[PATCH v13 0/6] USB DWC3 host wake up support from system suspend

[PATCH v13 0/6] USB DWC3 host wake up support from system suspend

[Sandeep Maheswaram]

Avoiding phy powerdown in host mode when wakeup capable devices are 
connected, so that it can be wake up by devices.
Keep usb30_prim gdsc active to retain controller status
during suspend/resume.

Changes in v13:
Moved the dt bindings patch to start.
Changed dwc3_set_phy_speed_mode to dwc3_check_phy_speed_mode.
Check wakep-source property for dwc3 core node to set the
wakeup capability. Drop the device_init_wakeup call from
runtime suspend and resume.
Added GENPD_FLAG_RPM_ALWAYS_ON and set GENPD_FLAG_ALWAYS_ON if
wakeup is supported.

Changes in v12:
Squashed PATCH 1/5 and 2/5 of v11.
Added dt bindings and device tree entry for wakeup-source property
for dwc3 core node.
Dropped redundant phy_set_mode call.


Changes in v11:
Moving back to v8 version
https://patchwork.kernel.org/project/linux-arm-msm/cover/1624882097-23265-1-git-send-email-sanm@codeaurora.org
as we are getting interrupts during suspend
when enabling both DP hs phy irq and DM hs phy irq.
Moved the set phy mode function to dwc3/core.c from xhci-plat.c
We didn't find any other option other than accessing xhci from dwc.

Changes in v10:
PATCH 1/6: Change device_set_wakeup_capable to device_set_wakeup_enable
PATCH 2/6: Remove redundant else part in dwc3_resume_common
PATCH 4/6: Change the irg flags
PATCH 5/6: Set flag GENPD_FLAG_ALWAYS_ON
PATCH 6/6: Remove disable interrupts function and enable
interrupts in probe.


Changes in v9:
Checking with device_may_makeup property instead of phy_power_off flag.
Changed the IRQ flags and removed hs_phy_mode variable.

Changes in v8:
Moved the dwc3 suspend quirk code in dwc3/host.c to xhci-plat.c
Checking phy_power_off flag instead of usb_wakeup_enabled_descendants 
to keep gdsc active.

Changes in v7:
Change in commit text and message in PATCH 1/5 and PATCH 5/5
as per Matthias suggestion.
Added curly braces for if and else if sections in PATCH 4/5.

Changes in v6:
Addressed comments in host.c and core.c
Separated the patches in dwc3-qcom.c to make it simple.
Dropped wakeup-source change as it is not related to this series.

Changes in v5:
Added phy_power_off flag to check presence of wakeup capable devices.
Dropped patch[v4,4/5] as it is present linux-next.
Addressed comments in host.c and dwc3-qcom.c.

Changes in v4:
Addressed Matthias comments raised in v3.

Changes in v3:
Removed need_phy_for_wakeup flag and by default avoiding phy powerdown.
Addressed Matthias comments and added entry for DEV_SUPERSPEED.
Added suspend_quirk in dwc3 host and moved the dwc3_set_phy_speed_flags.
Added wakeup-source dt entry and reading in dwc-qcom.c glue driver.

Changes in v2:
Dropped the patch in clock to set GENPD_FLAG_ACTIVE_WAKEUP flag and 
setting in usb dwc3 driver.
Separated the core patch and glue driver patch.
Made need_phy_for_wakeup flag part of dwc structure and 
hs_phy_flags as unsgined int.
Adrressed the comment on device_init_wakeup call.
Corrected offset for reading portsc register.
Added pacth to support wakeup in xo shutdown case.

Sandeep Maheswaram (6):
  dt-bindings: usb: dwc3: Add wakeup-source property support
  usb: dwc3: core: Host wake up support from system suspend
  usb: dwc3: qcom: Add helper functions to enable,disable wake irqs
  usb: dwc3: qcom: Configure wakeup interrupts during suspend
  usb: dwc3: qcom: Keep power domain on to retain controller status
  arm64: dts: qcom: sc7280: Add wakeup-source property for USB node

 .../devicetree/bindings/usb/snps,dwc3.yaml         |  5 ++
 arch/arm64/boot/dts/qcom/sc7280.dtsi               |  1 +
 drivers/usb/dwc3/core.c                            | 33 +++++----
 drivers/usb/dwc3/core.h                            |  4 ++
 drivers/usb/dwc3/dwc3-qcom.c                       | 79 +++++++++++++---------
 drivers/usb/dwc3/host.c                            | 25 +++++++
 6 files changed, 103 insertions(+), 44 deletions(-)

-- 
2.7.4

[PATCH v13 1/6] dt-bindings: usb: dwc3: Add wakeup-source property support

[Sandeep Maheswaram]

Added support for wakeup-source property. This property can be
used to check and power down the phy during system suspend if
wake up is not supported.

Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
---
 Documentation/devicetree/bindings/usb/snps,dwc3.yaml | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
index f4471f8..4d4de2f 100644
--- a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
+++ b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
@@ -341,6 +341,11 @@ properties:
       This port is used with the 'usb-role-switch' property  to connect the
       dwc3 to type C connector.
 
+  wakeup-source:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description:
+      Enable USB remote wakeup.
+
 unevaluatedProperties: false
 
 required:
-- 
2.7.4

[PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Sandeep Maheswaram]

During suspend read the status of all port and set hs phy mode
based on current speed. Use this hs phy mode to configure wakeup
interrupts in qcom glue driver.

Check wakep-source property for dwc3 core node to set the
wakeup capability. Drop the device_init_wakeup call from
runtime suspend and resume.

Also check during suspend if any wakeup capable devices are
connected to the controller (directly or through hubs), if there
are none set a flag to indicate that the PHY is powered
down during suspend.

Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
---
 drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++-------------
 drivers/usb/dwc3/core.h |  4 ++++
 drivers/usb/dwc3/host.c | 25 +++++++++++++++++++++++++
 3 files changed, 49 insertions(+), 13 deletions(-)

diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
index 1170b80..effaa43 100644
--- a/drivers/usb/dwc3/core.c
+++ b/drivers/usb/dwc3/core.c
@@ -32,6 +32,7 @@
 #include <linux/usb/gadget.h>
 #include <linux/usb/of.h>
 #include <linux/usb/otg.h>
+#include <linux/usb/hcd.h>
 
 #include "core.h"
 #include "gadget.h"
@@ -1723,6 +1724,7 @@ static int dwc3_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, dwc);
 	dwc3_cache_hwparams(dwc);
+	device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
 
 	spin_lock_init(&dwc->lock);
 	mutex_init(&dwc->mutex);
@@ -1865,6 +1867,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
 {
 	unsigned long	flags;
 	u32 reg;
+	struct usb_hcd  *hcd = platform_get_drvdata(dwc->xhci);
 
 	switch (dwc->current_dr_role) {
 	case DWC3_GCTL_PRTCAP_DEVICE:
@@ -1877,10 +1880,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
 		dwc3_core_exit(dwc);
 		break;
 	case DWC3_GCTL_PRTCAP_HOST:
-		if (!PMSG_IS_AUTO(msg)) {
-			dwc3_core_exit(dwc);
-			break;
-		}
+		dwc3_check_phy_speed_mode(dwc);
 
 		/* Let controller to suspend HSPHY before PHY driver suspends */
 		if (dwc->dis_u2_susphy_quirk ||
@@ -1896,6 +1896,16 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
 
 		phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
 		phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
+
+		if (!PMSG_IS_AUTO(msg)) {
+			if (device_may_wakeup(dwc->dev) &&
+			    usb_wakeup_enabled_descendants(hcd->self.root_hub)) {
+				dwc->phy_power_off = false;
+			} else {
+				dwc->phy_power_off = true;
+				dwc3_core_exit(dwc);
+			}
+		}
 		break;
 	case DWC3_GCTL_PRTCAP_OTG:
 		/* do nothing during runtime_suspend */
@@ -1939,11 +1949,12 @@ static int dwc3_resume_common(struct dwc3 *dwc, pm_message_t msg)
 		break;
 	case DWC3_GCTL_PRTCAP_HOST:
 		if (!PMSG_IS_AUTO(msg)) {
-			ret = dwc3_core_init_for_resume(dwc);
-			if (ret)
-				return ret;
-			dwc3_set_prtcap(dwc, DWC3_GCTL_PRTCAP_HOST);
-			break;
+			if (dwc->phy_power_off) {
+				ret = dwc3_core_init_for_resume(dwc);
+				if (ret)
+					return ret;
+				dwc3_set_prtcap(dwc, DWC3_GCTL_PRTCAP_HOST);
+			}
 		}
 		/* Restore GUSB2PHYCFG bits that were modified in suspend */
 		reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
@@ -2015,8 +2026,6 @@ static int dwc3_runtime_suspend(struct device *dev)
 	if (ret)
 		return ret;
 
-	device_init_wakeup(dev, true);
-
 	return 0;
 }
 
@@ -2025,8 +2034,6 @@ static int dwc3_runtime_resume(struct device *dev)
 	struct dwc3     *dwc = dev_get_drvdata(dev);
 	int		ret;
 
-	device_init_wakeup(dev, false);
-
 	ret = dwc3_resume_common(dwc, PMSG_AUTO_RESUME);
 	if (ret)
 		return ret;
diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
index 5c9d467..6a5845f 100644
--- a/drivers/usb/dwc3/core.h
+++ b/drivers/usb/dwc3/core.h
@@ -1154,6 +1154,9 @@ struct dwc3 {
 
 	bool			phys_ready;
 
+	unsigned int            hs_phy_mode;
+	bool			phy_power_off;
+
 	struct ulpi		*ulpi;
 	bool			ulpi_ready;
 
@@ -1537,6 +1540,7 @@ int dwc3_core_soft_reset(struct dwc3 *dwc);
 #if IS_ENABLED(CONFIG_USB_DWC3_HOST) || IS_ENABLED(CONFIG_USB_DWC3_DUAL_ROLE)
 int dwc3_host_init(struct dwc3 *dwc);
 void dwc3_host_exit(struct dwc3 *dwc);
+void dwc3_check_phy_speed_mode(struct dwc3 *dwc);
 #else
 static inline int dwc3_host_init(struct dwc3 *dwc)
 { return 0; }
diff --git a/drivers/usb/dwc3/host.c b/drivers/usb/dwc3/host.c
index eda8719..7f22675 100644
--- a/drivers/usb/dwc3/host.c
+++ b/drivers/usb/dwc3/host.c
@@ -13,6 +13,7 @@
 #include <linux/platform_device.h>
 
 #include "core.h"
+#include "../host/xhci.h"
 
 static void dwc3_host_fill_xhci_irq_res(struct dwc3 *dwc,
 					int irq, char *name)
@@ -138,3 +139,27 @@ void dwc3_host_exit(struct dwc3 *dwc)
 {
 	platform_device_unregister(dwc->xhci);
 }
+
+void dwc3_check_phy_speed_mode(struct dwc3 *dwc)
+{
+
+	int i, num_ports;
+	u32 reg;
+	struct usb_hcd	*hcd = platform_get_drvdata(dwc->xhci);
+	struct xhci_hcd	*xhci_hcd = hcd_to_xhci(hcd);
+
+	dwc->hs_phy_mode = 0;
+
+	reg = readl(&xhci_hcd->cap_regs->hcs_params1);
+
+	num_ports = HCS_MAX_PORTS(reg);
+	for (i = 0; i < num_ports; i++) {
+		reg = readl(&xhci_hcd->op_regs->port_status_base + i * NUM_PORT_REGS);
+		if (reg & PORT_PE) {
+			if (DEV_HIGHSPEED(reg) || DEV_FULLSPEED(reg))
+				dwc->hs_phy_mode |= PHY_MODE_USB_HOST_HS;
+			else if (DEV_LOWSPEED(reg))
+				dwc->hs_phy_mode |= PHY_MODE_USB_HOST_LS;
+		}
+	}
+}
-- 
2.7.4

[PATCH v13 3/6] usb: dwc3: qcom: Add helper functions to enable,disable wake irqs

[Sandeep Maheswaram]

Adding helper functions to enable,disable wake irqs to make
the code simple and readable.

Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
---
 drivers/usb/dwc3/dwc3-qcom.c | 58 ++++++++++++++++++++------------------------
 1 file changed, 26 insertions(+), 32 deletions(-)

diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c
index 6cba990..7352124 100644
--- a/drivers/usb/dwc3/dwc3-qcom.c
+++ b/drivers/usb/dwc3/dwc3-qcom.c
@@ -296,50 +296,44 @@ static void dwc3_qcom_interconnect_exit(struct dwc3_qcom *qcom)
 	icc_put(qcom->icc_path_apps);
 }
 
+static void dwc3_qcom_enable_wakeup_irq(int irq)
+{
+	if (!irq)
+		return;
+
+	enable_irq(irq);
+	enable_irq_wake(irq);
+}
+
+static void dwc3_qcom_disable_wakeup_irq(int irq)
+{
+	if (!irq)
+		return;
+
+	disable_irq_wake(irq);
+	disable_irq_nosync(irq);
+}
+
 static void dwc3_qcom_disable_interrupts(struct dwc3_qcom *qcom)
 {
-	if (qcom->hs_phy_irq) {
-		disable_irq_wake(qcom->hs_phy_irq);
-		disable_irq_nosync(qcom->hs_phy_irq);
-	}
+	dwc3_qcom_disable_wakeup_irq(qcom->hs_phy_irq);
 
-	if (qcom->dp_hs_phy_irq) {
-		disable_irq_wake(qcom->dp_hs_phy_irq);
-		disable_irq_nosync(qcom->dp_hs_phy_irq);
-	}
+	dwc3_qcom_disable_wakeup_irq(qcom->dp_hs_phy_irq);
 
-	if (qcom->dm_hs_phy_irq) {
-		disable_irq_wake(qcom->dm_hs_phy_irq);
-		disable_irq_nosync(qcom->dm_hs_phy_irq);
-	}
+	dwc3_qcom_disable_wakeup_irq(qcom->dm_hs_phy_irq);
 
-	if (qcom->ss_phy_irq) {
-		disable_irq_wake(qcom->ss_phy_irq);
-		disable_irq_nosync(qcom->ss_phy_irq);
-	}
+	dwc3_qcom_disable_wakeup_irq(qcom->ss_phy_irq);
 }
 
 static void dwc3_qcom_enable_interrupts(struct dwc3_qcom *qcom)
 {
-	if (qcom->hs_phy_irq) {
-		enable_irq(qcom->hs_phy_irq);
-		enable_irq_wake(qcom->hs_phy_irq);
-	}
+	dwc3_qcom_enable_wakeup_irq(qcom->hs_phy_irq);
 
-	if (qcom->dp_hs_phy_irq) {
-		enable_irq(qcom->dp_hs_phy_irq);
-		enable_irq_wake(qcom->dp_hs_phy_irq);
-	}
+	dwc3_qcom_enable_wakeup_irq(qcom->dp_hs_phy_irq);
 
-	if (qcom->dm_hs_phy_irq) {
-		enable_irq(qcom->dm_hs_phy_irq);
-		enable_irq_wake(qcom->dm_hs_phy_irq);
-	}
+	dwc3_qcom_enable_wakeup_irq(qcom->dm_hs_phy_irq);
 
-	if (qcom->ss_phy_irq) {
-		enable_irq(qcom->ss_phy_irq);
-		enable_irq_wake(qcom->ss_phy_irq);
-	}
+	dwc3_qcom_enable_wakeup_irq(qcom->ss_phy_irq);
 }
 
 static int dwc3_qcom_suspend(struct dwc3_qcom *qcom)
-- 
2.7.4

[PATCH v13 4/6] usb: dwc3: qcom: Configure wakeup interrupts during suspend

[Sandeep Maheswaram]

Configure DP/DM interrupts to detect line state changes based on
hs_phy_mode. Enable the triggers opposite of what the current
DP, DM levels. For HS/FS mode enable DM interrupt and for LS enable DP
interrupt.

Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
---
 drivers/usb/dwc3/dwc3-qcom.c | 26 ++++++++++++++++++++------
 1 file changed, 20 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c
index 7352124..9804a19 100644
--- a/drivers/usb/dwc3/dwc3-qcom.c
+++ b/drivers/usb/dwc3/dwc3-qcom.c
@@ -316,22 +316,36 @@ static void dwc3_qcom_disable_wakeup_irq(int irq)
 
 static void dwc3_qcom_disable_interrupts(struct dwc3_qcom *qcom)
 {
-	dwc3_qcom_disable_wakeup_irq(qcom->hs_phy_irq);
+	struct dwc3 *dwc = platform_get_drvdata(qcom->dwc3);
 
-	dwc3_qcom_disable_wakeup_irq(qcom->dp_hs_phy_irq);
+	dwc3_qcom_disable_wakeup_irq(qcom->hs_phy_irq);
 
-	dwc3_qcom_disable_wakeup_irq(qcom->dm_hs_phy_irq);
+	if (dwc->hs_phy_mode & PHY_MODE_USB_HOST_LS) {
+		dwc3_qcom_disable_wakeup_irq(qcom->dp_hs_phy_irq);
+	} else if (dwc->hs_phy_mode & PHY_MODE_USB_HOST_HS) {
+		dwc3_qcom_disable_wakeup_irq(qcom->dm_hs_phy_irq);
+	} else {
+		dwc3_qcom_disable_wakeup_irq(qcom->dp_hs_phy_irq);
+		dwc3_qcom_disable_wakeup_irq(qcom->dm_hs_phy_irq);
+	}
 
 	dwc3_qcom_disable_wakeup_irq(qcom->ss_phy_irq);
 }
 
 static void dwc3_qcom_enable_interrupts(struct dwc3_qcom *qcom)
 {
-	dwc3_qcom_enable_wakeup_irq(qcom->hs_phy_irq);
+	struct dwc3 *dwc = platform_get_drvdata(qcom->dwc3);
 
-	dwc3_qcom_enable_wakeup_irq(qcom->dp_hs_phy_irq);
+	dwc3_qcom_enable_wakeup_irq(qcom->hs_phy_irq);
 
-	dwc3_qcom_enable_wakeup_irq(qcom->dm_hs_phy_irq);
+	if (dwc->hs_phy_mode & PHY_MODE_USB_HOST_LS) {
+		dwc3_qcom_enable_wakeup_irq(qcom->dp_hs_phy_irq);
+	} else if (dwc->hs_phy_mode & PHY_MODE_USB_HOST_HS) {
+		dwc3_qcom_enable_wakeup_irq(qcom->dm_hs_phy_irq);
+	} else {
+		dwc3_qcom_enable_wakeup_irq(qcom->dp_hs_phy_irq);
+		dwc3_qcom_enable_wakeup_irq(qcom->dm_hs_phy_irq);
+	}
 
 	dwc3_qcom_enable_wakeup_irq(qcom->ss_phy_irq);
 }
-- 
2.7.4

[PATCH v13 5/6] usb: dwc3: qcom: Keep power domain on to retain controller status

[Sandeep Maheswaram]

Keep the power domain always on during runtime suspend or if the
controller supports wakeup in order to retain controller status
and to support wakeup from devices.

Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
---
 drivers/usb/dwc3/dwc3-qcom.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c
index 9804a19..9747be6 100644
--- a/drivers/usb/dwc3/dwc3-qcom.c
+++ b/drivers/usb/dwc3/dwc3-qcom.c
@@ -17,6 +17,7 @@
 #include <linux/of_platform.h>
 #include <linux/platform_device.h>
 #include <linux/phy/phy.h>
+#include <linux/pm_domain.h>
 #include <linux/usb/of.h>
 #include <linux/reset.h>
 #include <linux/iopoll.h>
@@ -724,6 +725,7 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
 	struct resource		*res, *parent_res = NULL;
 	int			ret, i;
 	bool			ignore_pipe_clk;
+	struct generic_pm_domain *genpd;
 
 	qcom = devm_kzalloc(&pdev->dev, sizeof(*qcom), GFP_KERNEL);
 	if (!qcom)
@@ -732,6 +734,8 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, qcom);
 	qcom->dev = &pdev->dev;
 
+	genpd = pd_to_genpd(qcom->dev->pm_domain);
+
 	if (has_acpi_companion(dev)) {
 		qcom->acpi_pdata = acpi_device_get_match_data(dev);
 		if (!qcom->acpi_pdata) {
@@ -839,7 +843,12 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
 	if (ret)
 		goto interconnect_exit;
 
-	device_init_wakeup(&pdev->dev, 1);
+	genpd->flags |= GENPD_FLAG_RPM_ALWAYS_ON;
+
+	if (device_may_wakeup(&qcom->dwc3->dev))
+		genpd->flags |= GENPD_FLAG_ALWAYS_ON;
+
+	device_init_wakeup(&pdev->dev, device_may_wakeup(&qcom->dwc3->dev));
 	qcom->is_suspended = false;
 	pm_runtime_set_active(dev);
 	pm_runtime_enable(dev);
-- 
2.7.4

[PATCH v13 6/6] arm64: dts: qcom: sc7280: Add wakeup-source property for USB node

[Sandeep Maheswaram]

Adding wakeup-source property for USB controller in SC7280.
This property is added to inform that the USB controller is
wake up capable and to conditionally power down the phy during
system suspend.

Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
---
 arch/arm64/boot/dts/qcom/sc7280.dtsi | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm64/boot/dts/qcom/sc7280.dtsi b/arch/arm64/boot/dts/qcom/sc7280.dtsi
index f0b64be..046262a 100644
--- a/arch/arm64/boot/dts/qcom/sc7280.dtsi
+++ b/arch/arm64/boot/dts/qcom/sc7280.dtsi
@@ -2916,6 +2916,7 @@
 				phys = <&usb_1_hsphy>, <&usb_1_ssphy>;
 				phy-names = "usb2-phy", "usb3-phy";
 				maximum-speed = "super-speed";
+				wakeup-source;
 			};
 		};
 
-- 
2.7.4

Re: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Matthias Kaehlcke]

On Tue, Apr 12, 2022 at 12:46:50AM +0530, Sandeep Maheswaram wrote:
> During suspend read the status of all port and set hs phy mode
> based on current speed. Use this hs phy mode to configure wakeup
> interrupts in qcom glue driver.
> 
> Check wakep-source property for dwc3 core node to set the

s/wakep/wakeup/

> wakeup capability. Drop the device_init_wakeup call from
> runtime suspend and resume.
> 
> Also check during suspend if any wakeup capable devices are
> connected to the controller (directly or through hubs), if there
> are none set a flag to indicate that the PHY is powered
> down during suspend.
> 
> Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
> ---

A per-patch change log would be really helpful for reviewers, even
if it doesn't include older versions.

>  drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++-------------
>  drivers/usb/dwc3/core.h |  4 ++++
>  drivers/usb/dwc3/host.c | 25 +++++++++++++++++++++++++
>  3 files changed, 49 insertions(+), 13 deletions(-)
> 
> diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> index 1170b80..effaa43 100644
> --- a/drivers/usb/dwc3/core.c
> +++ b/drivers/usb/dwc3/core.c
> @@ -32,6 +32,7 @@
>  #include <linux/usb/gadget.h>
>  #include <linux/usb/of.h>
>  #include <linux/usb/otg.h>
> +#include <linux/usb/hcd.h>
>  
>  #include "core.h"
>  #include "gadget.h"
> @@ -1723,6 +1724,7 @@ static int dwc3_probe(struct platform_device *pdev)
>  
>  	platform_set_drvdata(pdev, dwc);
>  	dwc3_cache_hwparams(dwc);
> +	device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
>  
>  	spin_lock_init(&dwc->lock);
>  	mutex_init(&dwc->mutex);
> @@ -1865,6 +1867,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>  {
>  	unsigned long	flags;
>  	u32 reg;
> +	struct usb_hcd  *hcd = platform_get_drvdata(dwc->xhci);
>  
>  	switch (dwc->current_dr_role) {
>  	case DWC3_GCTL_PRTCAP_DEVICE:
> @@ -1877,10 +1880,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>  		dwc3_core_exit(dwc);
>  		break;
>  	case DWC3_GCTL_PRTCAP_HOST:
> -		if (!PMSG_IS_AUTO(msg)) {
> -			dwc3_core_exit(dwc);
> -			break;
> -		}
> +		dwc3_check_phy_speed_mode(dwc);
>  
>  		/* Let controller to suspend HSPHY before PHY driver suspends */
>  		if (dwc->dis_u2_susphy_quirk ||
> @@ -1896,6 +1896,16 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>  
>  		phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
>  		phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
> +
> +		if (!PMSG_IS_AUTO(msg)) {
> +			if (device_may_wakeup(dwc->dev) &&
> +			    usb_wakeup_enabled_descendants(hcd->self.root_hub)) {

You did not answer my question on v12, reposting it:

  Did you ever try whether you could use device_children_wakeup_capable() from
  [1] instead of usb_wakeup_enabled_descendants()?

  [1] https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065

> +				dwc->phy_power_off = false;
> +			} else {
> +				dwc->phy_power_off = true;
> +				dwc3_core_exit(dwc);
> +			}
> +		}
>  		break;
>  	case DWC3_GCTL_PRTCAP_OTG:
>  		/* do nothing during runtime_suspend */
> @@ -1939,11 +1949,12 @@ static int dwc3_resume_common(struct dwc3 *dwc, pm_message_t msg)
>  		break;
>  	case DWC3_GCTL_PRTCAP_HOST:
>  		if (!PMSG_IS_AUTO(msg)) {
> -			ret = dwc3_core_init_for_resume(dwc);
> -			if (ret)
> -				return ret;
> -			dwc3_set_prtcap(dwc, DWC3_GCTL_PRTCAP_HOST);
> -			break;
> +			if (dwc->phy_power_off) {
> +				ret = dwc3_core_init_for_resume(dwc);
> +				if (ret)
> +					return ret;
> +				dwc3_set_prtcap(dwc, DWC3_GCTL_PRTCAP_HOST);
> +			}
>  		}
>  		/* Restore GUSB2PHYCFG bits that were modified in suspend */
>  		reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
> @@ -2015,8 +2026,6 @@ static int dwc3_runtime_suspend(struct device *dev)
>  	if (ret)
>  		return ret;
>  
> -	device_init_wakeup(dev, true);
> -
>  	return 0;
>  }
>  
> @@ -2025,8 +2034,6 @@ static int dwc3_runtime_resume(struct device *dev)
>  	struct dwc3     *dwc = dev_get_drvdata(dev);
>  	int		ret;
>  
> -	device_init_wakeup(dev, false);
> -
>  	ret = dwc3_resume_common(dwc, PMSG_AUTO_RESUME);
>  	if (ret)
>  		return ret;
> diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
> index 5c9d467..6a5845f 100644
> --- a/drivers/usb/dwc3/core.h
> +++ b/drivers/usb/dwc3/core.h
> @@ -1154,6 +1154,9 @@ struct dwc3 {
>  
>  	bool			phys_ready;
>  
> +	unsigned int            hs_phy_mode;
> +	bool			phy_power_off;
> +
>  	struct ulpi		*ulpi;
>  	bool			ulpi_ready;
>  
> @@ -1537,6 +1540,7 @@ int dwc3_core_soft_reset(struct dwc3 *dwc);
>  #if IS_ENABLED(CONFIG_USB_DWC3_HOST) || IS_ENABLED(CONFIG_USB_DWC3_DUAL_ROLE)
>  int dwc3_host_init(struct dwc3 *dwc);
>  void dwc3_host_exit(struct dwc3 *dwc);
> +void dwc3_check_phy_speed_mode(struct dwc3 *dwc);
>  #else
>  static inline int dwc3_host_init(struct dwc3 *dwc)
>  { return 0; }
> diff --git a/drivers/usb/dwc3/host.c b/drivers/usb/dwc3/host.c
> index eda8719..7f22675 100644
> --- a/drivers/usb/dwc3/host.c
> +++ b/drivers/usb/dwc3/host.c
> @@ -13,6 +13,7 @@
>  #include <linux/platform_device.h>
>  
>  #include "core.h"
> +#include "../host/xhci.h"
>  
>  static void dwc3_host_fill_xhci_irq_res(struct dwc3 *dwc,
>  					int irq, char *name)
> @@ -138,3 +139,27 @@ void dwc3_host_exit(struct dwc3 *dwc)
>  {
>  	platform_device_unregister(dwc->xhci);
>  }
> +
> +void dwc3_check_phy_speed_mode(struct dwc3 *dwc)
> +{
> +

delete empty line

> +	int i, num_ports;
> +	u32 reg;
> +	struct usb_hcd	*hcd = platform_get_drvdata(dwc->xhci);
> +	struct xhci_hcd	*xhci_hcd = hcd_to_xhci(hcd);
> +
> +	dwc->hs_phy_mode = 0;
> +
> +	reg = readl(&xhci_hcd->cap_regs->hcs_params1);
> +
> +	num_ports = HCS_MAX_PORTS(reg);
> +	for (i = 0; i < num_ports; i++) {
> +		reg = readl(&xhci_hcd->op_regs->port_status_base + i * NUM_PORT_REGS);
> +		if (reg & PORT_PE) {
> +			if (DEV_HIGHSPEED(reg) || DEV_FULLSPEED(reg))
> +				dwc->hs_phy_mode |= PHY_MODE_USB_HOST_HS;
> +			else if (DEV_LOWSPEED(reg))
> +				dwc->hs_phy_mode |= PHY_MODE_USB_HOST_LS;
> +		}
> +	}
> +}
> -- 
> 2.7.4
> 

Re: [PATCH v13 5/6] usb: dwc3: qcom: Keep power domain on to retain controller status

[Matthias Kaehlcke]

On Tue, Apr 12, 2022 at 12:46:53AM +0530, Sandeep Maheswaram wrote:
> Keep the power domain always on during runtime suspend or if the
> controller supports wakeup in order to retain controller status
> and to support wakeup from devices.
> 
> Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
> ---
>  drivers/usb/dwc3/dwc3-qcom.c | 11 ++++++++++-
>  1 file changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c
> index 9804a19..9747be6 100644
> --- a/drivers/usb/dwc3/dwc3-qcom.c
> +++ b/drivers/usb/dwc3/dwc3-qcom.c
> @@ -17,6 +17,7 @@
>  #include <linux/of_platform.h>
>  #include <linux/platform_device.h>
>  #include <linux/phy/phy.h>
> +#include <linux/pm_domain.h>
>  #include <linux/usb/of.h>
>  #include <linux/reset.h>
>  #include <linux/iopoll.h>
> @@ -724,6 +725,7 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
>  	struct resource		*res, *parent_res = NULL;
>  	int			ret, i;
>  	bool			ignore_pipe_clk;
> +	struct generic_pm_domain *genpd;

nit: I'm not really a fan of gazillions of whitespaces between the type
and the variable name, but if they are kept all variable names above
should move a tab to the right. In any case the style in this file isn't
consistent, so an alternative would be to just get rid of the alignment
completely.

>  	qcom = devm_kzalloc(&pdev->dev, sizeof(*qcom), GFP_KERNEL);
>  	if (!qcom)
> @@ -732,6 +734,8 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
>  	platform_set_drvdata(pdev, qcom);
>  	qcom->dev = &pdev->dev;
>  
> +	genpd = pd_to_genpd(qcom->dev->pm_domain);
> +
>  	if (has_acpi_companion(dev)) {
>  		qcom->acpi_pdata = acpi_device_get_match_data(dev);
>  		if (!qcom->acpi_pdata) {
> @@ -839,7 +843,12 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
>  	if (ret)
>  		goto interconnect_exit;
>  
> -	device_init_wakeup(&pdev->dev, 1);
> +	genpd->flags |= GENPD_FLAG_RPM_ALWAYS_ON;
> +
> +	if (device_may_wakeup(&qcom->dwc3->dev))
> +		genpd->flags |= GENPD_FLAG_ALWAYS_ON;

In v12 only GENPD_FLAG_ALWAYS_ON was set, not GENPD_FLAG_RPM_ALWAYS_ON.
I asked a few times for a change log, in this instance it would be
useful to explain why GENPD_FLAG_RPM_ALWAYS_ON is now set.

> +
> +	device_init_wakeup(&pdev->dev, device_may_wakeup(&qcom->dwc3->dev));

device_may_wakeup() isn't an expensive operation, but it's still not nice to
call it twice in three lines of code. Instead you could do this:

	if (device_may_wakeup(&qcom->dwc3->dev)) {
		genpd->flags |= GENPD_FLAG_ALWAYS_ON;
		device_init_wakeup(&pdev->dev, true);
	}

Re: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Sandeep Maheswaram (Temp)]

Hi Matthias,

On 4/12/2022 2:24 AM, Matthias Kaehlcke wrote:
> On Tue, Apr 12, 2022 at 12:46:50AM +0530, Sandeep Maheswaram wrote:
>> During suspend read the status of all port and set hs phy mode
>> based on current speed. Use this hs phy mode to configure wakeup
>> interrupts in qcom glue driver.
>>
>> Check wakep-source property for dwc3 core node to set the
> s/wakep/wakeup/
Okay. Will update in next version.
>
>> wakeup capability. Drop the device_init_wakeup call from
>> runtime suspend and resume.
>>
>> Also check during suspend if any wakeup capable devices are
>> connected to the controller (directly or through hubs), if there
>> are none set a flag to indicate that the PHY is powered
>> down during suspend.
>>
>> Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
>> ---
> A per-patch change log would be really helpful for reviewers, even
> if it doesn't include older versions.
Okay. Will update in next version.
>
>>   drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++-------------
>>   drivers/usb/dwc3/core.h |  4 ++++
>>   drivers/usb/dwc3/host.c | 25 +++++++++++++++++++++++++
>>   3 files changed, 49 insertions(+), 13 deletions(-)
>>
>> diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
>> index 1170b80..effaa43 100644
>> --- a/drivers/usb/dwc3/core.c
>> +++ b/drivers/usb/dwc3/core.c
>> @@ -32,6 +32,7 @@
>>   #include <linux/usb/gadget.h>
>>   #include <linux/usb/of.h>
>>   #include <linux/usb/otg.h>
>> +#include <linux/usb/hcd.h>
>>   
>>   #include "core.h"
>>   #include "gadget.h"
>> @@ -1723,6 +1724,7 @@ static int dwc3_probe(struct platform_device *pdev)
>>   
>>   	platform_set_drvdata(pdev, dwc);
>>   	dwc3_cache_hwparams(dwc);
>> +	device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
>>   
>>   	spin_lock_init(&dwc->lock);
>>   	mutex_init(&dwc->mutex);
>> @@ -1865,6 +1867,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>   {
>>   	unsigned long	flags;
>>   	u32 reg;
>> +	struct usb_hcd  *hcd = platform_get_drvdata(dwc->xhci);
>>   
>>   	switch (dwc->current_dr_role) {
>>   	case DWC3_GCTL_PRTCAP_DEVICE:
>> @@ -1877,10 +1880,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>   		dwc3_core_exit(dwc);
>>   		break;
>>   	case DWC3_GCTL_PRTCAP_HOST:
>> -		if (!PMSG_IS_AUTO(msg)) {
>> -			dwc3_core_exit(dwc);
>> -			break;
>> -		}
>> +		dwc3_check_phy_speed_mode(dwc);
>>   
>>   		/* Let controller to suspend HSPHY before PHY driver suspends */
>>   		if (dwc->dis_u2_susphy_quirk ||
>> @@ -1896,6 +1896,16 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>   
>>   		phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
>>   		phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
>> +
>> +		if (!PMSG_IS_AUTO(msg)) {
>> +			if (device_may_wakeup(dwc->dev) &&
>> +			    usb_wakeup_enabled_descendants(hcd->self.root_hub)) {
> You did not answer my question on v12, reposting it:
>
>    Did you ever try whether you could use device_children_wakeup_capable() from
>    [1] instead of usb_wakeup_enabled_descendants()?
>
>    [1] https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065

Sorry ..I have replied in mail yesterday but it is not showing up in 
patchwork link.

Tried with  device_children_wakeup_capable(dwc->dev) instead of 
usb_wakeup_enabled_descendants and it always returns true even

when no devices are connected.

>
>> +				dwc->phy_power_off = false;
>> +			} else {
>> +				dwc->phy_power_off = true;
>> +				dwc3_core_exit(dwc);
>> +			}
>> +		}
>>   		break;
>>   	case DWC3_GCTL_PRTCAP_OTG:
>>   		/* do nothing during runtime_suspend */
>> @@ -1939,11 +1949,12 @@ static int dwc3_resume_common(struct dwc3 *dwc, pm_message_t msg)
>>   		break;
>>   	case DWC3_GCTL_PRTCAP_HOST:
>>   		if (!PMSG_IS_AUTO(msg)) {
>> -			ret = dwc3_core_init_for_resume(dwc);
>> -			if (ret)
>> -				return ret;
>> -			dwc3_set_prtcap(dwc, DWC3_GCTL_PRTCAP_HOST);
>> -			break;
>> +			if (dwc->phy_power_off) {
>> +				ret = dwc3_core_init_for_resume(dwc);
>> +				if (ret)
>> +					return ret;
>> +				dwc3_set_prtcap(dwc, DWC3_GCTL_PRTCAP_HOST);
>> +			}
>>   		}
>>   		/* Restore GUSB2PHYCFG bits that were modified in suspend */
>>   		reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
>> @@ -2015,8 +2026,6 @@ static int dwc3_runtime_suspend(struct device *dev)
>>   	if (ret)
>>   		return ret;
>>   
>> -	device_init_wakeup(dev, true);
>> -
>>   	return 0;
>>   }
>>   
>> @@ -2025,8 +2034,6 @@ static int dwc3_runtime_resume(struct device *dev)
>>   	struct dwc3     *dwc = dev_get_drvdata(dev);
>>   	int		ret;
>>   
>> -	device_init_wakeup(dev, false);
>> -
>>   	ret = dwc3_resume_common(dwc, PMSG_AUTO_RESUME);
>>   	if (ret)
>>   		return ret;
>> diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
>> index 5c9d467..6a5845f 100644
>> --- a/drivers/usb/dwc3/core.h
>> +++ b/drivers/usb/dwc3/core.h
>> @@ -1154,6 +1154,9 @@ struct dwc3 {
>>   
>>   	bool			phys_ready;
>>   
>> +	unsigned int            hs_phy_mode;
>> +	bool			phy_power_off;
>> +
>>   	struct ulpi		*ulpi;
>>   	bool			ulpi_ready;
>>   
>> @@ -1537,6 +1540,7 @@ int dwc3_core_soft_reset(struct dwc3 *dwc);
>>   #if IS_ENABLED(CONFIG_USB_DWC3_HOST) || IS_ENABLED(CONFIG_USB_DWC3_DUAL_ROLE)
>>   int dwc3_host_init(struct dwc3 *dwc);
>>   void dwc3_host_exit(struct dwc3 *dwc);
>> +void dwc3_check_phy_speed_mode(struct dwc3 *dwc);
>>   #else
>>   static inline int dwc3_host_init(struct dwc3 *dwc)
>>   { return 0; }
>> diff --git a/drivers/usb/dwc3/host.c b/drivers/usb/dwc3/host.c
>> index eda8719..7f22675 100644
>> --- a/drivers/usb/dwc3/host.c
>> +++ b/drivers/usb/dwc3/host.c
>> @@ -13,6 +13,7 @@
>>   #include <linux/platform_device.h>
>>   
>>   #include "core.h"
>> +#include "../host/xhci.h"
>>   
>>   static void dwc3_host_fill_xhci_irq_res(struct dwc3 *dwc,
>>   					int irq, char *name)
>> @@ -138,3 +139,27 @@ void dwc3_host_exit(struct dwc3 *dwc)
>>   {
>>   	platform_device_unregister(dwc->xhci);
>>   }
>> +
>> +void dwc3_check_phy_speed_mode(struct dwc3 *dwc)
>> +{
>> +
> delete empty line
Okay. Will update in next version.
>
>> +	int i, num_ports;
>> +	u32 reg;
>> +	struct usb_hcd	*hcd = platform_get_drvdata(dwc->xhci);
>> +	struct xhci_hcd	*xhci_hcd = hcd_to_xhci(hcd);
>> +
>> +	dwc->hs_phy_mode = 0;
>> +
>> +	reg = readl(&xhci_hcd->cap_regs->hcs_params1);
>> +
>> +	num_ports = HCS_MAX_PORTS(reg);
>> +	for (i = 0; i < num_ports; i++) {
>> +		reg = readl(&xhci_hcd->op_regs->port_status_base + i * NUM_PORT_REGS);
>> +		if (reg & PORT_PE) {
>> +			if (DEV_HIGHSPEED(reg) || DEV_FULLSPEED(reg))
>> +				dwc->hs_phy_mode |= PHY_MODE_USB_HOST_HS;
>> +			else if (DEV_LOWSPEED(reg))
>> +				dwc->hs_phy_mode |= PHY_MODE_USB_HOST_LS;
>> +		}
>> +	}
>> +}
>> -- 
>> 2.7.4
>>

Re: [PATCH v13 5/6] usb: dwc3: qcom: Keep power domain on to retain controller status

[Sandeep Maheswaram (Temp)]

Hi,

On 4/12/2022 2:43 AM, Matthias Kaehlcke wrote:
> On Tue, Apr 12, 2022 at 12:46:53AM +0530, Sandeep Maheswaram wrote:
>> Keep the power domain always on during runtime suspend or if the
>> controller supports wakeup in order to retain controller status
>> and to support wakeup from devices.
>>
>> Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
>> ---
>>   drivers/usb/dwc3/dwc3-qcom.c | 11 ++++++++++-
>>   1 file changed, 10 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c
>> index 9804a19..9747be6 100644
>> --- a/drivers/usb/dwc3/dwc3-qcom.c
>> +++ b/drivers/usb/dwc3/dwc3-qcom.c
>> @@ -17,6 +17,7 @@
>>   #include <linux/of_platform.h>
>>   #include <linux/platform_device.h>
>>   #include <linux/phy/phy.h>
>> +#include <linux/pm_domain.h>
>>   #include <linux/usb/of.h>
>>   #include <linux/reset.h>
>>   #include <linux/iopoll.h>
>> @@ -724,6 +725,7 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
>>   	struct resource		*res, *parent_res = NULL;
>>   	int			ret, i;
>>   	bool			ignore_pipe_clk;
>> +	struct generic_pm_domain *genpd;
> nit: I'm not really a fan of gazillions of whitespaces between the type
> and the variable name, but if they are kept all variable names above
> should move a tab to the right. In any case the style in this file isn't
> consistent, so an alternative would be to just get rid of the alignment
> completely.
Okay. Will do it in next version.
>
>>   	qcom = devm_kzalloc(&pdev->dev, sizeof(*qcom), GFP_KERNEL);
>>   	if (!qcom)
>> @@ -732,6 +734,8 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
>>   	platform_set_drvdata(pdev, qcom);
>>   	qcom->dev = &pdev->dev;
>>   
>> +	genpd = pd_to_genpd(qcom->dev->pm_domain);
>> +
>>   	if (has_acpi_companion(dev)) {
>>   		qcom->acpi_pdata = acpi_device_get_match_data(dev);
>>   		if (!qcom->acpi_pdata) {
>> @@ -839,7 +843,12 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
>>   	if (ret)
>>   		goto interconnect_exit;
>>   
>> -	device_init_wakeup(&pdev->dev, 1);
>> +	genpd->flags |= GENPD_FLAG_RPM_ALWAYS_ON;
>> +
>> +	if (device_may_wakeup(&qcom->dwc3->dev))
>> +		genpd->flags |= GENPD_FLAG_ALWAYS_ON;
> In v12 only GENPD_FLAG_ALWAYS_ON was set, not GENPD_FLAG_RPM_ALWAYS_ON.
> I asked a few times for a change log, in this instance it would be
> useful to explain why GENPD_FLAG_RPM_ALWAYS_ON is now set.
Okay. Will add it in next version.
>
>> +
>> +	device_init_wakeup(&pdev->dev, device_may_wakeup(&qcom->dwc3->dev));
> device_may_wakeup() isn't an expensive operation, but it's still not nice to
> call it twice in three lines of code. Instead you could do this:
>
> 	if (device_may_wakeup(&qcom->dwc3->dev)) {
> 		genpd->flags |= GENPD_FLAG_ALWAYS_ON;
> 		device_init_wakeup(&pdev->dev, true);
> 	}
Okay. Will do it in next version.

Re: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Pavan Kondeti]

Hi Sandeep,

On Tue, Apr 12, 2022 at 10:16:39AM +0530, Sandeep Maheswaram (Temp) wrote:
> Hi Matthias,
> 
> On 4/12/2022 2:24 AM, Matthias Kaehlcke wrote:
> >On Tue, Apr 12, 2022 at 12:46:50AM +0530, Sandeep Maheswaram wrote:
> >>During suspend read the status of all port and set hs phy mode
> >>based on current speed. Use this hs phy mode to configure wakeup
> >>interrupts in qcom glue driver.
> >>
> >>Check wakep-source property for dwc3 core node to set the
> >s/wakep/wakeup/
> Okay. Will update in next version.
> >
> >>wakeup capability. Drop the device_init_wakeup call from
> >>runtime suspend and resume.
> >>
> >>Also check during suspend if any wakeup capable devices are
> >>connected to the controller (directly or through hubs), if there
> >>are none set a flag to indicate that the PHY is powered
> >>down during suspend.
> >>
> >>Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
> >>---
> >A per-patch change log would be really helpful for reviewers, even
> >if it doesn't include older versions.
> Okay. Will update in next version.
> >
> >>  drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++-------------
> >>  drivers/usb/dwc3/core.h |  4 ++++
> >>  drivers/usb/dwc3/host.c | 25 +++++++++++++++++++++++++
> >>  3 files changed, 49 insertions(+), 13 deletions(-)
> >>
> >>diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> >>index 1170b80..effaa43 100644
> >>--- a/drivers/usb/dwc3/core.c
> >>+++ b/drivers/usb/dwc3/core.c
> >>@@ -32,6 +32,7 @@
> >>  #include <linux/usb/gadget.h>
> >>  #include <linux/usb/of.h>
> >>  #include <linux/usb/otg.h>
> >>+#include <linux/usb/hcd.h>
> >>  #include "core.h"
> >>  #include "gadget.h"
> >>@@ -1723,6 +1724,7 @@ static int dwc3_probe(struct platform_device *pdev)
> >>  	platform_set_drvdata(pdev, dwc);
> >>  	dwc3_cache_hwparams(dwc);
> >>+	device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
> >>  	spin_lock_init(&dwc->lock);
> >>  	mutex_init(&dwc->mutex);
> >>@@ -1865,6 +1867,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> >>  {
> >>  	unsigned long	flags;
> >>  	u32 reg;
> >>+	struct usb_hcd  *hcd = platform_get_drvdata(dwc->xhci);
> >>  	switch (dwc->current_dr_role) {
> >>  	case DWC3_GCTL_PRTCAP_DEVICE:
> >>@@ -1877,10 +1880,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> >>  		dwc3_core_exit(dwc);
> >>  		break;
> >>  	case DWC3_GCTL_PRTCAP_HOST:
> >>-		if (!PMSG_IS_AUTO(msg)) {
> >>-			dwc3_core_exit(dwc);
> >>-			break;
> >>-		}
> >>+		dwc3_check_phy_speed_mode(dwc);
> >>  		/* Let controller to suspend HSPHY before PHY driver suspends */
> >>  		if (dwc->dis_u2_susphy_quirk ||
> >>@@ -1896,6 +1896,16 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> >>  		phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
> >>  		phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
> >>+
> >>+		if (!PMSG_IS_AUTO(msg)) {
> >>+			if (device_may_wakeup(dwc->dev) &&
> >>+			    usb_wakeup_enabled_descendants(hcd->self.root_hub)) {
> >You did not answer my question on v12, reposting it:
> >
> >   Did you ever try whether you could use device_children_wakeup_capable() from
> >   [1] instead of usb_wakeup_enabled_descendants()?
> >
> >   [1] https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
> 
> Sorry ..I have replied in mail yesterday but it is not showing up in
> patchwork link.
> 
> Tried with  device_children_wakeup_capable(dwc->dev) instead of
> usb_wakeup_enabled_descendants and it always returns true even
> 
> when no devices are connected.
> 

What do you mean by when no devices are connected? There is always
root hub connected and we should not power down the DWC3 here even
when remote wakeup for root hub is enabled. Essentially
usb_wakeup_enabled_descendants() returns true even without any
physical devices connected.

What does device_children_wakeup_capable() do? Sorry, I could not
find this function definition.

Thanks,
Pavan

Re: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Sandeep Maheswaram (Temp)]

Hi Pavan,

On 4/12/2022 10:30 AM, Pavan Kondeti wrote:
> Hi Sandeep,
>
> On Tue, Apr 12, 2022 at 10:16:39AM +0530, Sandeep Maheswaram (Temp) wrote:
>> Hi Matthias,
>>
>> On 4/12/2022 2:24 AM, Matthias Kaehlcke wrote:
>>> On Tue, Apr 12, 2022 at 12:46:50AM +0530, Sandeep Maheswaram wrote:
>>>> During suspend read the status of all port and set hs phy mode
>>>> based on current speed. Use this hs phy mode to configure wakeup
>>>> interrupts in qcom glue driver.
>>>>
>>>> Check wakep-source property for dwc3 core node to set the
>>> s/wakep/wakeup/
>> Okay. Will update in next version.
>>>> wakeup capability. Drop the device_init_wakeup call from
>>>> runtime suspend and resume.
>>>>
>>>> Also check during suspend if any wakeup capable devices are
>>>> connected to the controller (directly or through hubs), if there
>>>> are none set a flag to indicate that the PHY is powered
>>>> down during suspend.
>>>>
>>>> Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
>>>> ---
>>> A per-patch change log would be really helpful for reviewers, even
>>> if it doesn't include older versions.
>> Okay. Will update in next version.
>>>>   drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++-------------
>>>>   drivers/usb/dwc3/core.h |  4 ++++
>>>>   drivers/usb/dwc3/host.c | 25 +++++++++++++++++++++++++
>>>>   3 files changed, 49 insertions(+), 13 deletions(-)
>>>>
>>>> diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
>>>> index 1170b80..effaa43 100644
>>>> --- a/drivers/usb/dwc3/core.c
>>>> +++ b/drivers/usb/dwc3/core.c
>>>> @@ -32,6 +32,7 @@
>>>>   #include <linux/usb/gadget.h>
>>>>   #include <linux/usb/of.h>
>>>>   #include <linux/usb/otg.h>
>>>> +#include <linux/usb/hcd.h>
>>>>   #include "core.h"
>>>>   #include "gadget.h"
>>>> @@ -1723,6 +1724,7 @@ static int dwc3_probe(struct platform_device *pdev)
>>>>   	platform_set_drvdata(pdev, dwc);
>>>>   	dwc3_cache_hwparams(dwc);
>>>> +	device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
>>>>   	spin_lock_init(&dwc->lock);
>>>>   	mutex_init(&dwc->mutex);
>>>> @@ -1865,6 +1867,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>>>   {
>>>>   	unsigned long	flags;
>>>>   	u32 reg;
>>>> +	struct usb_hcd  *hcd = platform_get_drvdata(dwc->xhci);
>>>>   	switch (dwc->current_dr_role) {
>>>>   	case DWC3_GCTL_PRTCAP_DEVICE:
>>>> @@ -1877,10 +1880,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>>>   		dwc3_core_exit(dwc);
>>>>   		break;
>>>>   	case DWC3_GCTL_PRTCAP_HOST:
>>>> -		if (!PMSG_IS_AUTO(msg)) {
>>>> -			dwc3_core_exit(dwc);
>>>> -			break;
>>>> -		}
>>>> +		dwc3_check_phy_speed_mode(dwc);
>>>>   		/* Let controller to suspend HSPHY before PHY driver suspends */
>>>>   		if (dwc->dis_u2_susphy_quirk ||
>>>> @@ -1896,6 +1896,16 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>>>   		phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
>>>>   		phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
>>>> +
>>>> +		if (!PMSG_IS_AUTO(msg)) {
>>>> +			if (device_may_wakeup(dwc->dev) &&
>>>> +			    usb_wakeup_enabled_descendants(hcd->self.root_hub)) {
>>> You did not answer my question on v12, reposting it:
>>>
>>>    Did you ever try whether you could use device_children_wakeup_capable() from
>>>    [1] instead of usb_wakeup_enabled_descendants()?
>>>
>>>    [1] https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
>> Sorry ..I have replied in mail yesterday but it is not showing up in
>> patchwork link.
>>
>> Tried with  device_children_wakeup_capable(dwc->dev) instead of
>> usb_wakeup_enabled_descendants and it always returns true even
>>
>> when no devices are connected.
>>
> What do you mean by when no devices are connected? There is always
> root hub connected and we should not power down the DWC3 here even
> when remote wakeup for root hub is enabled. Essentially
> usb_wakeup_enabled_descendants() returns true even without any
> physical devices connected.
>
> What does device_children_wakeup_capable() do? Sorry, I could not
> find this function definition.
>
> Thanks,
> Pavan

usb_wakeup_enabled_descendants() doesn't consider hubs. It only returns true if any devices
are connected with wakeup capability apart from hubs. If we consider hubs also dwc3 core exit
and phy exit will never be called.

device_children_wakeup_capable() implementation was shared by Matthias in below thread
https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065

Probably device_children_wakeup_capable() is returning true because it considers hubs also.

Re: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Pavan Kondeti]

Hi Sandeep,

On Tue, Apr 12, 2022 at 12:08:02PM +0530, Sandeep Maheswaram (Temp) wrote:
> Hi Pavan,
> 
> On 4/12/2022 10:30 AM, Pavan Kondeti wrote:
> >Hi Sandeep,
> >
> >On Tue, Apr 12, 2022 at 10:16:39AM +0530, Sandeep Maheswaram (Temp) wrote:
> >>Hi Matthias,
> >>
> >>On 4/12/2022 2:24 AM, Matthias Kaehlcke wrote:
> >>>On Tue, Apr 12, 2022 at 12:46:50AM +0530, Sandeep Maheswaram wrote:
> >>>>During suspend read the status of all port and set hs phy mode
> >>>>based on current speed. Use this hs phy mode to configure wakeup
> >>>>interrupts in qcom glue driver.
> >>>>
> >>>>Check wakep-source property for dwc3 core node to set the
> >>>s/wakep/wakeup/
> >>Okay. Will update in next version.
> >>>>wakeup capability. Drop the device_init_wakeup call from
> >>>>runtime suspend and resume.
> >>>>
> >>>>Also check during suspend if any wakeup capable devices are
> >>>>connected to the controller (directly or through hubs), if there
> >>>>are none set a flag to indicate that the PHY is powered
> >>>>down during suspend.
> >>>>
> >>>>Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
> >>>>---
> >>>A per-patch change log would be really helpful for reviewers, even
> >>>if it doesn't include older versions.
> >>Okay. Will update in next version.
> >>>>  drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++-------------
> >>>>  drivers/usb/dwc3/core.h |  4 ++++
> >>>>  drivers/usb/dwc3/host.c | 25 +++++++++++++++++++++++++
> >>>>  3 files changed, 49 insertions(+), 13 deletions(-)
> >>>>
> >>>>diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> >>>>index 1170b80..effaa43 100644
> >>>>--- a/drivers/usb/dwc3/core.c
> >>>>+++ b/drivers/usb/dwc3/core.c
> >>>>@@ -32,6 +32,7 @@
> >>>>  #include <linux/usb/gadget.h>
> >>>>  #include <linux/usb/of.h>
> >>>>  #include <linux/usb/otg.h>
> >>>>+#include <linux/usb/hcd.h>
> >>>>  #include "core.h"
> >>>>  #include "gadget.h"
> >>>>@@ -1723,6 +1724,7 @@ static int dwc3_probe(struct platform_device *pdev)
> >>>>  	platform_set_drvdata(pdev, dwc);
> >>>>  	dwc3_cache_hwparams(dwc);
> >>>>+	device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
> >>>>  	spin_lock_init(&dwc->lock);
> >>>>  	mutex_init(&dwc->mutex);
> >>>>@@ -1865,6 +1867,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> >>>>  {
> >>>>  	unsigned long	flags;
> >>>>  	u32 reg;
> >>>>+	struct usb_hcd  *hcd = platform_get_drvdata(dwc->xhci);
> >>>>  	switch (dwc->current_dr_role) {
> >>>>  	case DWC3_GCTL_PRTCAP_DEVICE:
> >>>>@@ -1877,10 +1880,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> >>>>  		dwc3_core_exit(dwc);
> >>>>  		break;
> >>>>  	case DWC3_GCTL_PRTCAP_HOST:
> >>>>-		if (!PMSG_IS_AUTO(msg)) {
> >>>>-			dwc3_core_exit(dwc);
> >>>>-			break;
> >>>>-		}
> >>>>+		dwc3_check_phy_speed_mode(dwc);
> >>>>  		/* Let controller to suspend HSPHY before PHY driver suspends */
> >>>>  		if (dwc->dis_u2_susphy_quirk ||
> >>>>@@ -1896,6 +1896,16 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> >>>>  		phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
> >>>>  		phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
> >>>>+
> >>>>+		if (!PMSG_IS_AUTO(msg)) {
> >>>>+			if (device_may_wakeup(dwc->dev) &&
> >>>>+			    usb_wakeup_enabled_descendants(hcd->self.root_hub)) {
> >>>You did not answer my question on v12, reposting it:
> >>>
> >>>   Did you ever try whether you could use device_children_wakeup_capable() from
> >>>   [1] instead of usb_wakeup_enabled_descendants()?
> >>>
> >>>   [1] https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
> >>Sorry ..I have replied in mail yesterday but it is not showing up in
> >>patchwork link.
> >>
> >>Tried with  device_children_wakeup_capable(dwc->dev) instead of
> >>usb_wakeup_enabled_descendants and it always returns true even
> >>
> >>when no devices are connected.
> >>
> >What do you mean by when no devices are connected? There is always
> >root hub connected and we should not power down the DWC3 here even
> >when remote wakeup for root hub is enabled. Essentially
> >usb_wakeup_enabled_descendants() returns true even without any
> >physical devices connected.
> >
> >What does device_children_wakeup_capable() do? Sorry, I could not
> >find this function definition.
> >
> >Thanks,
> >Pavan
> 
> usb_wakeup_enabled_descendants() doesn't consider hubs. It only returns true if any devices
> are connected with wakeup capability apart from hubs. If we consider hubs also dwc3 core exit
> and phy exit will never be called.
> 

copy/pasting the code here. you are passing roothub as udev. As you can see
this function clearly takes the remote wakeup of root hub into account. Lets
take a simple case, we have a roothub with one port. nothing attached. so
we enter runtime suspend. Don't you want to keep PHY ON? otherwise, how can
you detect if a keyboard/mouse attached directly to root hub? 

In your configuration, is remote wakeup enabled for the roothub or not?

/* Count of wakeup-enabled devices at or below udev */
unsigned usb_wakeup_enabled_descendants(struct usb_device *udev)
{
	struct usb_hub *hub = usb_hub_to_struct_hub(udev);

	return udev->do_remote_wakeup +
			(hub ? hub->wakeup_enabled_descendants : 0);
}
> device_children_wakeup_capable() implementation was shared by Matthias in below thread
> https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
> 
> Probably device_children_wakeup_capable() is returning true because it considers hubs also.
> 

if you see choose_wakeup(), it derives the USB remote wakeup capability from
the device_may_wakeup(), so the patch suggested might work. Note that we have
to take wakeup capability of roothub into account even when there is no
USB device attached.

Thanks,
Pavan

Re: [PATCH v13 1/6] dt-bindings: usb: dwc3: Add wakeup-source property support

[Krzysztof Kozlowski]

On 11/04/2022 21:16, Sandeep Maheswaram wrote:
> Added support for wakeup-source property. This property can be
> used to check and power down the phy during system suspend if
> wake up is not supported.
> 
> Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
> Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
> ---
>  Documentation/devicetree/bindings/usb/snps,dwc3.yaml | 5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> index f4471f8..4d4de2f 100644
> --- a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> +++ b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml
> @@ -341,6 +341,11 @@ properties:
>        This port is used with the 'usb-role-switch' property  to connect the
>        dwc3 to type C connector.
>  
> +  wakeup-source:
> +    $ref: /schemas/types.yaml#/definitions/flag

I would prefer to use shorter "type: boolean", but in any case:

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>

> +    description:
> +      Enable USB remote wakeup.


Best regards,
Krzysztof

Re: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Matthias Kaehlcke]

On Tue, Apr 12, 2022 at 12:08:02PM +0530, Sandeep Maheswaram (Temp) wrote:
> Hi Pavan,
> 
> On 4/12/2022 10:30 AM, Pavan Kondeti wrote:
> > Hi Sandeep,
> > 
> > On Tue, Apr 12, 2022 at 10:16:39AM +0530, Sandeep Maheswaram (Temp) wrote:
> > > Hi Matthias,
> > > 
> > > On 4/12/2022 2:24 AM, Matthias Kaehlcke wrote:
> > > > On Tue, Apr 12, 2022 at 12:46:50AM +0530, Sandeep Maheswaram wrote:
> > > > > During suspend read the status of all port and set hs phy mode
> > > > > based on current speed. Use this hs phy mode to configure wakeup
> > > > > interrupts in qcom glue driver.
> > > > > 
> > > > > Check wakep-source property for dwc3 core node to set the
> > > > s/wakep/wakeup/
> > > Okay. Will update in next version.
> > > > > wakeup capability. Drop the device_init_wakeup call from
> > > > > runtime suspend and resume.
> > > > > 
> > > > > Also check during suspend if any wakeup capable devices are
> > > > > connected to the controller (directly or through hubs), if there
> > > > > are none set a flag to indicate that the PHY is powered
> > > > > down during suspend.
> > > > > 
> > > > > Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
> > > > > ---
> > > > A per-patch change log would be really helpful for reviewers, even
> > > > if it doesn't include older versions.
> > > Okay. Will update in next version.
> > > > >   drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++-------------
> > > > >   drivers/usb/dwc3/core.h |  4 ++++
> > > > >   drivers/usb/dwc3/host.c | 25 +++++++++++++++++++++++++
> > > > >   3 files changed, 49 insertions(+), 13 deletions(-)
> > > > > 
> > > > > diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> > > > > index 1170b80..effaa43 100644
> > > > > --- a/drivers/usb/dwc3/core.c
> > > > > +++ b/drivers/usb/dwc3/core.c
> > > > > @@ -32,6 +32,7 @@
> > > > >   #include <linux/usb/gadget.h>
> > > > >   #include <linux/usb/of.h>
> > > > >   #include <linux/usb/otg.h>
> > > > > +#include <linux/usb/hcd.h>
> > > > >   #include "core.h"
> > > > >   #include "gadget.h"
> > > > > @@ -1723,6 +1724,7 @@ static int dwc3_probe(struct platform_device *pdev)
> > > > >   	platform_set_drvdata(pdev, dwc);
> > > > >   	dwc3_cache_hwparams(dwc);
> > > > > +	device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
> > > > >   	spin_lock_init(&dwc->lock);
> > > > >   	mutex_init(&dwc->mutex);
> > > > > @@ -1865,6 +1867,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > > > >   {
> > > > >   	unsigned long	flags;
> > > > >   	u32 reg;
> > > > > +	struct usb_hcd  *hcd = platform_get_drvdata(dwc->xhci);
> > > > >   	switch (dwc->current_dr_role) {
> > > > >   	case DWC3_GCTL_PRTCAP_DEVICE:
> > > > > @@ -1877,10 +1880,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > > > >   		dwc3_core_exit(dwc);
> > > > >   		break;
> > > > >   	case DWC3_GCTL_PRTCAP_HOST:
> > > > > -		if (!PMSG_IS_AUTO(msg)) {
> > > > > -			dwc3_core_exit(dwc);
> > > > > -			break;
> > > > > -		}
> > > > > +		dwc3_check_phy_speed_mode(dwc);
> > > > >   		/* Let controller to suspend HSPHY before PHY driver suspends */
> > > > >   		if (dwc->dis_u2_susphy_quirk ||
> > > > > @@ -1896,6 +1896,16 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > > > >   		phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
> > > > >   		phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
> > > > > +
> > > > > +		if (!PMSG_IS_AUTO(msg)) {
> > > > > +			if (device_may_wakeup(dwc->dev) &&
> > > > > +			    usb_wakeup_enabled_descendants(hcd->self.root_hub)) {
> > > > You did not answer my question on v12, reposting it:
> > > > 
> > > >    Did you ever try whether you could use device_children_wakeup_capable() from
> > > >    [1] instead of usb_wakeup_enabled_descendants()?
> > > > 
> > > >    [1] https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
> > > Sorry ..I have replied in mail yesterday but it is not showing up in
> > > patchwork link.
> > > 
> > > Tried with  device_children_wakeup_capable(dwc->dev) instead of
> > > usb_wakeup_enabled_descendants and it always returns true even
> > > 
> > > when no devices are connected.
> > > 
> > What do you mean by when no devices are connected? There is always
> > root hub connected and we should not power down the DWC3 here even
> > when remote wakeup for root hub is enabled. Essentially
> > usb_wakeup_enabled_descendants() returns true even without any
> > physical devices connected.
> > 
> > What does device_children_wakeup_capable() do? Sorry, I could not
> > find this function definition.
> > 
> > Thanks,
> > Pavan
> 
> usb_wakeup_enabled_descendants() doesn't consider hubs. It only returns true if any devices
> are connected with wakeup capability apart from hubs.

Actually it considers hubs:

unsigned usb_wakeup_enabled_descendants(struct usb_device *udev)
{
	struct usb_hub *hub = usb_hub_to_struct_hub(udev);

	return udev->do_remote_wakeup +
		(hub ? hub->wakeup_enabled_descendants : 0);
}

'udev' may or may not be a hub, if 'do_remote_wakeup' is set then the
device is considered a wakeup enabled descendant.

And for system suspebd 'do_remote_wakeup' is set based on the wakeup
config of the device:

static void choose_wakeup(struct usb_device *udev, pm_message_t msg)
{
	...
	w = device_may_wakeup(&udev->dev);
	...
	udev->do_remote_wakeup = w;
}

I checked on three systems with different Linux distributions, on all of
the wakeup flag of a connected hub is 'disabled'. Wakeup still works, so
apparently that flag doesn't really have an impact for child ports.

> If we consider hubs also dwc3 core exit and phy exit will never be called.
> 
> device_children_wakeup_capable() implementation was shared by Matthias in below thread
> https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
> 
> Probably device_children_wakeup_capable() is returning true because it considers hubs also.

I thought I did a basic test when I sent the patch, I did another (?) one
with v13 of your patch set. In this tests with a hub connected the
function returns true when an HID device is connected, and false when
nothing is connected. The wakeup flag of the hub is disabled (default).

Sandeep, are the wakeup flags of the child hub(s) set to 'enabled' on
the system you tested on?

Re: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Sandeep Maheswaram (Temp)]

Hi Matthias,

On 4/13/2022 12:35 AM, Matthias Kaehlcke wrote:
> On Tue, Apr 12, 2022 at 12:08:02PM +0530, Sandeep Maheswaram (Temp) wrote:
>> Hi Pavan,
>>
>> On 4/12/2022 10:30 AM, Pavan Kondeti wrote:
>>> Hi Sandeep,
>>>
>>> On Tue, Apr 12, 2022 at 10:16:39AM +0530, Sandeep Maheswaram (Temp) wrote:
>>>> Hi Matthias,
>>>>
>>>> On 4/12/2022 2:24 AM, Matthias Kaehlcke wrote:
>>>>> On Tue, Apr 12, 2022 at 12:46:50AM +0530, Sandeep Maheswaram wrote:
>>>>>> During suspend read the status of all port and set hs phy mode
>>>>>> based on current speed. Use this hs phy mode to configure wakeup
>>>>>> interrupts in qcom glue driver.
>>>>>>
>>>>>> Check wakep-source property for dwc3 core node to set the
>>>>> s/wakep/wakeup/
>>>> Okay. Will update in next version.
>>>>>> wakeup capability. Drop the device_init_wakeup call from
>>>>>> runtime suspend and resume.
>>>>>>
>>>>>> Also check during suspend if any wakeup capable devices are
>>>>>> connected to the controller (directly or through hubs), if there
>>>>>> are none set a flag to indicate that the PHY is powered
>>>>>> down during suspend.
>>>>>>
>>>>>> Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
>>>>>> ---
>>>>> A per-patch change log would be really helpful for reviewers, even
>>>>> if it doesn't include older versions.
>>>> Okay. Will update in next version.
>>>>>>    drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++-------------
>>>>>>    drivers/usb/dwc3/core.h |  4 ++++
>>>>>>    drivers/usb/dwc3/host.c | 25 +++++++++++++++++++++++++
>>>>>>    3 files changed, 49 insertions(+), 13 deletions(-)
>>>>>>
>>>>>> diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
>>>>>> index 1170b80..effaa43 100644
>>>>>> --- a/drivers/usb/dwc3/core.c
>>>>>> +++ b/drivers/usb/dwc3/core.c
>>>>>> @@ -32,6 +32,7 @@
>>>>>>    #include <linux/usb/gadget.h>
>>>>>>    #include <linux/usb/of.h>
>>>>>>    #include <linux/usb/otg.h>
>>>>>> +#include <linux/usb/hcd.h>
>>>>>>    #include "core.h"
>>>>>>    #include "gadget.h"
>>>>>> @@ -1723,6 +1724,7 @@ static int dwc3_probe(struct platform_device *pdev)
>>>>>>    	platform_set_drvdata(pdev, dwc);
>>>>>>    	dwc3_cache_hwparams(dwc);
>>>>>> +	device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
>>>>>>    	spin_lock_init(&dwc->lock);
>>>>>>    	mutex_init(&dwc->mutex);
>>>>>> @@ -1865,6 +1867,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>>>>>    {
>>>>>>    	unsigned long	flags;
>>>>>>    	u32 reg;
>>>>>> +	struct usb_hcd  *hcd = platform_get_drvdata(dwc->xhci);
>>>>>>    	switch (dwc->current_dr_role) {
>>>>>>    	case DWC3_GCTL_PRTCAP_DEVICE:
>>>>>> @@ -1877,10 +1880,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>>>>>    		dwc3_core_exit(dwc);
>>>>>>    		break;
>>>>>>    	case DWC3_GCTL_PRTCAP_HOST:
>>>>>> -		if (!PMSG_IS_AUTO(msg)) {
>>>>>> -			dwc3_core_exit(dwc);
>>>>>> -			break;
>>>>>> -		}
>>>>>> +		dwc3_check_phy_speed_mode(dwc);
>>>>>>    		/* Let controller to suspend HSPHY before PHY driver suspends */
>>>>>>    		if (dwc->dis_u2_susphy_quirk ||
>>>>>> @@ -1896,6 +1896,16 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>>>>>    		phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
>>>>>>    		phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
>>>>>> +
>>>>>> +		if (!PMSG_IS_AUTO(msg)) {
>>>>>> +			if (device_may_wakeup(dwc->dev) &&
>>>>>> +			    usb_wakeup_enabled_descendants(hcd->self.root_hub)) {
>>>>> You did not answer my question on v12, reposting it:
>>>>>
>>>>>     Did you ever try whether you could use device_children_wakeup_capable() from
>>>>>     [1] instead of usb_wakeup_enabled_descendants()?
>>>>>
>>>>>     [1] https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
>>>> Sorry ..I have replied in mail yesterday but it is not showing up in
>>>> patchwork link.
>>>>
>>>> Tried with  device_children_wakeup_capable(dwc->dev) instead of
>>>> usb_wakeup_enabled_descendants and it always returns true even
>>>>
>>>> when no devices are connected.
>>>>
>>> What do you mean by when no devices are connected? There is always
>>> root hub connected and we should not power down the DWC3 here even
>>> when remote wakeup for root hub is enabled. Essentially
>>> usb_wakeup_enabled_descendants() returns true even without any
>>> physical devices connected.
>>>
>>> What does device_children_wakeup_capable() do? Sorry, I could not
>>> find this function definition.
>>>
>>> Thanks,
>>> Pavan
>> usb_wakeup_enabled_descendants() doesn't consider hubs. It only returns true if any devices
>> are connected with wakeup capability apart from hubs.
> Actually it considers hubs:
>
> unsigned usb_wakeup_enabled_descendants(struct usb_device *udev)
> {
> 	struct usb_hub *hub = usb_hub_to_struct_hub(udev);
>
> 	return udev->do_remote_wakeup +
> 		(hub ? hub->wakeup_enabled_descendants : 0);
> }
>
> 'udev' may or may not be a hub, if 'do_remote_wakeup' is set then the
> device is considered a wakeup enabled descendant.
>
> And for system suspebd 'do_remote_wakeup' is set based on the wakeup
> config of the device:
>
> static void choose_wakeup(struct usb_device *udev, pm_message_t msg)
> {
> 	...
> 	w = device_may_wakeup(&udev->dev);
> 	...
> 	udev->do_remote_wakeup = w;
> }
>
> I checked on three systems with different Linux distributions, on all of
> the wakeup flag of a connected hub is 'disabled'. Wakeup still works, so
> apparently that flag doesn't really have an impact for child ports.
>
>> If we consider hubs also dwc3 core exit and phy exit will never be called.
>>
>> device_children_wakeup_capable() implementation was shared by Matthias in below thread
>> https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
>>
>> Probably device_children_wakeup_capable() is returning true because it considers hubs also.
> I thought I did a basic test when I sent the patch, I did another (?) one
> with v13 of your patch set. In this tests with a hub connected the
> function returns true when an HID device is connected, and false when
> nothing is connected. The wakeup flag of the hub is disabled (default).
>
> Sandeep, are the wakeup flags of the child hub(s) set to 'enabled' on
> the system you tested on?

The wakeup flags of hub is 'disabled' on system I tested.

What is the input param you are giving to device_children_wakeup_capable() function ?

I was passing the dwc->dev like this device_children_wakeup_capable(dwc->dev) .Is it right?

Regards
Sandeep

Re: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Matthias Kaehlcke]

On Wed, Apr 13, 2022 at 02:38:33PM +0530, Sandeep Maheswaram (Temp) wrote:
> Hi Matthias,
> 
> On 4/13/2022 12:35 AM, Matthias Kaehlcke wrote:
> > On Tue, Apr 12, 2022 at 12:08:02PM +0530, Sandeep Maheswaram (Temp) wrote:
> > > Hi Pavan,
> > > 
> > > On 4/12/2022 10:30 AM, Pavan Kondeti wrote:
> > > > Hi Sandeep,
> > > > 
> > > > On Tue, Apr 12, 2022 at 10:16:39AM +0530, Sandeep Maheswaram (Temp) wrote:
> > > > > Hi Matthias,
> > > > > 
> > > > > On 4/12/2022 2:24 AM, Matthias Kaehlcke wrote:
> > > > > > On Tue, Apr 12, 2022 at 12:46:50AM +0530, Sandeep Maheswaram wrote:
> > > > > > > During suspend read the status of all port and set hs phy mode
> > > > > > > based on current speed. Use this hs phy mode to configure wakeup
> > > > > > > interrupts in qcom glue driver.
> > > > > > > 
> > > > > > > Check wakep-source property for dwc3 core node to set the
> > > > > > s/wakep/wakeup/
> > > > > Okay. Will update in next version.
> > > > > > > wakeup capability. Drop the device_init_wakeup call from
> > > > > > > runtime suspend and resume.
> > > > > > > 
> > > > > > > Also check during suspend if any wakeup capable devices are
> > > > > > > connected to the controller (directly or through hubs), if there
> > > > > > > are none set a flag to indicate that the PHY is powered
> > > > > > > down during suspend.
> > > > > > > 
> > > > > > > Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
> > > > > > > ---
> > > > > > A per-patch change log would be really helpful for reviewers, even
> > > > > > if it doesn't include older versions.
> > > > > Okay. Will update in next version.
> > > > > > >    drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++-------------
> > > > > > >    drivers/usb/dwc3/core.h |  4 ++++
> > > > > > >    drivers/usb/dwc3/host.c | 25 +++++++++++++++++++++++++
> > > > > > >    3 files changed, 49 insertions(+), 13 deletions(-)
> > > > > > > 
> > > > > > > diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> > > > > > > index 1170b80..effaa43 100644
> > > > > > > --- a/drivers/usb/dwc3/core.c
> > > > > > > +++ b/drivers/usb/dwc3/core.c
> > > > > > > @@ -32,6 +32,7 @@
> > > > > > >    #include <linux/usb/gadget.h>
> > > > > > >    #include <linux/usb/of.h>
> > > > > > >    #include <linux/usb/otg.h>
> > > > > > > +#include <linux/usb/hcd.h>
> > > > > > >    #include "core.h"
> > > > > > >    #include "gadget.h"
> > > > > > > @@ -1723,6 +1724,7 @@ static int dwc3_probe(struct platform_device *pdev)
> > > > > > >    	platform_set_drvdata(pdev, dwc);
> > > > > > >    	dwc3_cache_hwparams(dwc);
> > > > > > > +	device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
> > > > > > >    	spin_lock_init(&dwc->lock);
> > > > > > >    	mutex_init(&dwc->mutex);
> > > > > > > @@ -1865,6 +1867,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > > > > > >    {
> > > > > > >    	unsigned long	flags;
> > > > > > >    	u32 reg;
> > > > > > > +	struct usb_hcd  *hcd = platform_get_drvdata(dwc->xhci);
> > > > > > >    	switch (dwc->current_dr_role) {
> > > > > > >    	case DWC3_GCTL_PRTCAP_DEVICE:
> > > > > > > @@ -1877,10 +1880,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > > > > > >    		dwc3_core_exit(dwc);
> > > > > > >    		break;
> > > > > > >    	case DWC3_GCTL_PRTCAP_HOST:
> > > > > > > -		if (!PMSG_IS_AUTO(msg)) {
> > > > > > > -			dwc3_core_exit(dwc);
> > > > > > > -			break;
> > > > > > > -		}
> > > > > > > +		dwc3_check_phy_speed_mode(dwc);
> > > > > > >    		/* Let controller to suspend HSPHY before PHY driver suspends */
> > > > > > >    		if (dwc->dis_u2_susphy_quirk ||
> > > > > > > @@ -1896,6 +1896,16 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > > > > > >    		phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
> > > > > > >    		phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
> > > > > > > +
> > > > > > > +		if (!PMSG_IS_AUTO(msg)) {
> > > > > > > +			if (device_may_wakeup(dwc->dev) &&
> > > > > > > +			    usb_wakeup_enabled_descendants(hcd->self.root_hub)) {
> > > > > > You did not answer my question on v12, reposting it:
> > > > > > 
> > > > > >     Did you ever try whether you could use device_children_wakeup_capable() from
> > > > > >     [1] instead of usb_wakeup_enabled_descendants()?
> > > > > > 
> > > > > >     [1] https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
> > > > > Sorry ..I have replied in mail yesterday but it is not showing up in
> > > > > patchwork link.
> > > > > 
> > > > > Tried with  device_children_wakeup_capable(dwc->dev) instead of
> > > > > usb_wakeup_enabled_descendants and it always returns true even
> > > > > 
> > > > > when no devices are connected.
> > > > > 
> > > > What do you mean by when no devices are connected? There is always
> > > > root hub connected and we should not power down the DWC3 here even
> > > > when remote wakeup for root hub is enabled. Essentially
> > > > usb_wakeup_enabled_descendants() returns true even without any
> > > > physical devices connected.
> > > > 
> > > > What does device_children_wakeup_capable() do? Sorry, I could not
> > > > find this function definition.
> > > > 
> > > > Thanks,
> > > > Pavan
> > > usb_wakeup_enabled_descendants() doesn't consider hubs. It only returns true if any devices
> > > are connected with wakeup capability apart from hubs.
> > Actually it considers hubs:
> > 
> > unsigned usb_wakeup_enabled_descendants(struct usb_device *udev)
> > {
> > 	struct usb_hub *hub = usb_hub_to_struct_hub(udev);
> > 
> > 	return udev->do_remote_wakeup +
> > 		(hub ? hub->wakeup_enabled_descendants : 0);
> > }
> > 
> > 'udev' may or may not be a hub, if 'do_remote_wakeup' is set then the
> > device is considered a wakeup enabled descendant.
> > 
> > And for system suspebd 'do_remote_wakeup' is set based on the wakeup
> > config of the device:
> > 
> > static void choose_wakeup(struct usb_device *udev, pm_message_t msg)
> > {
> > 	...
> > 	w = device_may_wakeup(&udev->dev);
> > 	...
> > 	udev->do_remote_wakeup = w;
> > }
> > 
> > I checked on three systems with different Linux distributions, on all of
> > the wakeup flag of a connected hub is 'disabled'. Wakeup still works, so
> > apparently that flag doesn't really have an impact for child ports.
> > 
> > > If we consider hubs also dwc3 core exit and phy exit will never be called.
> > > 
> > > device_children_wakeup_capable() implementation was shared by Matthias in below thread
> > > https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
> > > 
> > > Probably device_children_wakeup_capable() is returning true because it considers hubs also.
> > I thought I did a basic test when I sent the patch, I did another (?) one
> > with v13 of your patch set. In this tests with a hub connected the
> > function returns true when an HID device is connected, and false when
> > nothing is connected. The wakeup flag of the hub is disabled (default).
> > 
> > Sandeep, are the wakeup flags of the child hub(s) set to 'enabled' on
> > the system you tested on?
> 
> The wakeup flags of hub is 'disabled' on system I tested.
> 
> What is the input param you are giving to device_children_wakeup_capable() function ?

I passed '&hcd->self.root_hub->dev'

> I was passing the dwc->dev like this device_children_wakeup_capable(dwc->dev) .Is it right?

The DWC doesn't know anything about the connected USB devices, so this only
checks wakeup capability/policy of the DWC itself.

Re: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Sandeep Maheswaram (Temp)]

Hi Matthias,

On 4/13/2022 9:26 PM, Matthias Kaehlcke wrote:
> On Wed, Apr 13, 2022 at 02:38:33PM +0530, Sandeep Maheswaram (Temp) wrote:
>> Hi Matthias,
>>
>> On 4/13/2022 12:35 AM, Matthias Kaehlcke wrote:
>>> On Tue, Apr 12, 2022 at 12:08:02PM +0530, Sandeep Maheswaram (Temp) wrote:
>>>> Hi Pavan,
>>>>
>>>> On 4/12/2022 10:30 AM, Pavan Kondeti wrote:
>>>>> Hi Sandeep,
>>>>>
>>>>> On Tue, Apr 12, 2022 at 10:16:39AM +0530, Sandeep Maheswaram (Temp) wrote:
>>>>>> Hi Matthias,
>>>>>>
>>>>>> On 4/12/2022 2:24 AM, Matthias Kaehlcke wrote:
>>>>>>> On Tue, Apr 12, 2022 at 12:46:50AM +0530, Sandeep Maheswaram wrote:
>>>>>>>> During suspend read the status of all port and set hs phy mode
>>>>>>>> based on current speed. Use this hs phy mode to configure wakeup
>>>>>>>> interrupts in qcom glue driver.
>>>>>>>>
>>>>>>>> Check wakep-source property for dwc3 core node to set the
>>>>>>> s/wakep/wakeup/
>>>>>> Okay. Will update in next version.
>>>>>>>> wakeup capability. Drop the device_init_wakeup call from
>>>>>>>> runtime suspend and resume.
>>>>>>>>
>>>>>>>> Also check during suspend if any wakeup capable devices are
>>>>>>>> connected to the controller (directly or through hubs), if there
>>>>>>>> are none set a flag to indicate that the PHY is powered
>>>>>>>> down during suspend.
>>>>>>>>
>>>>>>>> Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
>>>>>>>> ---
>>>>>>> A per-patch change log would be really helpful for reviewers, even
>>>>>>> if it doesn't include older versions.
>>>>>> Okay. Will update in next version.
>>>>>>>>     drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++-------------
>>>>>>>>     drivers/usb/dwc3/core.h |  4 ++++
>>>>>>>>     drivers/usb/dwc3/host.c | 25 +++++++++++++++++++++++++
>>>>>>>>     3 files changed, 49 insertions(+), 13 deletions(-)
>>>>>>>>
>>>>>>>> diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
>>>>>>>> index 1170b80..effaa43 100644
>>>>>>>> --- a/drivers/usb/dwc3/core.c
>>>>>>>> +++ b/drivers/usb/dwc3/core.c
>>>>>>>> @@ -32,6 +32,7 @@
>>>>>>>>     #include <linux/usb/gadget.h>
>>>>>>>>     #include <linux/usb/of.h>
>>>>>>>>     #include <linux/usb/otg.h>
>>>>>>>> +#include <linux/usb/hcd.h>
>>>>>>>>     #include "core.h"
>>>>>>>>     #include "gadget.h"
>>>>>>>> @@ -1723,6 +1724,7 @@ static int dwc3_probe(struct platform_device *pdev)
>>>>>>>>     	platform_set_drvdata(pdev, dwc);
>>>>>>>>     	dwc3_cache_hwparams(dwc);
>>>>>>>> +	device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
>>>>>>>>     	spin_lock_init(&dwc->lock);
>>>>>>>>     	mutex_init(&dwc->mutex);
>>>>>>>> @@ -1865,6 +1867,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>>>>>>>     {
>>>>>>>>     	unsigned long	flags;
>>>>>>>>     	u32 reg;
>>>>>>>> +	struct usb_hcd  *hcd = platform_get_drvdata(dwc->xhci);
>>>>>>>>     	switch (dwc->current_dr_role) {
>>>>>>>>     	case DWC3_GCTL_PRTCAP_DEVICE:
>>>>>>>> @@ -1877,10 +1880,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>>>>>>>     		dwc3_core_exit(dwc);
>>>>>>>>     		break;
>>>>>>>>     	case DWC3_GCTL_PRTCAP_HOST:
>>>>>>>> -		if (!PMSG_IS_AUTO(msg)) {
>>>>>>>> -			dwc3_core_exit(dwc);
>>>>>>>> -			break;
>>>>>>>> -		}
>>>>>>>> +		dwc3_check_phy_speed_mode(dwc);
>>>>>>>>     		/* Let controller to suspend HSPHY before PHY driver suspends */
>>>>>>>>     		if (dwc->dis_u2_susphy_quirk ||
>>>>>>>> @@ -1896,6 +1896,16 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
>>>>>>>>     		phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
>>>>>>>>     		phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
>>>>>>>> +
>>>>>>>> +		if (!PMSG_IS_AUTO(msg)) {
>>>>>>>> +			if (device_may_wakeup(dwc->dev) &&
>>>>>>>> +			    usb_wakeup_enabled_descendants(hcd->self.root_hub)) {
>>>>>>> You did not answer my question on v12, reposting it:
>>>>>>>
>>>>>>>      Did you ever try whether you could use device_children_wakeup_capable() from
>>>>>>>      [1] instead of usb_wakeup_enabled_descendants()?
>>>>>>>
>>>>>>>      [1] https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
>>>>>> Sorry ..I have replied in mail yesterday but it is not showing up in
>>>>>> patchwork link.
>>>>>>
>>>>>> Tried with  device_children_wakeup_capable(dwc->dev) instead of
>>>>>> usb_wakeup_enabled_descendants and it always returns true even
>>>>>>
>>>>>> when no devices are connected.
>>>>>>
>>>>> What do you mean by when no devices are connected? There is always
>>>>> root hub connected and we should not power down the DWC3 here even
>>>>> when remote wakeup for root hub is enabled. Essentially
>>>>> usb_wakeup_enabled_descendants() returns true even without any
>>>>> physical devices connected.
>>>>>
>>>>> What does device_children_wakeup_capable() do? Sorry, I could not
>>>>> find this function definition.
>>>>>
>>>>> Thanks,
>>>>> Pavan
>>>> usb_wakeup_enabled_descendants() doesn't consider hubs. It only returns true if any devices
>>>> are connected with wakeup capability apart from hubs.
>>> Actually it considers hubs:
>>>
>>> unsigned usb_wakeup_enabled_descendants(struct usb_device *udev)
>>> {
>>> 	struct usb_hub *hub = usb_hub_to_struct_hub(udev);
>>>
>>> 	return udev->do_remote_wakeup +
>>> 		(hub ? hub->wakeup_enabled_descendants : 0);
>>> }
>>>
>>> 'udev' may or may not be a hub, if 'do_remote_wakeup' is set then the
>>> device is considered a wakeup enabled descendant.
>>>
>>> And for system suspebd 'do_remote_wakeup' is set based on the wakeup
>>> config of the device:
>>>
>>> static void choose_wakeup(struct usb_device *udev, pm_message_t msg)
>>> {
>>> 	...
>>> 	w = device_may_wakeup(&udev->dev);
>>> 	...
>>> 	udev->do_remote_wakeup = w;
>>> }
>>>
>>> I checked on three systems with different Linux distributions, on all of
>>> the wakeup flag of a connected hub is 'disabled'. Wakeup still works, so
>>> apparently that flag doesn't really have an impact for child ports.
>>>
>>>> If we consider hubs also dwc3 core exit and phy exit will never be called.
>>>>
>>>> device_children_wakeup_capable() implementation was shared by Matthias in below thread
>>>> https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
>>>>
>>>> Probably device_children_wakeup_capable() is returning true because it considers hubs also.
>>> I thought I did a basic test when I sent the patch, I did another (?) one
>>> with v13 of your patch set. In this tests with a hub connected the
>>> function returns true when an HID device is connected, and false when
>>> nothing is connected. The wakeup flag of the hub is disabled (default).
>>>
>>> Sandeep, are the wakeup flags of the child hub(s) set to 'enabled' on
>>> the system you tested on?
>> The wakeup flags of hub is 'disabled' on system I tested.
>>
>> What is the input param you are giving to device_children_wakeup_capable() function ?
> I passed '&hcd->self.root_hub->dev'

Thanks. It is working with this change device_children_wakeup_capable 
(&hcd->self.root_hub->dev).

But I am not sure if it is better than usb_wakeup_enabled_descendants. 
Still we are accessing xhci layer

from dwc which Felipe suggested to avoid.

>
>> I was passing the dwc->dev like this device_children_wakeup_capable(dwc->dev) .Is it right?
> The DWC doesn't know anything about the connected USB devices, so this only
> checks wakeup capability/policy of the DWC itself.

Re: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Matthias Kaehlcke]

On Thu, Apr 14, 2022 at 11:27:31AM +0530, Sandeep Maheswaram (Temp) wrote:
> Hi Matthias,
> 
> On 4/13/2022 9:26 PM, Matthias Kaehlcke wrote:
> > On Wed, Apr 13, 2022 at 02:38:33PM +0530, Sandeep Maheswaram (Temp) wrote:
> > > Hi Matthias,
> > > 
> > > On 4/13/2022 12:35 AM, Matthias Kaehlcke wrote:
> > > > On Tue, Apr 12, 2022 at 12:08:02PM +0530, Sandeep Maheswaram (Temp) wrote:
> > > > > Hi Pavan,
> > > > > 
> > > > > On 4/12/2022 10:30 AM, Pavan Kondeti wrote:
> > > > > > Hi Sandeep,
> > > > > > 
> > > > > > On Tue, Apr 12, 2022 at 10:16:39AM +0530, Sandeep Maheswaram (Temp) wrote:
> > > > > > > Hi Matthias,
> > > > > > > 
> > > > > > > On 4/12/2022 2:24 AM, Matthias Kaehlcke wrote:
> > > > > > > > On Tue, Apr 12, 2022 at 12:46:50AM +0530, Sandeep Maheswaram wrote:
> > > > > > > > > During suspend read the status of all port and set hs phy mode
> > > > > > > > > based on current speed. Use this hs phy mode to configure wakeup
> > > > > > > > > interrupts in qcom glue driver.
> > > > > > > > > 
> > > > > > > > > Check wakep-source property for dwc3 core node to set the
> > > > > > > > s/wakep/wakeup/
> > > > > > > Okay. Will update in next version.
> > > > > > > > > wakeup capability. Drop the device_init_wakeup call from
> > > > > > > > > runtime suspend and resume.
> > > > > > > > > 
> > > > > > > > > Also check during suspend if any wakeup capable devices are
> > > > > > > > > connected to the controller (directly or through hubs), if there
> > > > > > > > > are none set a flag to indicate that the PHY is powered
> > > > > > > > > down during suspend.
> > > > > > > > > 
> > > > > > > > > Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
> > > > > > > > > ---
> > > > > > > > A per-patch change log would be really helpful for reviewers, even
> > > > > > > > if it doesn't include older versions.
> > > > > > > Okay. Will update in next version.
> > > > > > > > >     drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++-------------
> > > > > > > > >     drivers/usb/dwc3/core.h |  4 ++++
> > > > > > > > >     drivers/usb/dwc3/host.c | 25 +++++++++++++++++++++++++
> > > > > > > > >     3 files changed, 49 insertions(+), 13 deletions(-)
> > > > > > > > > 
> > > > > > > > > diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> > > > > > > > > index 1170b80..effaa43 100644
> > > > > > > > > --- a/drivers/usb/dwc3/core.c
> > > > > > > > > +++ b/drivers/usb/dwc3/core.c
> > > > > > > > > @@ -32,6 +32,7 @@
> > > > > > > > >     #include <linux/usb/gadget.h>
> > > > > > > > >     #include <linux/usb/of.h>
> > > > > > > > >     #include <linux/usb/otg.h>
> > > > > > > > > +#include <linux/usb/hcd.h>
> > > > > > > > >     #include "core.h"
> > > > > > > > >     #include "gadget.h"
> > > > > > > > > @@ -1723,6 +1724,7 @@ static int dwc3_probe(struct platform_device *pdev)
> > > > > > > > >     	platform_set_drvdata(pdev, dwc);
> > > > > > > > >     	dwc3_cache_hwparams(dwc);
> > > > > > > > > +	device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
> > > > > > > > >     	spin_lock_init(&dwc->lock);
> > > > > > > > >     	mutex_init(&dwc->mutex);
> > > > > > > > > @@ -1865,6 +1867,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > > > > > > > >     {
> > > > > > > > >     	unsigned long	flags;
> > > > > > > > >     	u32 reg;
> > > > > > > > > +	struct usb_hcd  *hcd = platform_get_drvdata(dwc->xhci);
> > > > > > > > >     	switch (dwc->current_dr_role) {
> > > > > > > > >     	case DWC3_GCTL_PRTCAP_DEVICE:
> > > > > > > > > @@ -1877,10 +1880,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > > > > > > > >     		dwc3_core_exit(dwc);
> > > > > > > > >     		break;
> > > > > > > > >     	case DWC3_GCTL_PRTCAP_HOST:
> > > > > > > > > -		if (!PMSG_IS_AUTO(msg)) {
> > > > > > > > > -			dwc3_core_exit(dwc);
> > > > > > > > > -			break;
> > > > > > > > > -		}
> > > > > > > > > +		dwc3_check_phy_speed_mode(dwc);
> > > > > > > > >     		/* Let controller to suspend HSPHY before PHY driver suspends */
> > > > > > > > >     		if (dwc->dis_u2_susphy_quirk ||
> > > > > > > > > @@ -1896,6 +1896,16 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > > > > > > > >     		phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
> > > > > > > > >     		phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
> > > > > > > > > +
> > > > > > > > > +		if (!PMSG_IS_AUTO(msg)) {
> > > > > > > > > +			if (device_may_wakeup(dwc->dev) &&
> > > > > > > > > +			    usb_wakeup_enabled_descendants(hcd->self.root_hub)) {
> > > > > > > > You did not answer my question on v12, reposting it:
> > > > > > > > 
> > > > > > > >      Did you ever try whether you could use device_children_wakeup_capable() from
> > > > > > > >      [1] instead of usb_wakeup_enabled_descendants()?
> > > > > > > > 
> > > > > > > >      [1] https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
> > > > > > > Sorry ..I have replied in mail yesterday but it is not showing up in
> > > > > > > patchwork link.
> > > > > > > 
> > > > > > > Tried with  device_children_wakeup_capable(dwc->dev) instead of
> > > > > > > usb_wakeup_enabled_descendants and it always returns true even
> > > > > > > 
> > > > > > > when no devices are connected.
> > > > > > > 
> > > > > > What do you mean by when no devices are connected? There is always
> > > > > > root hub connected and we should not power down the DWC3 here even
> > > > > > when remote wakeup for root hub is enabled. Essentially
> > > > > > usb_wakeup_enabled_descendants() returns true even without any
> > > > > > physical devices connected.
> > > > > > 
> > > > > > What does device_children_wakeup_capable() do? Sorry, I could not
> > > > > > find this function definition.
> > > > > > 
> > > > > > Thanks,
> > > > > > Pavan
> > > > > usb_wakeup_enabled_descendants() doesn't consider hubs. It only returns true if any devices
> > > > > are connected with wakeup capability apart from hubs.
> > > > Actually it considers hubs:
> > > > 
> > > > unsigned usb_wakeup_enabled_descendants(struct usb_device *udev)
> > > > {
> > > > 	struct usb_hub *hub = usb_hub_to_struct_hub(udev);
> > > > 
> > > > 	return udev->do_remote_wakeup +
> > > > 		(hub ? hub->wakeup_enabled_descendants : 0);
> > > > }
> > > > 
> > > > 'udev' may or may not be a hub, if 'do_remote_wakeup' is set then the
> > > > device is considered a wakeup enabled descendant.
> > > > 
> > > > And for system suspebd 'do_remote_wakeup' is set based on the wakeup
> > > > config of the device:
> > > > 
> > > > static void choose_wakeup(struct usb_device *udev, pm_message_t msg)
> > > > {
> > > > 	...
> > > > 	w = device_may_wakeup(&udev->dev);
> > > > 	...
> > > > 	udev->do_remote_wakeup = w;
> > > > }
> > > > 
> > > > I checked on three systems with different Linux distributions, on all of
> > > > the wakeup flag of a connected hub is 'disabled'. Wakeup still works, so
> > > > apparently that flag doesn't really have an impact for child ports.
> > > > 
> > > > > If we consider hubs also dwc3 core exit and phy exit will never be called.
> > > > > 
> > > > > device_children_wakeup_capable() implementation was shared by Matthias in below thread
> > > > > https://patchwork.kernel.org/project/linux-usb/patch/1635753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
> > > > > 
> > > > > Probably device_children_wakeup_capable() is returning true because it considers hubs also.
> > > > I thought I did a basic test when I sent the patch, I did another (?) one
> > > > with v13 of your patch set. In this tests with a hub connected the
> > > > function returns true when an HID device is connected, and false when
> > > > nothing is connected. The wakeup flag of the hub is disabled (default).
> > > > 
> > > > Sandeep, are the wakeup flags of the child hub(s) set to 'enabled' on
> > > > the system you tested on?
> > > The wakeup flags of hub is 'disabled' on system I tested.
> > > 
> > > What is the input param you are giving to device_children_wakeup_capable() function ?
> > I passed '&hcd->self.root_hub->dev'
> 
> Thanks. It is working with this change device_children_wakeup_capable
> (&hcd->self.root_hub->dev).
> 
> But I am not sure if it is better than usb_wakeup_enabled_descendants. Still
> we are accessing xhci layer
> 
> from dwc which Felipe suggested to avoid.

True, it still needs access to the data structure(s), even though it doesn't
use a USB specific API.

Would be good to get feedback from Felipe on the current approach in general,
we haven't heard from him in some time.

RE: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from system suspend

[Sandeep Maheswaram (Temp) (QUIC)]

Hi Matthias,

> -----Original Message-----
> From: Matthias Kaehlcke <mka@chromium.org>
> Sent: Saturday, April 16, 2022 6:00 AM
> To: Sandeep Maheswaram (Temp) (QUIC) <quic_c_sanm@quicinc.com>
> Cc: Pavan Kumar Kondeti (QUIC) <quic_pkondeti@quicinc.com>; Rob Herring
> <robh+dt@kernel.org>; Andy Gross <agross@kernel.org>;
> bjorn.andersson@linaro.org; Greg Kroah-Hartman
> <gregkh@linuxfoundation.org>; Felipe Balbi <balbi@kernel.org>; Stephen
> Boyd <swboyd@chromium.org>; Doug Anderson
> <dianders@chromium.org>; Mathias Nyman <mathias.nyman@intel.com>;
> Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>;
> devicetree@vger.kernel.org; linux-arm-msm@vger.kernel.org; linux-
> usb@vger.kernel.org; linux-kernel@vger.kernel.org; Pratham Pratap (QUIC)
> <quic_ppratap@quicinc.com>; Krishna Kurapati PSSNV (QUIC)
> <quic_kriskura@quicinc.com>; Vidya Sagar Pulyala (Temp) (QUIC)
> <quic_vpulyala@quicinc.com>
> Subject: Re: [PATCH v13 2/6] usb: dwc3: core: Host wake up support from
> system suspend
> 
> WARNING: This email originated from outside of Qualcomm. Please be wary
> of any links or attachments, and do not enable macros.
> 
> On Thu, Apr 14, 2022 at 11:27:31AM +0530, Sandeep Maheswaram (Temp)
> wrote:
> > Hi Matthias,
> >
> > On 4/13/2022 9:26 PM, Matthias Kaehlcke wrote:
> > > On Wed, Apr 13, 2022 at 02:38:33PM +0530, Sandeep Maheswaram
> (Temp) wrote:
> > > > Hi Matthias,
> > > >
> > > > On 4/13/2022 12:35 AM, Matthias Kaehlcke wrote:
> > > > > On Tue, Apr 12, 2022 at 12:08:02PM +0530, Sandeep Maheswaram
> (Temp) wrote:
> > > > > > Hi Pavan,
> > > > > >
> > > > > > On 4/12/2022 10:30 AM, Pavan Kondeti wrote:
> > > > > > > Hi Sandeep,
> > > > > > >
> > > > > > > On Tue, Apr 12, 2022 at 10:16:39AM +0530, Sandeep Maheswaram
> (Temp) wrote:
> > > > > > > > Hi Matthias,
> > > > > > > >
> > > > > > > > On 4/12/2022 2:24 AM, Matthias Kaehlcke wrote:
> > > > > > > > > On Tue, Apr 12, 2022 at 12:46:50AM +0530, Sandeep
> Maheswaram wrote:
> > > > > > > > > > During suspend read the status of all port and set hs
> > > > > > > > > > phy mode based on current speed. Use this hs phy mode
> > > > > > > > > > to configure wakeup interrupts in qcom glue driver.
> > > > > > > > > >
> > > > > > > > > > Check wakep-source property for dwc3 core node to set
> > > > > > > > > > the
> > > > > > > > > s/wakep/wakeup/
> > > > > > > > Okay. Will update in next version.
> > > > > > > > > > wakeup capability. Drop the device_init_wakeup call
> > > > > > > > > > from runtime suspend and resume.
> > > > > > > > > >
> > > > > > > > > > Also check during suspend if any wakeup capable
> > > > > > > > > > devices are connected to the controller (directly or
> > > > > > > > > > through hubs), if there are none set a flag to
> > > > > > > > > > indicate that the PHY is powered down during suspend.
> > > > > > > > > >
> > > > > > > > > > Signed-off-by: Sandeep Maheswaram
> > > > > > > > > > <quic_c_sanm@quicinc.com>
> > > > > > > > > > ---
> > > > > > > > > A per-patch change log would be really helpful for
> > > > > > > > > reviewers, even if it doesn't include older versions.
> > > > > > > > Okay. Will update in next version.
> > > > > > > > > >     drivers/usb/dwc3/core.c | 33 ++++++++++++++++++++---
> ----------
> > > > > > > > > >     drivers/usb/dwc3/core.h |  4 ++++
> > > > > > > > > >     drivers/usb/dwc3/host.c | 25
> +++++++++++++++++++++++++
> > > > > > > > > >     3 files changed, 49 insertions(+), 13 deletions(-)
> > > > > > > > > >
> > > > > > > > > > diff --git a/drivers/usb/dwc3/core.c
> > > > > > > > > > b/drivers/usb/dwc3/core.c index 1170b80..effaa43
> > > > > > > > > > 100644
> > > > > > > > > > --- a/drivers/usb/dwc3/core.c
> > > > > > > > > > +++ b/drivers/usb/dwc3/core.c
> > > > > > > > > > @@ -32,6 +32,7 @@
> > > > > > > > > >     #include <linux/usb/gadget.h>
> > > > > > > > > >     #include <linux/usb/of.h>
> > > > > > > > > >     #include <linux/usb/otg.h>
> > > > > > > > > > +#include <linux/usb/hcd.h>
> > > > > > > > > >     #include "core.h"
> > > > > > > > > >     #include "gadget.h"
> > > > > > > > > > @@ -1723,6 +1724,7 @@ static int dwc3_probe(struct
> platform_device *pdev)
> > > > > > > > > >       platform_set_drvdata(pdev, dwc);
> > > > > > > > > >       dwc3_cache_hwparams(dwc);
> > > > > > > > > > +     device_init_wakeup(&pdev->dev,
> > > > > > > > > > + of_property_read_bool(dev->of_node,
> > > > > > > > > > + "wakeup-source"));
> > > > > > > > > >       spin_lock_init(&dwc->lock);
> > > > > > > > > >       mutex_init(&dwc->mutex); @@ -1865,6 +1867,7 @@
> > > > > > > > > > static int dwc3_suspend_common(struct dwc3 *dwc,
> pm_message_t msg)
> > > > > > > > > >     {
> > > > > > > > > >       unsigned long   flags;
> > > > > > > > > >       u32 reg;
> > > > > > > > > > +     struct usb_hcd  *hcd =
> > > > > > > > > > + platform_get_drvdata(dwc->xhci);
> > > > > > > > > >       switch (dwc->current_dr_role) {
> > > > > > > > > >       case DWC3_GCTL_PRTCAP_DEVICE:
> > > > > > > > > > @@ -1877,10 +1880,7 @@ static int
> dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > > > > > > > > >               dwc3_core_exit(dwc);
> > > > > > > > > >               break;
> > > > > > > > > >       case DWC3_GCTL_PRTCAP_HOST:
> > > > > > > > > > -             if (!PMSG_IS_AUTO(msg)) {
> > > > > > > > > > -                     dwc3_core_exit(dwc);
> > > > > > > > > > -                     break;
> > > > > > > > > > -             }
> > > > > > > > > > +             dwc3_check_phy_speed_mode(dwc);
> > > > > > > > > >               /* Let controller to suspend HSPHY before PHY driver
> suspends */
> > > > > > > > > >               if (dwc->dis_u2_susphy_quirk || @@
> > > > > > > > > > -1896,6 +1896,16 @@ static int
> dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > > > > > > > > >               phy_pm_runtime_put_sync(dwc-
> >usb2_generic_phy);
> > > > > > > > > >
> > > > > > > > > > phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
> > > > > > > > > > +
> > > > > > > > > > +             if (!PMSG_IS_AUTO(msg)) {
> > > > > > > > > > +                     if (device_may_wakeup(dwc->dev) &&
> > > > > > > > > > +
> > > > > > > > > > + usb_wakeup_enabled_descendants(hcd->self.root_hub))
> > > > > > > > > > + {
> > > > > > > > > You did not answer my question on v12, reposting it:
> > > > > > > > >
> > > > > > > > >      Did you ever try whether you could use
> device_children_wakeup_capable() from
> > > > > > > > >      [1] instead of usb_wakeup_enabled_descendants()?
> > > > > > > > >
> > > > > > > > >      [1]
> > > > > > > > > https://patchwork.kernel.org/project/linux-usb/patch/163
> > > > > > > > > 5753224-23975-2-git-send-email-quic_c_sanm@quicinc.com/#
> > > > > > > > > 24566065
> > > > > > > > Sorry ..I have replied in mail yesterday but it is not
> > > > > > > > showing up in patchwork link.
> > > > > > > >
> > > > > > > > Tried with  device_children_wakeup_capable(dwc->dev)
> > > > > > > > instead of usb_wakeup_enabled_descendants and it always
> > > > > > > > returns true even
> > > > > > > >
> > > > > > > > when no devices are connected.
> > > > > > > >
> > > > > > > What do you mean by when no devices are connected? There is
> > > > > > > always root hub connected and we should not power down the
> > > > > > > DWC3 here even when remote wakeup for root hub is enabled.
> > > > > > > Essentially
> > > > > > > usb_wakeup_enabled_descendants() returns true even without
> > > > > > > any physical devices connected.
> > > > > > >
> > > > > > > What does device_children_wakeup_capable() do? Sorry, I
> > > > > > > could not find this function definition.
> > > > > > >
> > > > > > > Thanks,
> > > > > > > Pavan
> > > > > > usb_wakeup_enabled_descendants() doesn't consider hubs. It
> > > > > > only returns true if any devices are connected with wakeup
> capability apart from hubs.
> > > > > Actually it considers hubs:
> > > > >
> > > > > unsigned usb_wakeup_enabled_descendants(struct usb_device
> *udev)
> > > > > {
> > > > >         struct usb_hub *hub = usb_hub_to_struct_hub(udev);
> > > > >
> > > > >         return udev->do_remote_wakeup +
> > > > >                 (hub ? hub->wakeup_enabled_descendants : 0); }
> > > > >
> > > > > 'udev' may or may not be a hub, if 'do_remote_wakeup' is set
> > > > > then the device is considered a wakeup enabled descendant.
> > > > >
> > > > > And for system suspebd 'do_remote_wakeup' is set based on the
> > > > > wakeup config of the device:
> > > > >
> > > > > static void choose_wakeup(struct usb_device *udev, pm_message_t
> > > > > msg) {
> > > > >         ...
> > > > >         w = device_may_wakeup(&udev->dev);
> > > > >         ...
> > > > >         udev->do_remote_wakeup = w; }
> > > > >
> > > > > I checked on three systems with different Linux distributions,
> > > > > on all of the wakeup flag of a connected hub is 'disabled'.
> > > > > Wakeup still works, so apparently that flag doesn't really have an
> impact for child ports.
> > > > >
> > > > > > If we consider hubs also dwc3 core exit and phy exit will never be
> called.
> > > > > >
> > > > > > device_children_wakeup_capable() implementation was shared by
> > > > > > Matthias in below thread
> > > > > > https://patchwork.kernel.org/project/linux-usb/patch/163575322
> > > > > > 4-23975-2-git-send-email-quic_c_sanm@quicinc.com/#24566065
> > > > > >
> > > > > > Probably device_children_wakeup_capable() is returning true
> because it considers hubs also.
> > > > > I thought I did a basic test when I sent the patch, I did
> > > > > another (?) one with v13 of your patch set. In this tests with a
> > > > > hub connected the function returns true when an HID device is
> > > > > connected, and false when nothing is connected. The wakeup flag of
> the hub is disabled (default).
> > > > >
> > > > > Sandeep, are the wakeup flags of the child hub(s) set to
> > > > > 'enabled' on the system you tested on?
> > > > The wakeup flags of hub is 'disabled' on system I tested.
> > > >
> > > > What is the input param you are giving to
> device_children_wakeup_capable() function ?
> > > I passed '&hcd->self.root_hub->dev'
> >
> > Thanks. It is working with this change device_children_wakeup_capable
> > (&hcd->self.root_hub->dev).
> >
> > But I am not sure if it is better than usb_wakeup_enabled_descendants.
> > Still we are accessing xhci layer
> >
> > from dwc which Felipe suggested to avoid.
> 
> True, it still needs access to the data structure(s), even though it doesn't use
> a USB specific API.
> 
> Would be good to get feedback from Felipe on the current approach in
> general, we haven't heard from him in some time.

Will send the new version with device_children_wakeup_capable including your patch
And ask Felipe for his opinion.

Regards
Sandeep

Re: (subset) [PATCH v13 6/6] arm64: dts: qcom: sc7280: Add wakeup-source property for USB node

[Bjorn Andersson]

On Tue, 12 Apr 2022 00:46:54 +0530, Sandeep Maheswaram wrote:
> Adding wakeup-source property for USB controller in SC7280.
> This property is added to inform that the USB controller is
> wake up capable and to conditionally power down the phy during
> system suspend.
> 
> 

Applied, thanks!

[6/6] arm64: dts: qcom: sc7280: Add wakeup-source property for USB node
      commit: 97276cbfb4fbe33c3ab8b5f5277a73c0b5f3732b

Best regards,
-- 
Bjorn Andersson <bjorn.andersson@linaro.org>