[PATCH v2 0/1] PCI: qcom: Add support for system suspend and resume

[PATCH v2 0/1] PCI: qcom: Add support for system suspend and resume

[Manivannan Sadhasivam]

Hello,

This series (a single patch) adds the system suspend and resume support
to the Qualcomm PCIe RC controller.

Background
==========

There were previous attempts [1][2] to add system suspend and resume
support to this driver.

In previous versions, the controller was put into low power mode by turning
OFF the resources even if there were active PCIe devices connected. Thanks
to Qualcomm's internal power topology, the link did not enter L2/L3 state
and the devices were still powered ON. But during very late end of suspend
cycle, kernel tried to disable MSIs of the PCIe devices. This caused access
violations as the resources needed to access the PCIe devices config space
were turned OFF. Series [1] worked around this issue by not accessing the
PCIe config space if the link was down in dw_msi_{un}mask_irq() functions.
But that approach was not accepted.

Then, series [2] implemented the suspend and resume operations using the
syscore framework that disabled the resources at the end of the suspend
cycle. But that approach also did not get much acceptance.

Proposal
========

So the proposal here is to just vote for minimal interconnect bandwidth and
not turn OFF the resources if there are active PCIe devices connected to
the controllers. This avoids the access violation issue during suspend and
also saves some power due to the lower interconnect bandwidth used.

Then if there are no active PCIe devices connected to the controller,
the resources are turned OFF completely and brought back during resume.
This also saves power if there are controllers in a system without any
devices connected.

Testing
=======

This series has been tested on Lenovo Thinkpad X13s.

Thanks,
Mani

[1] https://lore.kernel.org/linux-pci/1656055682-18817-1-git-send-email-quic_krichai@quicinc.com/
[2] https://lore.kernel.org/linux-pci/1663669347-29308-1-git-send-email-quic_krichai@quicinc.com/

Changes in v2:

* Used minimum icc vote to keep data path functional during suspend
* Collected Ack

Manivannan Sadhasivam (1):
  PCI: qcom: Add support for system suspend and resume

 drivers/pci/controller/dwc/pcie-qcom.c | 53 ++++++++++++++++++++++++++
 1 file changed, 53 insertions(+)

-- 
2.25.1

[PATCH v2 1/1] PCI: qcom: Add support for system suspend and resume

[Manivannan Sadhasivam]

During the system suspend, vote for minimal interconnect bandwidth and
also turn OFF the resources like clock and PHY if there are no active
devices connected to the controller. For the controllers with active
devices, the resources are kept ON as removing the resources will
trigger access violation during the late end of suspend cycle as kernel
tries to access the config space of PCIe devices to mask the MSIs.

Also, it is not desirable to put the link into L2/L3 state as that
implies VDD supply will be removed and the devices may go into powerdown
state. This will affect the lifetime of storage devices like NVMe.

And finally, during resume, turn ON the resources if the controller was
truly suspended (resources OFF) and update the interconnect bandwidth
based on PCIe Gen speed.

Suggested-by: Krishna chaitanya chundru <quic_krichai@quicinc.com>
Acked-by: Dhruva Gole <d-gole@ti.com>
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
---
 drivers/pci/controller/dwc/pcie-qcom.c | 53 ++++++++++++++++++++++++++
 1 file changed, 53 insertions(+)

diff --git a/drivers/pci/controller/dwc/pcie-qcom.c b/drivers/pci/controller/dwc/pcie-qcom.c
index a232b04af048..7147f0103026 100644
--- a/drivers/pci/controller/dwc/pcie-qcom.c
+++ b/drivers/pci/controller/dwc/pcie-qcom.c
@@ -227,6 +227,7 @@ struct qcom_pcie {
 	struct gpio_desc *reset;
 	struct icc_path *icc_mem;
 	const struct qcom_pcie_cfg *cfg;
+	bool suspended;
 };
 
 #define to_qcom_pcie(x)		dev_get_drvdata((x)->dev)
@@ -1820,6 +1821,53 @@ static int qcom_pcie_probe(struct platform_device *pdev)
 	return ret;
 }
 
+static int qcom_pcie_suspend_noirq(struct device *dev)
+{
+	struct qcom_pcie *pcie = dev_get_drvdata(dev);
+	int ret;
+
+	/* Set minimum bandwidth required to keep data path functional during suspend */
+	ret = icc_set_bw(pcie->icc_mem, 0, MBps_to_icc(250));
+	if (ret) {
+		dev_err(pcie->pci->dev, "Failed to set interconnect bandwidth: %d\n", ret);
+		return ret;
+	}
+
+	/*
+	 * Turn OFF the resources only for controllers without active PCIe devices. For controllers
+	 * with active devices, the resources are kept ON and the link is expected to be in L0/L1
+	 * (sub)states.
+	 *
+	 * Turning OFF the resources for controllers with active PCIe devices will trigger access
+	 * violation during the end of the suspend cycle, as kernel tries to access the PCIe devices
+	 * config space for masking MSIs.
+	 *
+	 * Also, it is not desirable to put the link into L2/L3 state as that implies VDD supply
+	 * will be removed and the devices may go into powerdown state. This will affect the
+	 * lifetime of the storage devices like NVMe.
+	 */
+	if (!dw_pcie_link_up(pcie->pci)) {
+		qcom_pcie_host_deinit(&pcie->pci->pp);
+		pcie->suspended = true;
+	}
+
+	return 0;
+}
+
+static int qcom_pcie_resume_noirq(struct device *dev)
+{
+	struct qcom_pcie *pcie = dev_get_drvdata(dev);
+
+	if (pcie->suspended) {
+		qcom_pcie_host_init(&pcie->pci->pp);
+		pcie->suspended = false;
+	}
+
+	qcom_pcie_icc_update(pcie);
+
+	return 0;
+}
+
 static const struct of_device_id qcom_pcie_match[] = {
 	{ .compatible = "qcom,pcie-apq8064", .data = &cfg_2_1_0 },
 	{ .compatible = "qcom,pcie-apq8084", .data = &cfg_1_0_0 },
@@ -1856,12 +1904,17 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_QCOM, 0x0302, qcom_fixup_class);
 DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_QCOM, 0x1000, qcom_fixup_class);
 DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_QCOM, 0x1001, qcom_fixup_class);
 
+static const struct dev_pm_ops qcom_pcie_pm_ops = {
+	NOIRQ_SYSTEM_SLEEP_PM_OPS(qcom_pcie_suspend_noirq, qcom_pcie_resume_noirq)
+};
+
 static struct platform_driver qcom_pcie_driver = {
 	.probe = qcom_pcie_probe,
 	.driver = {
 		.name = "qcom-pcie",
 		.suppress_bind_attrs = true,
 		.of_match_table = qcom_pcie_match,
+		.pm = &qcom_pcie_pm_ops,
 	},
 };
 builtin_platform_driver(qcom_pcie_driver);
-- 
2.25.1

Re: [PATCH v2 1/1] PCI: qcom: Add support for system suspend and resume

[Johan Hovold]

On Thu, Mar 02, 2023 at 02:06:25PM +0530, Manivannan Sadhasivam wrote:
> During the system suspend, vote for minimal interconnect bandwidth and
> also turn OFF the resources like clock and PHY if there are no active
> devices connected to the controller. For the controllers with active
> devices, the resources are kept ON as removing the resources will
> trigger access violation during the late end of suspend cycle as kernel
> tries to access the config space of PCIe devices to mask the MSIs.
> 
> Also, it is not desirable to put the link into L2/L3 state as that
> implies VDD supply will be removed and the devices may go into powerdown
> state. This will affect the lifetime of storage devices like NVMe.
> 
> And finally, during resume, turn ON the resources if the controller was
> truly suspended (resources OFF) and update the interconnect bandwidth
> based on PCIe Gen speed.
> 
> Suggested-by: Krishna chaitanya chundru <quic_krichai@quicinc.com>
> Acked-by: Dhruva Gole <d-gole@ti.com>
> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
> ---
>  drivers/pci/controller/dwc/pcie-qcom.c | 53 ++++++++++++++++++++++++++
>  1 file changed, 53 insertions(+)
> 
> diff --git a/drivers/pci/controller/dwc/pcie-qcom.c b/drivers/pci/controller/dwc/pcie-qcom.c
> index a232b04af048..7147f0103026 100644
> --- a/drivers/pci/controller/dwc/pcie-qcom.c
> +++ b/drivers/pci/controller/dwc/pcie-qcom.c
> @@ -227,6 +227,7 @@ struct qcom_pcie {
>  	struct gpio_desc *reset;
>  	struct icc_path *icc_mem;
>  	const struct qcom_pcie_cfg *cfg;
> +	bool suspended;
>  };
>  
>  #define to_qcom_pcie(x)		dev_get_drvdata((x)->dev)
> @@ -1820,6 +1821,53 @@ static int qcom_pcie_probe(struct platform_device *pdev)
>  	return ret;
>  }
>  
> +static int qcom_pcie_suspend_noirq(struct device *dev)
> +{
> +	struct qcom_pcie *pcie = dev_get_drvdata(dev);
> +	int ret;
> +
> +	/* Set minimum bandwidth required to keep data path functional during suspend */
> +	ret = icc_set_bw(pcie->icc_mem, 0, MBps_to_icc(250));

I can again confirm that this now passes basic smoke testing on my CRD
(unlike v1 which set bw to 0) and X13s.

> +	if (ret) {
> +		dev_err(pcie->pci->dev, "Failed to set interconnect bandwidth: %d\n", ret);

You already have a pointer to dev.

> +		return ret;
> +	}

And please try to stay within 80 columns width unless breaking it
improves readability (generally does not apply comments).

> +
> +	/*
> +	 * Turn OFF the resources only for controllers without active PCIe devices. For controllers
> +	 * with active devices, the resources are kept ON and the link is expected to be in L0/L1
> +	 * (sub)states.
> +	 *
> +	 * Turning OFF the resources for controllers with active PCIe devices will trigger access
> +	 * violation during the end of the suspend cycle, as kernel tries to access the PCIe devices
> +	 * config space for masking MSIs.
> +	 *
> +	 * Also, it is not desirable to put the link into L2/L3 state as that implies VDD supply
> +	 * will be removed and the devices may go into powerdown state. This will affect the
> +	 * lifetime of the storage devices like NVMe.
> +	 */

Same here.

> +	if (!dw_pcie_link_up(pcie->pci)) {
> +		qcom_pcie_host_deinit(&pcie->pci->pp);
> +		pcie->suspended = true;
> +	}
> +
> +	return 0;
> +}
> +
> +static int qcom_pcie_resume_noirq(struct device *dev)
> +{
> +	struct qcom_pcie *pcie = dev_get_drvdata(dev);
> +
> +	if (pcie->suspended) {
> +		qcom_pcie_host_init(&pcie->pci->pp);

Looks like error handling is missing here.

> +		pcie->suspended = false;
> +	}
> +
> +	qcom_pcie_icc_update(pcie);
> +
> +	return 0;
> +}
> +
>  static const struct of_device_id qcom_pcie_match[] = {
>  	{ .compatible = "qcom,pcie-apq8064", .data = &cfg_2_1_0 },
>  	{ .compatible = "qcom,pcie-apq8084", .data = &cfg_1_0_0 },
> @@ -1856,12 +1904,17 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_QCOM, 0x0302, qcom_fixup_class);
>  DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_QCOM, 0x1000, qcom_fixup_class);
>  DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_QCOM, 0x1001, qcom_fixup_class);
>  
> +static const struct dev_pm_ops qcom_pcie_pm_ops = {
> +	NOIRQ_SYSTEM_SLEEP_PM_OPS(qcom_pcie_suspend_noirq, qcom_pcie_resume_noirq)
> +};
> +
>  static struct platform_driver qcom_pcie_driver = {
>  	.probe = qcom_pcie_probe,
>  	.driver = {
>  		.name = "qcom-pcie",
>  		.suppress_bind_attrs = true,
>  		.of_match_table = qcom_pcie_match,
> +		.pm = &qcom_pcie_pm_ops,
>  	},
>  };
>  builtin_platform_driver(qcom_pcie_driver);

Johan