Skip to contentSkip to conversation

usb: host: xhci-plat: fix support for XHCI_SKIP_PHY_INIT quirk

PATCH Superseded Pali Rohár 4 participants 15 messages Dec 21, 2020 – Feb 2, 2021

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 usb: host: xhci-plat: fix support for XHCI_SKIP_PHY_INIT quirk

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set.

Fixes f768e718911e0 ("usb: host: xhci-plat: add priv quirk for skip PHY initialization")
---
 drivers/usb/host/xhci-plat.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/usb/host/xhci-plat.c +7 -7
index 4d34f6005381..58704c5b002b 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,6 +323,13 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
 	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
 		hcd->skip_phy_initialization = 1;
 
Signature · 1 line
2.20.1
#1 lore raw
Quoted 44 lines · Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set.

Fixes f768e718911e0 ("usb: host: xhci-plat: add priv quirk for skip PHY initialization")
---
 drivers/usb/host/xhci-plat.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/usb/host/xhci-plat.c +0 -7
index 4d34f6005381..58704c5b002b 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,6 +323,13 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;

drivers/usb/host/xhci-plat.c static int xhci_plat_probe(struct platform_device *pdev) · line 335

Quoted 10 lines
drivers/usb/host/xhci-plat.c +7 -0
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
 	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
 		hcd->skip_phy_initialization = 1;
 

Hi Pali,

What's problem you have met? In structure xhci_plat_priv, the quirks are defined at .quirks entry which is got at below code. .init_quirk is the routine if special initializations are needed.

	if (pdev->dev.of_node)
		priv_match = of_device_get_match_data(&pdev->dev);
	else
		priv_match = dev_get_platdata(&pdev->dev);

	if (priv_match) {
		priv = hcd_to_xhci_priv(hcd);
		/* Just copy data for now */
		*priv = *priv_match;
	}
Signature · 2 lines
Thanks,
Peter Chen
#2 lore raw
Quoted 58 lines · On Tuesday 22 December 2020 02:14:45 Peter Chen wrote:
Quoted 44 lines · Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set.

Fixes f768e718911e0 ("usb: host: xhci-plat: add priv quirk for skip PHY initialization")
---
 drivers/usb/host/xhci-plat.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/usb/host/xhci-plat.c +0 -7
index 4d34f6005381..58704c5b002b 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,6 +323,13 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;

drivers/usb/host/xhci-plat.c static int xhci_plat_probe(struct platform_device *pdev) · line 335

Quoted 10 lines
drivers/usb/host/xhci-plat.c +7 -0
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
 	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
 		hcd->skip_phy_initialization = 1;
 

Hi Pali,

Quoted 3 lines

What's problem you have met? In structure xhci_plat_priv, the quirks are defined at .quirks entry which is got at below code. .init_quirk is the routine if special initializations are needed.

Hello!

I'm talking about .init_quirk. And if XHCI_SKIP_PHY_INIT quirk is set in this function then has no effect.

I'm working currently on patch for xhci mvebu which conditionally enable or disable XHCI_SKIP_PHY_INIT quirk (it is going to fix existing regression since v5.1 kernel) and without this patch XHCI_SKIP_PHY_INIT quirk from the init_quirk does not work.

Quoted 14 lines
	if (pdev->dev.of_node)
		priv_match = of_device_get_match_data(&pdev->dev);
	else
		priv_match = dev_get_platdata(&pdev->dev);

	if (priv_match) {
		priv = hcd_to_xhci_priv(hcd);
		/* Just copy data for now */
		*priv = *priv_match;
	}
Signature · 2 lines
Thanks,
Peter Chen
#3 lore raw
Quoted 59 lines · On Tuesday 22 December 2020 10:23:27 Pali Rohár wrote:
Quoted 58 lines · On Tuesday 22 December 2020 02:14:45 Peter Chen wrote:
Quoted 44 lines · Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set.

Fixes f768e718911e0 ("usb: host: xhci-plat: add priv quirk for skip PHY initialization")
---
 drivers/usb/host/xhci-plat.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/usb/host/xhci-plat.c +0 -7
index 4d34f6005381..58704c5b002b 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,6 +323,13 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;

drivers/usb/host/xhci-plat.c static int xhci_plat_probe(struct platform_device *pdev) · line 335

Quoted 10 lines
drivers/usb/host/xhci-plat.c +7 -0
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
 	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
 		hcd->skip_phy_initialization = 1;
 

Hi Pali,

Quoted 8 lines
Quoted 3 lines

What's problem you have met? In structure xhci_plat_priv, the quirks are defined at .quirks entry which is got at below code. .init_quirk is the routine if special initializations are needed.

Hello!

I'm talking about .init_quirk. And if XHCI_SKIP_PHY_INIT quirk is set in this function then has no effect.

Ok, this patch is not enough, I will send V2.

Quoted 4 lines

I'm working currently on patch for xhci mvebu which conditionally enable or disable XHCI_SKIP_PHY_INIT quirk (it is going to fix existing regression since v5.1 kernel) and without this patch XHCI_SKIP_PHY_INIT quirk from the init_quirk does not work.

And now I have tested V2 with my mvebu regression fix. I will send it to mailing list for review.

Quoted 14 lines
Quoted 14 lines
	if (pdev->dev.of_node)
		priv_match = of_device_get_match_data(&pdev->dev);
	else
		priv_match = dev_get_platdata(&pdev->dev);

	if (priv_match) {
		priv = hcd_to_xhci_priv(hcd);
		/* Just copy data for now */
		*priv = *priv_match;
	}
Signature · 2 lines
Thanks,
Peter Chen
#4 lore raw
Quoted 15 lines · Pali Rohár
Quoted 14 lines · On Tuesday 22 December 2020 10:23:27 Pali Rohár wrote:
Quoted 13 lines · On Tuesday 22 December 2020 02:14:45 Peter Chen wrote:
Quoted 12 lines · Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set.

Fixes f768e718911e0 ("usb: host: xhci-plat: add priv quirk for skip PHY initialization")

Hi Pali,

I know your case, you need to choose XHCI_SKIP_PHY_INIT quirk conditionally, but you may can't add Fixes tag at your patch since your issue was existed before my patch.

Peter

Quoted 72 lines
Quoted 39 lines
Quoted 39 lines
Quoted 26 lines
---
 drivers/usb/host/xhci-plat.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/usb/host/xhci-plat.c +0 -7
index 4d34f6005381..58704c5b002b 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,6 +323,13 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;

drivers/usb/host/xhci-plat.c static int xhci_plat_probe(struct platform_device *pdev) · line 335

Quoted 10 lines
drivers/usb/host/xhci-plat.c +7 -0
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
 	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
 		hcd->skip_phy_initialization = 1;
 

Hi Pali,

Quoted 8 lines
Quoted 3 lines

What's problem you have met? In structure xhci_plat_priv, the quirks are defined at .quirks entry which is got at below code. .init_quirk is the routine if special initializations are needed.

Hello!

I'm talking about .init_quirk. And if XHCI_SKIP_PHY_INIT quirk is set in this function then has no effect.

Ok, this patch is not enough, I will send V2.

Quoted 4 lines

I'm working currently on patch for xhci mvebu which conditionally enable or disable XHCI_SKIP_PHY_INIT quirk (it is going to fix existing regression since v5.1 kernel) and without this patch XHCI_SKIP_PHY_INIT quirk from the init_quirk does not work.

And now I have tested V2 with my mvebu regression fix. I will send it to mailing list for review.

Quoted 14 lines
Quoted 14 lines
	if (pdev->dev.of_node)
		priv_match = of_device_get_match_data(&pdev->dev);
	else
		priv_match = dev_get_platdata(&pdev->dev);

	if (priv_match) {
		priv = hcd_to_xhci_priv(hcd);
		/* Just copy data for now */
		*priv = *priv_match;
	}
Signature · 2 lines
Thanks,
Peter Chen
Signature · 2 lines
Thanks,
Peter Chen
#5 lore raw
Quoted 20 lines · On Wednesday 23 December 2020 01:02:43 Peter Chen wrote:
Quoted 19 lines · Pali Rohár
Quoted 18 lines · On Tuesday 22 December 2020 10:23:27 Pali Rohár wrote:
Quoted 17 lines · On Tuesday 22 December 2020 02:14:45 Peter Chen wrote:
Quoted 16 lines · Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set.

Quoted 8 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Fixes f768e718911e0 ("usb: host: xhci-plat: add priv quirk for skip PHY initialization")

Hi Pali,

I know your case, you need to choose XHCI_SKIP_PHY_INIT quirk conditionally, but you may can't add Fixes tag at your patch since your issue was existed before my patch.

Ok, no problem, in V2 I will not add it.

Quoted 80 lines

Peter

Quoted 72 lines
Quoted 39 lines
Quoted 39 lines
Quoted 26 lines
---
 drivers/usb/host/xhci-plat.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/usb/host/xhci-plat.c +0 -7
index 4d34f6005381..58704c5b002b 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,6 +323,13 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;

drivers/usb/host/xhci-plat.c static int xhci_plat_probe(struct platform_device *pdev) · line 335

Quoted 10 lines
drivers/usb/host/xhci-plat.c +7 -0
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
 	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
 		hcd->skip_phy_initialization = 1;
 

Hi Pali,

Quoted 8 lines
Quoted 3 lines

What's problem you have met? In structure xhci_plat_priv, the quirks are defined at .quirks entry which is got at below code. .init_quirk is the routine if special initializations are needed.

Hello!

I'm talking about .init_quirk. And if XHCI_SKIP_PHY_INIT quirk is set in this function then has no effect.

Ok, this patch is not enough, I will send V2.

Quoted 4 lines

I'm working currently on patch for xhci mvebu which conditionally enable or disable XHCI_SKIP_PHY_INIT quirk (it is going to fix existing regression since v5.1 kernel) and without this patch XHCI_SKIP_PHY_INIT quirk from the init_quirk does not work.

And now I have tested V2 with my mvebu regression fix. I will send it to mailing list for review.

Quoted 14 lines
Quoted 14 lines
	if (pdev->dev.of_node)
		priv_match = of_device_get_match_data(&pdev->dev);
	else
		priv_match = dev_get_platdata(&pdev->dev);

	if (priv_match) {
		priv = hcd_to_xhci_priv(hcd);
		/* Just copy data for now */
		*priv = *priv_match;
	}
Signature · 2 lines
Thanks,
Peter Chen
Signature · 2 lines
Thanks,
Peter Chen
#6 lore raw

PATCH v2 1/1 usb: host: xhci-plat: fix support for XHCI_SKIP_PHY_INIT quirk

In reply to Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set. Also checking if either xhci->quirks or priv->quirks contains this XHCI_SKIP_PHY_INIT quirk.

---
Changes in v2:
* Check also xhci->quirks as xhci_priv_init_quirk() callbacks are setting xhci->quirks
* Tested with "usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720" patch
* Removed Fixes: line

---
 drivers/usb/host/xhci-plat.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/usb/host/xhci-plat.c +8 -8
index 4d34f6005381..0eab7cb5a767 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,7 +323,14 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;
-	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
+	if ((xhci->quirks & XHCI_SKIP_PHY_INIT) || (priv && (priv->quirks & XHCI_SKIP_PHY_INIT)))
 		hcd->skip_phy_initialization = 1;
 
 	if (priv && (priv->quirks & XHCI_SG_TRB_CACHE_SIZE_QUIRK))
Signature · 1 line
2.20.1
#7 lore raw
Quoted 50 lines · Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set. Also checking if either xhci->quirks or priv->quirks contains this XHCI_SKIP_PHY_INIT quirk.

---
Changes in v2:
* Check also xhci->quirks as xhci_priv_init_quirk() callbacks are setting xhci->quirks
* Tested with "usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720" patch
* Removed Fixes: line

---
 drivers/usb/host/xhci-plat.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/usb/host/xhci-plat.c +0 -7
index 4d34f6005381..0eab7cb5a767 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,7 +323,14 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;

drivers/usb/host/xhci-plat.c static int xhci_plat_probe(struct platform_device *pdev) · line 334

Quoted 10 lines
drivers/usb/host/xhci-plat.c +8 -1
-	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
+	if ((xhci->quirks & XHCI_SKIP_PHY_INIT) || (priv && (priv->quirks & XHCI_SKIP_PHY_INIT)))
 		hcd->skip_phy_initialization = 1;

I am not sure if others agree with you move the position of xhci_priv_init_quirk, Let's see Mathias opinion.

Signature · 2 lines
Thanks,
Peter Chen
#8 lore raw
Quoted 49 lines · On Thursday 24 December 2020 05:59:05 Peter Chen wrote:
Quoted 48 lines · Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set. Also checking if either xhci->quirks or priv->quirks contains this XHCI_SKIP_PHY_INIT quirk.

---
Changes in v2:
* Check also xhci->quirks as xhci_priv_init_quirk() callbacks are setting xhci->quirks
* Tested with "usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720" patch
* Removed Fixes: line

---
 drivers/usb/host/xhci-plat.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/usb/host/xhci-plat.c +0 -7
index 4d34f6005381..0eab7cb5a767 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,7 +323,14 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
Quoted 15 lines

drivers/usb/host/xhci-plat.c static int xhci_plat_probe(struct platform_device *pdev) · line 334

Quoted 12 lines
drivers/usb/host/xhci-plat.c +8 -1
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;
-	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
+	if ((xhci->quirks & XHCI_SKIP_PHY_INIT) || (priv && (priv->quirks & XHCI_SKIP_PHY_INIT)))
 		hcd->skip_phy_initialization = 1;

I am not sure if others agree with you move the position of xhci_priv_init_quirk, Let's see Mathias opinion.

Hello! Do you have an opinion how to handle this issue? As currently it is needed for another patch which is fixing issue/regression in xhci-mvebu:
https://lore.kernel.org/linux-usb/[email protected]/

#9 lore raw
Quoted 66 lines · Pali Rohár
Quoted 49 lines · On Thursday 24 December 2020 05:59:05 Peter Chen wrote:
Quoted 48 lines · Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set. Also checking if either xhci->quirks or priv->quirks contains this XHCI_SKIP_PHY_INIT quirk.

---
Changes in v2:
* Check also xhci->quirks as xhci_priv_init_quirk() callbacks are setting xhci->quirks
* Tested with "usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720" patch
* Removed Fixes: line

---
 drivers/usb/host/xhci-plat.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/usb/host/xhci-plat.c +0 -7
index 4d34f6005381..0eab7cb5a767 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,7 +323,14 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
Quoted 15 lines

drivers/usb/host/xhci-plat.c static int xhci_plat_probe(struct platform_device *pdev) · line 334

Quoted 12 lines
drivers/usb/host/xhci-plat.c +8 -1
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;
-	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
+	if ((xhci->quirks & XHCI_SKIP_PHY_INIT) || (priv && (priv->quirks & XHCI_SKIP_PHY_INIT)))
 		hcd->skip_phy_initialization = 1;

I am not sure if others agree with you move the position of xhci_priv_init_quirk, Let's see Mathias opinion.

Quoted 4 lines

Hello! Do you have an opinion how to handle this issue? As currently it is needed for another patch which is fixing issue/regression in xhci-mvebu:
https://lore.kernel.org/linux-usb/[email protected]/

I can see the benefit in this.
In the xhci-plat case usb_create_hcd and usb_add_hcd are separate steps, and we could both copy the xhci_plat_priv .quirks and run the .init_qurks before adding the hcd.
I guess the current way is inherited from pci case where the earliest place to do this after hcd is created is the hcd->driver->reset callback (which is set to xhci_pci_setup() or xhci_plat_setup()).

xhci-rcar.c is using the .init_quirk to load firmware, we need to check with them if this change is ok. (added Yoshihiro Shimoda to cc)
Their firmware would be loaded before phy parts are initialized, usb bus registered, or roothub device allocated.

Thanks
-Mathias

#10 lore raw
Quoted 80 lines · On Friday 15 January 2021 15:32:30 Mathias Nyman wrote:
Quoted 66 lines · Pali Rohár
Quoted 49 lines · On Thursday 24 December 2020 05:59:05 Peter Chen wrote:
Quoted 48 lines · Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set. Also checking if either xhci->quirks or priv->quirks contains this XHCI_SKIP_PHY_INIT quirk.

---
Changes in v2:
* Check also xhci->quirks as xhci_priv_init_quirk() callbacks are setting xhci->quirks
* Tested with "usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720" patch
* Removed Fixes: line

---
 drivers/usb/host/xhci-plat.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/usb/host/xhci-plat.c +0 -7
index 4d34f6005381..0eab7cb5a767 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,7 +323,14 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
Quoted 15 lines

drivers/usb/host/xhci-plat.c static int xhci_plat_probe(struct platform_device *pdev) · line 334

Quoted 12 lines
drivers/usb/host/xhci-plat.c +8 -1
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;
-	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
+	if ((xhci->quirks & XHCI_SKIP_PHY_INIT) || (priv && (priv->quirks & XHCI_SKIP_PHY_INIT)))
 		hcd->skip_phy_initialization = 1;

I am not sure if others agree with you move the position of xhci_priv_init_quirk, Let's see Mathias opinion.

Quoted 4 lines

Hello! Do you have an opinion how to handle this issue? As currently it is needed for another patch which is fixing issue/regression in xhci-mvebu:
https://lore.kernel.org/linux-usb/[email protected]/

I can see the benefit in this.
In the xhci-plat case usb_create_hcd and usb_add_hcd are separate steps, and we could both copy the xhci_plat_priv .quirks and run the .init_qurks before adding the hcd.
I guess the current way is inherited from pci case where the earliest place to do this after hcd is created is the hcd->driver->reset callback (which is set to xhci_pci_setup() or xhci_plat_setup()).

Quoted 2 lines

xhci-rcar.c is using the .init_quirk to load firmware, we need to check with them if this change is ok. (added Yoshihiro Shimoda to cc)

Yoshihiro, is this change OK?

Can we move forward? I really need to now how to handle regression in xhci-mvebu driver. And one option is with this patch...

Quoted 5 lines

Their firmware would be loaded before phy parts are initialized, usb bus registered, or roothub device allocated.

Thanks
-Mathias

#12 lore raw
In reply to Yoshihiro Shimoda
Quoted 93 lines · On Tuesday 26 January 2021 04:27:37 Yoshihiro Shimoda wrote:

Hi Pali,

Quoted 72 lines

From: Pali Rohár, Sent: Monday, January 25, 2021 11:20 PM

Quoted 70 lines · On Friday 15 January 2021 15:32:30 Mathias Nyman wrote:
Quoted 69 lines · Pali Rohár
Quoted 49 lines · On Thursday 24 December 2020 05:59:05 Peter Chen wrote:
Quoted 48 lines · Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set. Also checking if either xhci->quirks or priv->quirks contains this XHCI_SKIP_PHY_INIT quirk.

---
Changes in v2:
* Check also xhci->quirks as xhci_priv_init_quirk() callbacks are setting xhci->quirks
* Tested with "usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720" patch
* Removed Fixes: line

---
 drivers/usb/host/xhci-plat.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/usb/host/xhci-plat.c +0 -7
index 4d34f6005381..0eab7cb5a767 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,7 +323,14 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
Quoted 15 lines

drivers/usb/host/xhci-plat.c static int xhci_plat_probe(struct platform_device *pdev) · line 334

Quoted 12 lines
drivers/usb/host/xhci-plat.c +8 -1
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;
-	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
+	if ((xhci->quirks & XHCI_SKIP_PHY_INIT) || (priv && (priv->quirks & XHCI_SKIP_PHY_INIT)))
 		hcd->skip_phy_initialization = 1;

I am not sure if others agree with you move the position of xhci_priv_init_quirk, Let's see Mathias opinion.

Hello! Do you have an opinion how to handle this issue? As currently it is needed for another patch which is fixing issue/regression in xhci-mvebu:

<snip>

Quoted 10 lines
Quoted 10 lines
Quoted 1 line

I can see the benefit in this.
In the xhci-plat case usb_create_hcd and usb_add_hcd are separate steps, and we could both copy the xhci_plat_priv .quirks and run the .init_qurks before adding the hcd.
I guess the current way is inherited from pci case where the earliest place to do this after hcd is created is the hcd->driver->reset callback (which is set to xhci_pci_setup() or xhci_plat_setup()).

Quoted 7 lines
Quoted 2 lines

xhci-rcar.c is using the .init_quirk to load firmware, we need to check with them if this change is ok. (added Yoshihiro Shimoda to cc)

Yoshihiro, is this change OK?

Can we move forward? I really need to now how to handle regression in xhci-mvebu driver. And one option is with this patch...

Quoted 7 lines

Thank you for asking me about this topic. I tested the patch, but unfortunately, this patch is possible to break a rcar platform because a phy initialization is needed before the firmware loading if the platform uses the phy. (Note that upstream code (salvator-common.dtsi) doesn't use the phy for xhci. But, if we use the phy on other board with this patch, the xhci will not work.)

So, I think we need to add a new function pointer for your case.

Ok, thank you for testing! I will try to come up with other solution to mentioned mvebu-xhci issue.

Quoted 8 lines

Best regards,
Yoshihiro Shimoda

Quoted 5 lines
Quoted 5 lines

Their firmware would be loaded before phy parts are initialized, usb bus registered, or roothub device allocated.

Thanks
-Mathias

#11 lore raw
In reply to Pali Rohár

Hi Pali,

Quoted 72 lines

From: Pali Rohár, Sent: Monday, January 25, 2021 11:20 PM

Quoted 70 lines · On Friday 15 January 2021 15:32:30 Mathias Nyman wrote:
Quoted 69 lines · Pali Rohár
Quoted 49 lines · On Thursday 24 December 2020 05:59:05 Peter Chen wrote:
Quoted 48 lines · Pali Rohár

Currently init_quirk callbacks for xhci platform drivers are called xhci_plat_setup() function which is called after chip reset completes. It happens in the middle of the usb_add_hcd() function.

But XHCI_SKIP_PHY_INIT quirk is checked in the xhci_plat_probe() function prior calling usb_add_hcd() function. Therefore this XHCI_SKIP_PHY_INIT currently does nothing as prior xhci_plat_setup() it is not set.

Quirk XHCI_SKIP_PHY_INIT is only setting hcd->skip_phy_initialization value which really needs to be set prior calling usb_add_hcd() as this function at its beginning skips PHY init if this member is set.

This patch fixes implementation of the XHCI_SKIP_PHY_INIT quirk by calling init_quirk callbacks (via xhci_priv_init_quirk()) prior checking if XHCI_SKIP_PHY_INIT is set. Also checking if either xhci->quirks or priv->quirks contains this XHCI_SKIP_PHY_INIT quirk.

---
Changes in v2:
* Check also xhci->quirks as xhci_priv_init_quirk() callbacks are setting xhci->quirks
* Tested with "usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720" patch
* Removed Fixes: line

---
 drivers/usb/host/xhci-plat.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/usb/host/xhci-plat.c +0 -7
index 4d34f6005381..0eab7cb5a767 100644
@@ -89,13 +89,6 @@ static void xhci_plat_quirks(struct device *dev, struct xhci_hcd *xhci)
 /* called during probe() after chip reset completes */
 static int xhci_plat_setup(struct usb_hcd *hcd)
 {
-	int ret;
-
-
-	ret = xhci_priv_init_quirk(hcd);
-	if (ret)
-		return ret;
-
 	return xhci_gen_setup(hcd, xhci_plat_quirks);
 }
 
@@ -330,7 +323,14 @@ static int xhci_plat_probe(struct platform_device *pdev)
 
Quoted 15 lines

drivers/usb/host/xhci-plat.c static int xhci_plat_probe(struct platform_device *pdev) · line 334

Quoted 12 lines
drivers/usb/host/xhci-plat.c +8 -1
 	hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node);
 	xhci->shared_hcd->tpl_support = hcd->tpl_support;
-	if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT))
+
+	if (priv) {
+		ret = xhci_priv_init_quirk(hcd);
+		if (ret)
+			goto disable_usb_phy;
+	}
+
+	if ((xhci->quirks & XHCI_SKIP_PHY_INIT) || (priv && (priv->quirks & XHCI_SKIP_PHY_INIT)))
 		hcd->skip_phy_initialization = 1;

I am not sure if others agree with you move the position of xhci_priv_init_quirk, Let's see Mathias opinion.

Hello! Do you have an opinion how to handle this issue? As currently it is needed for another patch which is fixing issue/regression in xhci-mvebu:

<snip>

Quoted 10 lines
Quoted 10 lines
Quoted 1 line

I can see the benefit in this.
In the xhci-plat case usb_create_hcd and usb_add_hcd are separate steps, and we could both copy the xhci_plat_priv .quirks and run the .init_qurks before adding the hcd.
I guess the current way is inherited from pci case where the earliest place to do this after hcd is created is the hcd->driver->reset callback (which is set to xhci_pci_setup() or xhci_plat_setup()).

Quoted 7 lines
Quoted 2 lines

xhci-rcar.c is using the .init_quirk to load firmware, we need to check with them if this change is ok. (added Yoshihiro Shimoda to cc)

Yoshihiro, is this change OK?

Can we move forward? I really need to now how to handle regression in xhci-mvebu driver. And one option is with this patch...

Thank you for asking me about this topic. I tested the patch, but unfortunately, this patch is possible to break a rcar platform because a phy initialization is needed before the firmware loading if the platform uses the phy. (Note that upstream code (salvator-common.dtsi) doesn't use the phy for xhci. But, if we use the phy on other board with this patch, the xhci will not work.)

So, I think we need to add a new function pointer for your case.

Best regards,
Yoshihiro Shimoda

Quoted 5 lines
Quoted 5 lines

Their firmware would be loaded before phy parts are initialized, usb bus registered, or roothub device allocated.

Thanks
-Mathias

#13 lore raw
In reply to Pali Rohár
Quoted 24 lines · On Tuesday 26 January 2021 10:06:06 Pali Rohár wrote:
Quoted 23 lines · On Tuesday 26 January 2021 04:27:37 Yoshihiro Shimoda wrote:

Hi Pali,

Quoted 8 lines
Quoted 8 lines

I can see the benefit in this.
In the xhci-plat case usb_create_hcd and usb_add_hcd are separate steps, and we could both copy the xhci_plat_priv .quirks and run the .init_qurks before adding the hcd.
I guess the current way is inherited from pci case where the earliest place to do this after hcd is created is the hcd->driver->reset callback (which is set to xhci_pci_setup() or xhci_plat_setup()).

Quoted 7 lines
Quoted 2 lines

xhci-rcar.c is using the .init_quirk to load firmware, we need to check with them if this change is ok. (added Yoshihiro Shimoda to cc)

Yoshihiro, is this change OK?

Can we move forward? I really need to now how to handle regression in xhci-mvebu driver. And one option is with this patch...

Thank you for asking me about this topic. I tested the patch, but unfortunately, this patch is possible to break a rcar platform because a phy initialization is needed before the firmware loading if the platform uses the phy. (Note that upstream code (salvator-common.dtsi) doesn't use the phy for xhci. But, if we use the phy on other board with this patch, the xhci will not work.)

Quoted 4 lines
Quoted 1 line

So, I think we need to add a new function pointer for your case.

Ok, thank you for testing! I will try to come up with other solution to mentioned mvebu-xhci issue.

Hello! New version of this patch is in following thread, please review it:
"[PATCH v2] usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720"
https://lore.kernel.org/linux-usb/[email protected]/

Quoted 8 lines
Quoted 8 lines

Best regards,
Yoshihiro Shimoda

Quoted 5 lines
Quoted 5 lines

Their firmware would be loaded before phy parts are initialized, usb bus registered, or roothub device allocated.

Thanks
-Mathias

#14 lore raw
Quoted 26 lines

From: Pali Rohár, Sent: Tuesday, February 2, 2021 12:12 AM

Quoted 24 lines · On Tuesday 26 January 2021 10:06:06 Pali Rohár wrote:
Quoted 23 lines · On Tuesday 26 January 2021 04:27:37 Yoshihiro Shimoda wrote:

Hi Pali,

Quoted 8 lines
Quoted 8 lines

I can see the benefit in this.
In the xhci-plat case usb_create_hcd and usb_add_hcd are separate steps, and we could both copy the xhci_plat_priv .quirks and run the .init_qurks before adding the hcd.
I guess the current way is inherited from pci case where the earliest place to do this after hcd is created is the hcd->driver->reset callback (which is set to xhci_pci_setup() or xhci_plat_setup()).

Quoted 7 lines
Quoted 2 lines

xhci-rcar.c is using the .init_quirk to load firmware, we need to check with them if this change is ok. (added Yoshihiro Shimoda to cc)

Yoshihiro, is this change OK?

Can we move forward? I really need to now how to handle regression in xhci-mvebu driver. And one option is with this patch...

Thank you for asking me about this topic. I tested the patch, but unfortunately, this patch is possible to break a rcar platform because a phy initialization is needed before the firmware loading if the platform uses the phy. (Note that upstream code (salvator-common.dtsi) doesn't use the phy for xhci. But, if we use the phy on other board with this patch, the xhci will not work.)

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

So, I think we need to add a new function pointer for your case.

Ok, thank you for testing! I will try to come up with other solution to mentioned mvebu-xhci issue.

Hello! New version of this patch is in following thread, please review it:
"[PATCH v2] usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720"

Thanks! I'll review it.

Best regards,
Yoshihiro Shimoda