus

since this falls into "has never worked before" I will apply them for v3.10. If you have any objections, let me know.

Signature · 1 line

balbi

A message that is not in the archive (sent off-list or not yet received): 20130320125425.GQ29659-S8G//mZuvNWo5Im9Ml3/[email protected]

Felipe Balbi

2013-03-20 13:08 UTC #14 lore raw

Quoted 31 lines · Felipe Balbi

Quoted 21 lines · Grazvydas Ignotas
I have a pandora board which has similar musb setup to beagleboard (OMAP3530 + TWL4030) and musb never worked well on it for me in mainline. Well it usually works if you plug the cable once, but as soon as you start replugging cables and mixing host adapter into the game it totally breaks and reboot is then needed. Host mode is especially broken, any replugs after musb has been in host mode result in dead port that needs reboot to recover.
With this series I can switch host/peripheral cables any way I like and even suspend works with cable plugged with musb in peripheral mode! ("ARM: OMAP3: hwmod data: keep MIDLEMODE in force-standby for musb" is needed that was sent separately). This also fixes power drain when cable is plugged an no gadget driver is loaded.
Changed since v1:
- rebased on Felipe's testing branch
- added locking for patch 4 to take care of possible races between work item and IRQ
- changed patch 6 to only disable VBUS if not runtime suspended, otherwise we get data abort on OMAP3
Quoted 8 lines
Grazvydas Ignotas (7):
usb: phy: twl4030-usb: don't enable PHY during init
usb: phy: twl4030-usb: ignore duplicate events
usb: phy: twl4030-usb: don't switch the phy on/off needlessly
usb: phy: twl4030-usb: poll for ID disconnect
usb: phy: twl4030-usb: check if vbus is driven by twl itself
usb: musb: omap2430: turn off vbus on cable disconnect
usb: musb: gadget: use platform callback to enable vbus

Quoted 2 lines

since this falls into "has never worked before" I will apply them for v3.10. If you have any objections, let me know.

I have applied up to 'poll for ID disconnect'. Please rebase the others against my 'testing' branch once you update based on my comments.

Signature · 1 line

balbi

Grazvydas Ignotas

author

2013-03-17 18:23 UTC #15 lore raw

PATCH v2 3/7 usb: phy: twl4030-usb: don't switch the phy on/off needlessly

In reply to Grazvydas Ignotas

With runtime_pm in place there is no longer need to turn the phy on/off in OTG layer on cable connect/disconnect, OMAP glue does this through otg.set_suspend() callback after it's called through omap_musb_mailbox() on VBUS/ID interrupt. Not doing this will save power when cable is connected but no gadget driver is loaded.

This will also have side effect of automatic USB charging no longer working without twl4030_charger driver, because a regulator needed for charging will no longer be enabled, so be sure to enable charger driver if charging is needed.

Signed-off-by Grazvydas Ignotas <[email protected]>

---
 drivers/usb/phy/phy-twl4030-usb.c |    6 ------
 1 file changed, 6 deletions(-)

drivers/usb/phy/phy-twl4030-usb.c +0 -6

index 305463b..b53a2a2 100644
@@ -506,12 +506,6 @@ static irqreturn_t twl4030_usb_irq(int irq, void *_twl)
 		 * USB_LINK_VBUS state.  musb_hdrc won't care until it
 		 * starts to handle softconnect right.
 		 */
-		if (status == OMAP_MUSB_VBUS_OFF ||
-				status == OMAP_MUSB_ID_FLOAT)
-			twl4030_phy_suspend(twl, 0);
-		else
-			twl4030_phy_resume(twl);
-
 		omap_musb_mailbox(twl->linkstat);
 	}
 	sysfs_notify(&twl->dev->kobj, NULL, "vbus");

Signature · 1 line

1.7.9.5

Grazvydas Ignotas

author

2013-03-17 18:23 UTC #16 lore raw

PATCH v2 6/7 usb: musb: omap2430: turn off vbus on cable disconnect

In reply to Grazvydas Ignotas

On USB_EVENT_ID event the musb glue enables VBUS by calling omap2430_musb_set_vbus(musb, 1) that sets the session bit, but on USB_EVENT_NONE reverse action is never made, and that breaks PM.

Disable VBUS on USB_EVENT_NONE to be sure musb session is ended on cable unplug so that PM works.

Signed-off-by Grazvydas Ignotas <[email protected]>

---
 drivers/usb/musb/omap2430.c |    1 +
 1 file changed, 1 insertion(+)

drivers/usb/musb/omap2430.c +1 -0

index ec460ea..780a750 100644
@@ -291,6 +291,7 @@ static void omap_musb_set_mailbox(struct omap2430_glue *glue)
 
 		musb->xceiv->last_event = USB_EVENT_NONE;
 		if (musb->gadget_driver) {
+			omap2430_musb_set_vbus(musb, 0);
 			pm_runtime_mark_last_busy(dev);
 			pm_runtime_put_autosuspend(dev);
 		}

Signature · 1 line

1.7.9.5

Grazvydas Ignotas

author

2013-03-17 18:23 UTC #17 lore raw

PATCH v2 7/7 usb: musb: gadget: use platform callback to enable vbus

In reply to Grazvydas Ignotas

On some platform configurations (like OMAP3+twl4030) it's the platform code that enables VBUS, not OTG transceiver, so call vbus platform callback instead, it will then call the transceiver if needed.

This fixes a use case where USB cable is plugged first and gadget driver is loaded later after that.

Signed-off-by Grazvydas Ignotas <[email protected]>

---
 drivers/usb/musb/musb_gadget.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

drivers/usb/musb/musb_gadget.c +2 -3

index ae59ee6..60eef20 100644
@@ -1799,9 +1799,8 @@ static int musb_gadget_start(struct usb_gadget *g,
 		goto err;
 	}
 
-	if ((musb->xceiv->last_event == USB_EVENT_ID)
-				&& otg->set_vbus)
-		otg_set_vbus(otg, 1);
+	if (musb->xceiv->last_event == USB_EVENT_ID)
+		musb_platform_set_vbus(musb, 1);
 
 	hcd->self.uses_pio_for_control = 1;

Signature · 1 line

1.7.9.5