[PATCH v9 1/2] dt-bindings: rockchip: Add DesignWare based PCIe controller

Subsystems: open firmware and flattened device tree bindings, pci native host bridge and endpoint drivers, pci subsystem, the rest

STALE1768d REVIEWED: 7 (6M)

2 review trailers (1 from subsystem maintainers).

8 messages, 5 authors, 2021-10-27 · open the first message on its own page

[PATCH v9 1/2] dt-bindings: rockchip: Add DesignWare based PCIe controller

From: Simon Xue <hidden>
Date: 2021-05-06 02:35:34

Document DT bindings for PCIe controller found on Rockchip SoC.

Reviewed-by: Rob Herring <robh@kernel.org>
Reviewed-by: Kever Yang <redacted>
Signed-off-by: Simon Xue <redacted>
---
 .../bindings/pci/rockchip-dw-pcie.yaml        | 141 ++++++++++++++++++
 1 file changed, 141 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/pci/rockchip-dw-pcie.yaml
diff --git a/Documentation/devicetree/bindings/pci/rockchip-dw-pcie.yaml b/Documentation/devicetree/bindings/pci/rockchip-dw-pcie.yaml
new file mode 100644
index 000000000000..142bbe577763
--- /dev/null
+++ b/Documentation/devicetree/bindings/pci/rockchip-dw-pcie.yaml
@@ -0,0 +1,141 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/pci/rockchip-dw-pcie.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: DesignWare based PCIe controller on Rockchip SoCs
+
+maintainers:
+  - Shawn Lin <shawn.lin@rock-chips.com>
+  - Simon Xue <xxm@rock-chips.com>
+  - Heiko Stuebner <heiko@sntech.de>
+
+description: |+
+  RK3568 SoC PCIe host controller is based on the Synopsys DesignWare
+  PCIe IP and thus inherits all the common properties defined in
+  designware-pcie.txt.
+
+allOf:
+  - $ref: /schemas/pci/pci-bus.yaml#
+
+# We need a select here so we don't match all nodes with 'snps,dw-pcie'
+select:
+  properties:
+    compatible:
+      contains:
+        const: rockchip,rk3568-pcie
+  required:
+    - compatible
+
+properties:
+  compatible:
+    items:
+      - const: rockchip,rk3568-pcie
+      - const: snps,dw-pcie
+
+  reg:
+    items:
+      - description: Data Bus Interface (DBI) registers
+      - description: Rockchip designed configuration registers
+      - description: Config registers
+
+  reg-names:
+    items:
+      - const: dbi
+      - const: apb
+      - const: config
+
+  clocks:
+    items:
+      - description: AHB clock for PCIe master
+      - description: AHB clock for PCIe slave
+      - description: AHB clock for PCIe dbi
+      - description: APB clock for PCIe
+      - description: Auxiliary clock for PCIe
+
+  clock-names:
+    items:
+      - const: aclk_mst
+      - const: aclk_slv
+      - const: aclk_dbi
+      - const: pclk
+      - const: aux
+
+  msi-map: true
+
+  num-lanes: true
+
+  phys:
+    maxItems: 1
+
+  phy-names:
+    const: pcie-phy
+
+  power-domains:
+    maxItems: 1
+
+  ranges:
+    maxItems: 2
+
+  resets:
+    maxItems: 1
+
+  reset-names:
+    const: pipe
+
+  vpcie3v3-supply: true
+
+required:
+  - compatible
+  - reg
+  - reg-names
+  - clocks
+  - clock-names
+  - msi-map
+  - num-lanes
+  - phys
+  - phy-names
+  - power-domains
+  - resets
+  - reset-names
+
+unevaluatedProperties: false
+
+examples:
+  - |
+
+    bus {
+        #address-cells = <2>;
+        #size-cells = <2>;
+
+        pcie3x2: pcie@fe280000 {
+            compatible = "rockchip,rk3568-pcie", "snps,dw-pcie";
+            reg = <0x3 0xc0800000 0x0 0x390000>,
+                  <0x0 0xfe280000 0x0 0x10000>,
+                  <0x3 0x80000000 0x0 0x100000>;
+            reg-names = "dbi", "apb", "config";
+            bus-range = <0x20 0x2f>;
+            clocks = <&cru 143>, <&cru 144>,
+                     <&cru 145>, <&cru 146>,
+                     <&cru 147>;
+            clock-names = "aclk_mst", "aclk_slv",
+                          "aclk_dbi", "pclk",
+                          "aux";
+            device_type = "pci";
+            linux,pci-domain = <2>;
+            max-link-speed = <2>;
+            msi-map = <0x2000 &its 0x2000 0x1000>;
+            num-lanes = <2>;
+            phys = <&pcie30phy>;
+            phy-names = "pcie-phy";
+            power-domains = <&power 15>;
+            ranges = <0x81000000 0x0 0x80800000 0x3 0x80800000 0x0 0x100000>,
+                     <0x83000000 0x0 0x80900000 0x3 0x80900000 0x0 0x3f700000>;
+            resets = <&cru 193>;
+            reset-names = "pipe";
+            #address-cells = <3>;
+            #size-cells = <2>;
+        };
+    };
+...
-- 
2.25.1


[PATCH v9 2/2] PCI: rockchip: Add Rockchip RK356X host controller driver

From: Simon Xue <hidden>
Date: 2021-05-06 02:36:35

Add a driver for the DesignWare-based PCIe controller found on
RK356X. The existing pcie-rockchip-host driver is only used for
the Rockchip-designed IP found on RK3399.

Tested-by: Peter Geis <redacted>
Reviewed-by: Kever Yang <redacted>
Reported-by: kernel test robot <redacted>
Signed-off-by: Simon Xue <redacted>
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
---
 drivers/pci/controller/dwc/Kconfig            |  11 +
 drivers/pci/controller/dwc/Makefile           |   1 +
 drivers/pci/controller/dwc/pcie-dw-rockchip.c | 277 ++++++++++++++++++
 3 files changed, 289 insertions(+)
 create mode 100644 drivers/pci/controller/dwc/pcie-dw-rockchip.c
diff --git a/drivers/pci/controller/dwc/Kconfig b/drivers/pci/controller/dwc/Kconfig
index 423d35872ce4..60d3dde9ca39 100644
--- a/drivers/pci/controller/dwc/Kconfig
+++ b/drivers/pci/controller/dwc/Kconfig
@@ -214,6 +214,17 @@ config PCIE_ARTPEC6_EP
 	  Enables support for the PCIe controller in the ARTPEC-6 SoC to work in
 	  endpoint mode. This uses the DesignWare core.
 
+config PCIE_ROCKCHIP_DW_HOST
+	bool "Rockchip DesignWare PCIe controller"
+	select PCIE_DW
+	select PCIE_DW_HOST
+	depends on PCI_MSI_IRQ_DOMAIN
+	depends on ARCH_ROCKCHIP || COMPILE_TEST
+	depends on OF
+	help
+	  Enables support for the DesignWare PCIe controller in the
+	  Rockchip SoC except RK3399.
+
 config PCIE_INTEL_GW
 	bool "Intel Gateway PCIe host controller support"
 	depends on OF && (X86 || COMPILE_TEST)
diff --git a/drivers/pci/controller/dwc/Makefile b/drivers/pci/controller/dwc/Makefile
index eca805c1a023..689b06a7d334 100644
--- a/drivers/pci/controller/dwc/Makefile
+++ b/drivers/pci/controller/dwc/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_PCI_LAYERSCAPE_EP) += pci-layerscape-ep.o
 obj-$(CONFIG_PCIE_QCOM) += pcie-qcom.o
 obj-$(CONFIG_PCIE_ARMADA_8K) += pcie-armada8k.o
 obj-$(CONFIG_PCIE_ARTPEC6) += pcie-artpec6.o
+obj-$(CONFIG_PCIE_ROCKCHIP_DW_HOST) += pcie-dw-rockchip.o
 obj-$(CONFIG_PCIE_INTEL_GW) += pcie-intel-gw.o
 obj-$(CONFIG_PCIE_KIRIN) += pcie-kirin.o
 obj-$(CONFIG_PCIE_HISI_STB) += pcie-histb.o
diff --git a/drivers/pci/controller/dwc/pcie-dw-rockchip.c b/drivers/pci/controller/dwc/pcie-dw-rockchip.c
new file mode 100644
index 000000000000..3f060144eeab
--- /dev/null
+++ b/drivers/pci/controller/dwc/pcie-dw-rockchip.c
@@ -0,0 +1,277 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * PCIe host controller driver for Rockchip SoCs.
+ *
+ * Copyright (C) 2021 Rockchip Electronics Co., Ltd.
+ *		http://www.rock-chips.com
+ *
+ * Author: Simon Xue <xxm@rock-chips.com>
+ */
+
+#include <linux/clk.h>
+#include <linux/gpio/consumer.h>
+#include <linux/mfd/syscon.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/phy/phy.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/reset.h>
+
+#include "pcie-designware.h"
+
+/*
+ * The upper 16 bits of PCIE_CLIENT_CONFIG are a write
+ * mask for the lower 16 bits.
+ */
+#define HIWORD_UPDATE(mask, val) (((mask) << 16) | (val))
+#define HIWORD_UPDATE_BIT(val)	HIWORD_UPDATE(val, val)
+
+#define to_rockchip_pcie(x) dev_get_drvdata((x)->dev)
+
+#define PCIE_CLIENT_RC_MODE		HIWORD_UPDATE_BIT(0x40)
+#define PCIE_CLIENT_ENABLE_LTSSM	HIWORD_UPDATE_BIT(0xc)
+#define PCIE_SMLH_LINKUP		BIT(16)
+#define PCIE_RDLH_LINKUP		BIT(17)
+#define PCIE_LINKUP			(PCIE_SMLH_LINKUP | PCIE_RDLH_LINKUP)
+#define PCIE_L0S_ENTRY			0x11
+#define PCIE_CLIENT_GENERAL_CONTROL	0x0
+#define PCIE_CLIENT_GENERAL_DEBUG	0x104
+#define PCIE_CLIENT_HOT_RESET_CTRL      0x180
+#define PCIE_CLIENT_LTSSM_STATUS	0x300
+#define PCIE_LTSSM_ENABLE_ENHANCE       BIT(4)
+
+struct rockchip_pcie {
+	struct dw_pcie			pci;
+	void __iomem			*apb_base;
+	struct phy			*phy;
+	struct clk_bulk_data		*clks;
+	unsigned int			clk_cnt;
+	struct reset_control		*rst;
+	struct gpio_desc		*rst_gpio;
+	struct regulator                *vpcie3v3;
+};
+
+static int rockchip_pcie_readl_apb(struct rockchip_pcie *rockchip,
+					     u32 reg)
+{
+	return readl(rockchip->apb_base + reg);
+}
+
+static void rockchip_pcie_writel_apb(struct rockchip_pcie *rockchip,
+						u32 val, u32 reg)
+{
+	writel(val, rockchip->apb_base + reg);
+}
+
+static void rockchip_pcie_enable_ltssm(struct rockchip_pcie *rockchip)
+{
+	rockchip_pcie_writel_apb(rockchip, PCIE_CLIENT_ENABLE_LTSSM,
+				 PCIE_CLIENT_GENERAL_CONTROL);
+}
+
+static int rockchip_pcie_link_up(struct dw_pcie *pci)
+{
+	struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+	u32 val = rockchip_pcie_readl_apb(rockchip, PCIE_CLIENT_LTSSM_STATUS);
+
+	if ((val & (PCIE_RDLH_LINKUP | PCIE_SMLH_LINKUP)) == PCIE_LINKUP &&
+	    (val & GENMASK(5, 0)) == PCIE_L0S_ENTRY)
+		return 1;
+
+	return 0;
+}
+
+static int rockchip_pcie_start_link(struct dw_pcie *pci)
+{
+	struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+
+	/* Reset device */
+	gpiod_set_value_cansleep(rockchip->rst_gpio, 0);
+
+	rockchip_pcie_enable_ltssm(rockchip);
+
+	/*
+	 * PCIe requires the refclk to be stable for 100µs prior to releasing
+	 * PERST. See table 2-4 in section 2.6.2 AC Specifications of the PCI
+	 * Express Card Electromechanical Specification, 1.1. However, we don't
+	 * know if the refclk is coming from RC's PHY or external OSC. If it's
+	 * from RC, so enabling LTSSM is the just right place to release #PERST.
+	 * We need more extra time as before, rather than setting just
+	 * 100us as we don't know how long should the device need to reset.
+	 */
+	msleep(100);
+	gpiod_set_value_cansleep(rockchip->rst_gpio, 1);
+
+	return 0;
+}
+
+static int rockchip_pcie_host_init(struct pcie_port *pp)
+{
+	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
+	struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+	u32 val;
+
+	/* LTSSM enable control mode */
+	val = rockchip_pcie_readl_apb(rockchip, PCIE_CLIENT_HOT_RESET_CTRL);
+	val |= PCIE_LTSSM_ENABLE_ENHANCE | (PCIE_LTSSM_ENABLE_ENHANCE << 16);
+	rockchip_pcie_writel_apb(rockchip, val, PCIE_CLIENT_HOT_RESET_CTRL);
+
+	rockchip_pcie_writel_apb(rockchip, PCIE_CLIENT_RC_MODE,
+				 PCIE_CLIENT_GENERAL_CONTROL);
+
+	return 0;
+}
+
+static const struct dw_pcie_host_ops rockchip_pcie_host_ops = {
+	.host_init = rockchip_pcie_host_init,
+};
+
+static int rockchip_pcie_clk_init(struct rockchip_pcie *rockchip)
+{
+	struct device *dev = rockchip->pci.dev;
+	int ret;
+
+	ret = devm_clk_bulk_get_all(dev, &rockchip->clks);
+	if (ret < 0)
+		return ret;
+
+	rockchip->clk_cnt = ret;
+
+	return clk_bulk_prepare_enable(rockchip->clk_cnt, rockchip->clks);
+}
+
+static int rockchip_pcie_resource_get(struct platform_device *pdev,
+				      struct rockchip_pcie *rockchip)
+{
+	rockchip->apb_base = devm_platform_ioremap_resource_byname(pdev, "apb");
+	if (IS_ERR(rockchip->apb_base))
+		return PTR_ERR(rockchip->apb_base);
+
+	rockchip->rst_gpio = devm_gpiod_get_optional(&pdev->dev, "reset",
+						     GPIOD_OUT_HIGH);
+	if (IS_ERR(rockchip->rst_gpio))
+		return PTR_ERR(rockchip->rst_gpio);
+
+	return 0;
+}
+
+static int rockchip_pcie_phy_init(struct rockchip_pcie *rockchip)
+{
+	struct device *dev = rockchip->pci.dev;
+	int ret;
+
+	rockchip->phy = devm_phy_get(dev, "pcie-phy");
+	if (IS_ERR(rockchip->phy))
+		return dev_err_probe(dev, PTR_ERR(rockchip->phy),
+				     "missing PHY\n");
+
+	ret = phy_init(rockchip->phy);
+	if (ret < 0)
+		return ret;
+
+	ret = phy_power_on(rockchip->phy);
+	if (ret)
+		phy_exit(rockchip->phy);
+
+	return ret;
+}
+
+static void rockchip_pcie_phy_deinit(struct rockchip_pcie *rockchip)
+{
+	phy_exit(rockchip->phy);
+	phy_power_off(rockchip->phy);
+}
+
+static int rockchip_pcie_reset_control_release(struct rockchip_pcie *rockchip)
+{
+	struct device *dev = rockchip->pci.dev;
+
+	rockchip->rst = devm_reset_control_array_get_exclusive(dev);
+	if (IS_ERR(rockchip->rst))
+		return dev_err_probe(dev, PTR_ERR(rockchip->rst),
+				     "failed to get reset lines\n");
+
+	return reset_control_deassert(rockchip->rst);
+}
+
+static const struct dw_pcie_ops dw_pcie_ops = {
+	.link_up = rockchip_pcie_link_up,
+	.start_link = rockchip_pcie_start_link,
+};
+
+static int rockchip_pcie_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct rockchip_pcie *rockchip;
+	struct pcie_port *pp;
+	int ret;
+
+	rockchip = devm_kzalloc(dev, sizeof(*rockchip), GFP_KERNEL);
+	if (!rockchip)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, rockchip);
+
+	rockchip->pci.dev = dev;
+	rockchip->pci.ops = &dw_pcie_ops;
+
+	pp = &rockchip->pci.pp;
+	pp->ops = &rockchip_pcie_host_ops;
+
+	ret = rockchip_pcie_resource_get(pdev, rockchip);
+	if (ret)
+		return ret;
+
+	/* DON'T MOVE ME: must be enable before PHY init */
+	rockchip->vpcie3v3 = devm_regulator_get_optional(dev, "vpcie3v3");
+	if (IS_ERR(rockchip->vpcie3v3))
+		if (PTR_ERR(rockchip->vpcie3v3) != -ENODEV)
+			return dev_err_probe(dev, PTR_ERR(rockchip->vpcie3v3),
+					"failed to get vpcie3v3 regulator\n");
+
+	ret = regulator_enable(rockchip->vpcie3v3);
+	if (ret) {
+		dev_err(dev, "failed to enable vpcie3v3 regulator\n");
+		return ret;
+	}
+
+	ret = rockchip_pcie_phy_init(rockchip);
+	if (ret)
+		goto disable_regulator;
+
+	ret = rockchip_pcie_reset_control_release(rockchip);
+	if (ret)
+		goto deinit_phy;
+
+	ret = rockchip_pcie_clk_init(rockchip);
+	if (ret)
+		goto deinit_phy;
+
+	ret = dw_pcie_host_init(pp);
+	if (!ret)
+		return 0;
+
+	clk_bulk_disable_unprepare(rockchip->clk_cnt, rockchip->clks);
+deinit_phy:
+	rockchip_pcie_phy_deinit(rockchip);
+disable_regulator:
+	regulator_disable(rockchip->vpcie3v3);
+
+	return ret;
+}
+
+static const struct of_device_id rockchip_pcie_of_match[] = {
+	{ .compatible = "rockchip,rk3568-pcie", },
+	{},
+};
+
+static struct platform_driver rockchip_pcie_driver = {
+	.driver = {
+		.name	= "rockchip-dw-pcie",
+		.of_match_table = rockchip_pcie_of_match,
+		.suppress_bind_attrs = true,
+	},
+	.probe = rockchip_pcie_probe,
+};
+builtin_platform_driver(rockchip_pcie_driver);
-- 
2.25.1


Re: [PATCH v9 2/2] PCI: rockchip: Add Rockchip RK356X host controller driver

From: Lorenzo Pieralisi <hidden>
Date: 2021-06-23 14:33:42

[+Pali]

On Thu, May 06, 2021 at 10:35:44AM +0800, Simon Xue wrote:
Add a driver for the DesignWare-based PCIe controller found on
RK356X. The existing pcie-rockchip-host driver is only used for
the Rockchip-designed IP found on RK3399.

Tested-by: Peter Geis <redacted>
Reviewed-by: Kever Yang <redacted>
Reported-by: kernel test robot <redacted>
I will remove this tag - it does not make sense on a patch adding
a new driver.

[...]
+static int rockchip_pcie_start_link(struct dw_pcie *pci)
+{
+	struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+
+	/* Reset device */
+	gpiod_set_value_cansleep(rockchip->rst_gpio, 0);
+
+	rockchip_pcie_enable_ltssm(rockchip);
+
+	/*
+	 * PCIe requires the refclk to be stable for 100µs prior to releasing
+	 * PERST. See table 2-4 in section 2.6.2 AC Specifications of the PCI
+	 * Express Card Electromechanical Specification, 1.1. However, we don't
+	 * know if the refclk is coming from RC's PHY or external OSC. If it's
+	 * from RC, so enabling LTSSM is the just right place to release #PERST.
+	 * We need more extra time as before, rather than setting just
+	 * 100us as we don't know how long should the device need to reset.
+	 */
+	msleep(100);
Any rationale behind the time chosen ?

Ongoing discussion:

https://lore.kernel.org/linux-pci/20210531090540.2663171-1-luca@lucaceresoli.net

Lorenzo
+	gpiod_set_value_cansleep(rockchip->rst_gpio, 1);
+
+	return 0;
+}
+
+static int rockchip_pcie_host_init(struct pcie_port *pp)
+{
+	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
+	struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+	u32 val;
+
+	/* LTSSM enable control mode */
+	val = rockchip_pcie_readl_apb(rockchip, PCIE_CLIENT_HOT_RESET_CTRL);
+	val |= PCIE_LTSSM_ENABLE_ENHANCE | (PCIE_LTSSM_ENABLE_ENHANCE << 16);
+	rockchip_pcie_writel_apb(rockchip, val, PCIE_CLIENT_HOT_RESET_CTRL);
+
+	rockchip_pcie_writel_apb(rockchip, PCIE_CLIENT_RC_MODE,
+				 PCIE_CLIENT_GENERAL_CONTROL);
+
+	return 0;
+}
+
+static const struct dw_pcie_host_ops rockchip_pcie_host_ops = {
+	.host_init = rockchip_pcie_host_init,
+};
+
+static int rockchip_pcie_clk_init(struct rockchip_pcie *rockchip)
+{
+	struct device *dev = rockchip->pci.dev;
+	int ret;
+
+	ret = devm_clk_bulk_get_all(dev, &rockchip->clks);
+	if (ret < 0)
+		return ret;
+
+	rockchip->clk_cnt = ret;
+
+	return clk_bulk_prepare_enable(rockchip->clk_cnt, rockchip->clks);
+}
+
+static int rockchip_pcie_resource_get(struct platform_device *pdev,
+				      struct rockchip_pcie *rockchip)
+{
+	rockchip->apb_base = devm_platform_ioremap_resource_byname(pdev, "apb");
+	if (IS_ERR(rockchip->apb_base))
+		return PTR_ERR(rockchip->apb_base);
+
+	rockchip->rst_gpio = devm_gpiod_get_optional(&pdev->dev, "reset",
+						     GPIOD_OUT_HIGH);
+	if (IS_ERR(rockchip->rst_gpio))
+		return PTR_ERR(rockchip->rst_gpio);
+
+	return 0;
+}
+
+static int rockchip_pcie_phy_init(struct rockchip_pcie *rockchip)
+{
+	struct device *dev = rockchip->pci.dev;
+	int ret;
+
+	rockchip->phy = devm_phy_get(dev, "pcie-phy");
+	if (IS_ERR(rockchip->phy))
+		return dev_err_probe(dev, PTR_ERR(rockchip->phy),
+				     "missing PHY\n");
+
+	ret = phy_init(rockchip->phy);
+	if (ret < 0)
+		return ret;
+
+	ret = phy_power_on(rockchip->phy);
+	if (ret)
+		phy_exit(rockchip->phy);
+
+	return ret;
+}
+
+static void rockchip_pcie_phy_deinit(struct rockchip_pcie *rockchip)
+{
+	phy_exit(rockchip->phy);
+	phy_power_off(rockchip->phy);
+}
+
+static int rockchip_pcie_reset_control_release(struct rockchip_pcie *rockchip)
+{
+	struct device *dev = rockchip->pci.dev;
+
+	rockchip->rst = devm_reset_control_array_get_exclusive(dev);
+	if (IS_ERR(rockchip->rst))
+		return dev_err_probe(dev, PTR_ERR(rockchip->rst),
+				     "failed to get reset lines\n");
+
+	return reset_control_deassert(rockchip->rst);
+}
+
+static const struct dw_pcie_ops dw_pcie_ops = {
+	.link_up = rockchip_pcie_link_up,
+	.start_link = rockchip_pcie_start_link,
+};
+
+static int rockchip_pcie_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct rockchip_pcie *rockchip;
+	struct pcie_port *pp;
+	int ret;
+
+	rockchip = devm_kzalloc(dev, sizeof(*rockchip), GFP_KERNEL);
+	if (!rockchip)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, rockchip);
+
+	rockchip->pci.dev = dev;
+	rockchip->pci.ops = &dw_pcie_ops;
+
+	pp = &rockchip->pci.pp;
+	pp->ops = &rockchip_pcie_host_ops;
+
+	ret = rockchip_pcie_resource_get(pdev, rockchip);
+	if (ret)
+		return ret;
+
+	/* DON'T MOVE ME: must be enable before PHY init */
+	rockchip->vpcie3v3 = devm_regulator_get_optional(dev, "vpcie3v3");
+	if (IS_ERR(rockchip->vpcie3v3))
+		if (PTR_ERR(rockchip->vpcie3v3) != -ENODEV)
+			return dev_err_probe(dev, PTR_ERR(rockchip->vpcie3v3),
+					"failed to get vpcie3v3 regulator\n");
+
+	ret = regulator_enable(rockchip->vpcie3v3);
+	if (ret) {
+		dev_err(dev, "failed to enable vpcie3v3 regulator\n");
+		return ret;
+	}
+
+	ret = rockchip_pcie_phy_init(rockchip);
+	if (ret)
+		goto disable_regulator;
+
+	ret = rockchip_pcie_reset_control_release(rockchip);
+	if (ret)
+		goto deinit_phy;
+
+	ret = rockchip_pcie_clk_init(rockchip);
+	if (ret)
+		goto deinit_phy;
+
+	ret = dw_pcie_host_init(pp);
+	if (!ret)
+		return 0;
+
+	clk_bulk_disable_unprepare(rockchip->clk_cnt, rockchip->clks);
+deinit_phy:
+	rockchip_pcie_phy_deinit(rockchip);
+disable_regulator:
+	regulator_disable(rockchip->vpcie3v3);
+
+	return ret;
+}
+
+static const struct of_device_id rockchip_pcie_of_match[] = {
+	{ .compatible = "rockchip,rk3568-pcie", },
+	{},
+};
+
+static struct platform_driver rockchip_pcie_driver = {
+	.driver = {
+		.name	= "rockchip-dw-pcie",
+		.of_match_table = rockchip_pcie_of_match,
+		.suppress_bind_attrs = true,
+	},
+	.probe = rockchip_pcie_probe,
+};
+builtin_platform_driver(rockchip_pcie_driver);
-- 
2.25.1

Re: [PATCH v9 2/2] PCI: rockchip: Add Rockchip RK356X host controller driver

From: Rob Herring <robh+dt@kernel.org>
Date: 2021-06-23 21:15:19

 == On Wed, May 5, 2021 at 8:35 PM Simon Xue [off-list ref] wrote:
quoted hunk
Add a driver for the DesignWare-based PCIe controller found on
RK356X. The existing pcie-rockchip-host driver is only used for
the Rockchip-designed IP found on RK3399.

Tested-by: Peter Geis <redacted>
Reviewed-by: Kever Yang <redacted>
Reported-by: kernel test robot <redacted>
Signed-off-by: Simon Xue <redacted>
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
---
 drivers/pci/controller/dwc/Kconfig            |  11 +
 drivers/pci/controller/dwc/Makefile           |   1 +
 drivers/pci/controller/dwc/pcie-dw-rockchip.c | 277 ++++++++++++++++++
 3 files changed, 289 insertions(+)
 create mode 100644 drivers/pci/controller/dwc/pcie-dw-rockchip.c
diff --git a/drivers/pci/controller/dwc/Kconfig b/drivers/pci/controller/dwc/Kconfig
index 423d35872ce4..60d3dde9ca39 100644
--- a/drivers/pci/controller/dwc/Kconfig
+++ b/drivers/pci/controller/dwc/Kconfig
@@ -214,6 +214,17 @@ config PCIE_ARTPEC6_EP
          Enables support for the PCIe controller in the ARTPEC-6 SoC to work in
          endpoint mode. This uses the DesignWare core.

+config PCIE_ROCKCHIP_DW_HOST
+       bool "Rockchip DesignWare PCIe controller"
+       select PCIE_DW
+       select PCIE_DW_HOST
+       depends on PCI_MSI_IRQ_DOMAIN
+       depends on ARCH_ROCKCHIP || COMPILE_TEST
+       depends on OF
+       help
+         Enables support for the DesignWare PCIe controller in the
+         Rockchip SoC except RK3399.
+
 config PCIE_INTEL_GW
        bool "Intel Gateway PCIe host controller support"
        depends on OF && (X86 || COMPILE_TEST)
diff --git a/drivers/pci/controller/dwc/Makefile b/drivers/pci/controller/dwc/Makefile
index eca805c1a023..689b06a7d334 100644
--- a/drivers/pci/controller/dwc/Makefile
+++ b/drivers/pci/controller/dwc/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_PCI_LAYERSCAPE_EP) += pci-layerscape-ep.o
 obj-$(CONFIG_PCIE_QCOM) += pcie-qcom.o
 obj-$(CONFIG_PCIE_ARMADA_8K) += pcie-armada8k.o
 obj-$(CONFIG_PCIE_ARTPEC6) += pcie-artpec6.o
+obj-$(CONFIG_PCIE_ROCKCHIP_DW_HOST) += pcie-dw-rockchip.o
 obj-$(CONFIG_PCIE_INTEL_GW) += pcie-intel-gw.o
 obj-$(CONFIG_PCIE_KIRIN) += pcie-kirin.o
 obj-$(CONFIG_PCIE_HISI_STB) += pcie-histb.o
diff --git a/drivers/pci/controller/dwc/pcie-dw-rockchip.c b/drivers/pci/controller/dwc/pcie-dw-rockchip.c
new file mode 100644
index 000000000000..3f060144eeab
--- /dev/null
+++ b/drivers/pci/controller/dwc/pcie-dw-rockchip.c
@@ -0,0 +1,277 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * PCIe host controller driver for Rockchip SoCs.
+ *
+ * Copyright (C) 2021 Rockchip Electronics Co., Ltd.
+ *             http://www.rock-chips.com
+ *
+ * Author: Simon Xue <xxm@rock-chips.com>
+ */
+
+#include <linux/clk.h>
+#include <linux/gpio/consumer.h>
+#include <linux/mfd/syscon.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/phy/phy.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/reset.h>
+
+#include "pcie-designware.h"
+
+/*
+ * The upper 16 bits of PCIE_CLIENT_CONFIG are a write
+ * mask for the lower 16 bits.
+ */
+#define HIWORD_UPDATE(mask, val) (((mask) << 16) | (val))
+#define HIWORD_UPDATE_BIT(val) HIWORD_UPDATE(val, val)
+
+#define to_rockchip_pcie(x) dev_get_drvdata((x)->dev)
+
+#define PCIE_CLIENT_RC_MODE            HIWORD_UPDATE_BIT(0x40)
+#define PCIE_CLIENT_ENABLE_LTSSM       HIWORD_UPDATE_BIT(0xc)
+#define PCIE_SMLH_LINKUP               BIT(16)
+#define PCIE_RDLH_LINKUP               BIT(17)
+#define PCIE_LINKUP                    (PCIE_SMLH_LINKUP | PCIE_RDLH_LINKUP)
+#define PCIE_L0S_ENTRY                 0x11
+#define PCIE_CLIENT_GENERAL_CONTROL    0x0
+#define PCIE_CLIENT_GENERAL_DEBUG      0x104
+#define PCIE_CLIENT_HOT_RESET_CTRL      0x180
+#define PCIE_CLIENT_LTSSM_STATUS       0x300
+#define PCIE_LTSSM_ENABLE_ENHANCE       BIT(4)
+
+struct rockchip_pcie {
+       struct dw_pcie                  pci;
+       void __iomem                    *apb_base;
+       struct phy                      *phy;
+       struct clk_bulk_data            *clks;
+       unsigned int                    clk_cnt;
+       struct reset_control            *rst;
+       struct gpio_desc                *rst_gpio;
+       struct regulator                *vpcie3v3;
+};
+
+static int rockchip_pcie_readl_apb(struct rockchip_pcie *rockchip,
+                                            u32 reg)
+{
+       return readl(rockchip->apb_base + reg);
Use _relaxed variant.
+}
+
+static void rockchip_pcie_writel_apb(struct rockchip_pcie *rockchip,
+                                               u32 val, u32 reg)
+{
+       writel(val, rockchip->apb_base + reg);
Use _relaxed variant.
+}
+
+static void rockchip_pcie_enable_ltssm(struct rockchip_pcie *rockchip)
+{
+       rockchip_pcie_writel_apb(rockchip, PCIE_CLIENT_ENABLE_LTSSM,
+                                PCIE_CLIENT_GENERAL_CONTROL);
+}
+
+static int rockchip_pcie_link_up(struct dw_pcie *pci)
+{
+       struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+       u32 val = rockchip_pcie_readl_apb(rockchip, PCIE_CLIENT_LTSSM_STATUS);
+
+       if ((val & (PCIE_RDLH_LINKUP | PCIE_SMLH_LINKUP)) == PCIE_LINKUP &&
((val & PCIE_LINKUP) == PCIE_LINKUP)
+           (val & GENMASK(5, 0)) == PCIE_L0S_ENTRY)
Perhaps a define for GENMASK(5, 0)
+               return 1;
+
+       return 0;
+}
+
+static int rockchip_pcie_start_link(struct dw_pcie *pci)
+{
+       struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+
+       /* Reset device */
+       gpiod_set_value_cansleep(rockchip->rst_gpio, 0);
With gpiod, 0 means inactive which is reset deasserted. Your DT should
have an ACTIVE_LOW flag.
+
+       rockchip_pcie_enable_ltssm(rockchip);
+
+       /*
+        * PCIe requires the refclk to be stable for 100µs prior to releasing
+        * PERST. See table 2-4 in section 2.6.2 AC Specifications of the PCI
+        * Express Card Electromechanical Specification, 1.1. However, we don't
+        * know if the refclk is coming from RC's PHY or external OSC. If it's
+        * from RC, so enabling LTSSM is the just right place to release #PERST.
+        * We need more extra time as before, rather than setting just
+        * 100us as we don't know how long should the device need to reset.
+        */
+       msleep(100);
+       gpiod_set_value_cansleep(rockchip->rst_gpio, 1);
+
+       return 0;
+}
+
+static int rockchip_pcie_host_init(struct pcie_port *pp)
+{
+       struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
+       struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+       u32 val;
+
+       /* LTSSM enable control mode */
+       val = rockchip_pcie_readl_apb(rockchip, PCIE_CLIENT_HOT_RESET_CTRL);
+       val |= PCIE_LTSSM_ENABLE_ENHANCE | (PCIE_LTSSM_ENABLE_ENHANCE << 16);
Since you have a write mask, why do you need to do a read-mod-write?
And why not use HIWORD_UPDATE_BIT?
+       rockchip_pcie_writel_apb(rockchip, val, PCIE_CLIENT_HOT_RESET_CTRL);
+
+       rockchip_pcie_writel_apb(rockchip, PCIE_CLIENT_RC_MODE,
+                                PCIE_CLIENT_GENERAL_CONTROL);
+
+       return 0;
+}

Re: [PATCH v9 2/2] PCI: rockchip: Add Rockchip RK356X host controller driver

From: xxm <hidden>
Date: 2021-06-24 02:16:03

Hi,

在 2021/6/23 22:33, Lorenzo Pieralisi 写道:
[+Pali]

On Thu, May 06, 2021 at 10:35:44AM +0800, Simon Xue wrote:
quoted
Add a driver for the DesignWare-based PCIe controller found on
RK356X. The existing pcie-rockchip-host driver is only used for
the Rockchip-designed IP found on RK3399.

Tested-by: Peter Geis <redacted>
Reviewed-by: Kever Yang <redacted>
Reported-by: kernel test robot <redacted>
I will remove this tag - it does not make sense on a patch adding
a new driver.

[...]
quoted
+static int rockchip_pcie_start_link(struct dw_pcie *pci)
+{
+	struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+
+	/* Reset device */
+	gpiod_set_value_cansleep(rockchip->rst_gpio, 0);
+
+	rockchip_pcie_enable_ltssm(rockchip);
+
+	/*
+	 * PCIe requires the refclk to be stable for 100µs prior to releasing
+	 * PERST. See table 2-4 in section 2.6.2 AC Specifications of the PCI
+	 * Express Card Electromechanical Specification, 1.1. However, we don't
+	 * know if the refclk is coming from RC's PHY or external OSC. If it's
+	 * from RC, so enabling LTSSM is the just right place to release #PERST.
+	 * We need more extra time as before, rather than setting just
+	 * 100us as we don't know how long should the device need to reset.
+	 */
+	msleep(100);
Any rationale behind the time chosen ?
We found some device need about 30ms, so 100ms here just leave more room 
for other devices.
Ongoing discussion:

https://lore.kernel.org/linux-pci/20210531090540.2663171-1-luca@lucaceresoli.net

Lorenzo
quoted
+	gpiod_set_value_cansleep(rockchip->rst_gpio, 1);
+
+	return 0;
+}
+
+static int rockchip_pcie_host_init(struct pcie_port *pp)
+{
+	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
+	struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+	u32 val;
+
+	/* LTSSM enable control mode */
+	val = rockchip_pcie_readl_apb(rockchip, PCIE_CLIENT_HOT_RESET_CTRL);
+	val |= PCIE_LTSSM_ENABLE_ENHANCE | (PCIE_LTSSM_ENABLE_ENHANCE << 16);
+	rockchip_pcie_writel_apb(rockchip, val, PCIE_CLIENT_HOT_RESET_CTRL);
+
+	rockchip_pcie_writel_apb(rockchip, PCIE_CLIENT_RC_MODE,
+				 PCIE_CLIENT_GENERAL_CONTROL);
+
+	return 0;
+}
+
+static const struct dw_pcie_host_ops rockchip_pcie_host_ops = {
+	.host_init = rockchip_pcie_host_init,
+};
+
+static int rockchip_pcie_clk_init(struct rockchip_pcie *rockchip)
+{
+	struct device *dev = rockchip->pci.dev;
+	int ret;
+
+	ret = devm_clk_bulk_get_all(dev, &rockchip->clks);
+	if (ret < 0)
+		return ret;
+
+	rockchip->clk_cnt = ret;
+
+	return clk_bulk_prepare_enable(rockchip->clk_cnt, rockchip->clks);
+}
+
+static int rockchip_pcie_resource_get(struct platform_device *pdev,
+				      struct rockchip_pcie *rockchip)
+{
+	rockchip->apb_base = devm_platform_ioremap_resource_byname(pdev, "apb");
+	if (IS_ERR(rockchip->apb_base))
+		return PTR_ERR(rockchip->apb_base);
+
+	rockchip->rst_gpio = devm_gpiod_get_optional(&pdev->dev, "reset",
+						     GPIOD_OUT_HIGH);
+	if (IS_ERR(rockchip->rst_gpio))
+		return PTR_ERR(rockchip->rst_gpio);
+
+	return 0;
+}
+
+static int rockchip_pcie_phy_init(struct rockchip_pcie *rockchip)
+{
+	struct device *dev = rockchip->pci.dev;
+	int ret;
+
+	rockchip->phy = devm_phy_get(dev, "pcie-phy");
+	if (IS_ERR(rockchip->phy))
+		return dev_err_probe(dev, PTR_ERR(rockchip->phy),
+				     "missing PHY\n");
+
+	ret = phy_init(rockchip->phy);
+	if (ret < 0)
+		return ret;
+
+	ret = phy_power_on(rockchip->phy);
+	if (ret)
+		phy_exit(rockchip->phy);
+
+	return ret;
+}
+
+static void rockchip_pcie_phy_deinit(struct rockchip_pcie *rockchip)
+{
+	phy_exit(rockchip->phy);
+	phy_power_off(rockchip->phy);
+}
+
+static int rockchip_pcie_reset_control_release(struct rockchip_pcie *rockchip)
+{
+	struct device *dev = rockchip->pci.dev;
+
+	rockchip->rst = devm_reset_control_array_get_exclusive(dev);
+	if (IS_ERR(rockchip->rst))
+		return dev_err_probe(dev, PTR_ERR(rockchip->rst),
+				     "failed to get reset lines\n");
+
+	return reset_control_deassert(rockchip->rst);
+}
+
+static const struct dw_pcie_ops dw_pcie_ops = {
+	.link_up = rockchip_pcie_link_up,
+	.start_link = rockchip_pcie_start_link,
+};
+
+static int rockchip_pcie_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct rockchip_pcie *rockchip;
+	struct pcie_port *pp;
+	int ret;
+
+	rockchip = devm_kzalloc(dev, sizeof(*rockchip), GFP_KERNEL);
+	if (!rockchip)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, rockchip);
+
+	rockchip->pci.dev = dev;
+	rockchip->pci.ops = &dw_pcie_ops;
+
+	pp = &rockchip->pci.pp;
+	pp->ops = &rockchip_pcie_host_ops;
+
+	ret = rockchip_pcie_resource_get(pdev, rockchip);
+	if (ret)
+		return ret;
+
+	/* DON'T MOVE ME: must be enable before PHY init */
+	rockchip->vpcie3v3 = devm_regulator_get_optional(dev, "vpcie3v3");
+	if (IS_ERR(rockchip->vpcie3v3))
+		if (PTR_ERR(rockchip->vpcie3v3) != -ENODEV)
+			return dev_err_probe(dev, PTR_ERR(rockchip->vpcie3v3),
+					"failed to get vpcie3v3 regulator\n");
+
+	ret = regulator_enable(rockchip->vpcie3v3);
+	if (ret) {
+		dev_err(dev, "failed to enable vpcie3v3 regulator\n");
+		return ret;
+	}
+
+	ret = rockchip_pcie_phy_init(rockchip);
+	if (ret)
+		goto disable_regulator;
+
+	ret = rockchip_pcie_reset_control_release(rockchip);
+	if (ret)
+		goto deinit_phy;
+
+	ret = rockchip_pcie_clk_init(rockchip);
+	if (ret)
+		goto deinit_phy;
+
+	ret = dw_pcie_host_init(pp);
+	if (!ret)
+		return 0;
+
+	clk_bulk_disable_unprepare(rockchip->clk_cnt, rockchip->clks);
+deinit_phy:
+	rockchip_pcie_phy_deinit(rockchip);
+disable_regulator:
+	regulator_disable(rockchip->vpcie3v3);
+
+	return ret;
+}
+
+static const struct of_device_id rockchip_pcie_of_match[] = {
+	{ .compatible = "rockchip,rk3568-pcie", },
+	{},
+};
+
+static struct platform_driver rockchip_pcie_driver = {
+	.driver = {
+		.name	= "rockchip-dw-pcie",
+		.of_match_table = rockchip_pcie_of_match,
+		.suppress_bind_attrs = true,
+	},
+	.probe = rockchip_pcie_probe,
+};
+builtin_platform_driver(rockchip_pcie_driver);
-- 
2.25.1


Re: [PATCH v9 2/2] PCI: rockchip: Add Rockchip RK356X host controller driver

From: Pali Rohár <pali@kernel.org>
Date: 2021-06-24 08:23:18

On Thursday 24 June 2021 10:08:54 xxm wrote:
在 2021/6/23 22:33, Lorenzo Pieralisi 写道:
quoted
On Thu, May 06, 2021 at 10:35:44AM +0800, Simon Xue wrote:
quoted
+static int rockchip_pcie_start_link(struct dw_pcie *pci)
+{
+	struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+
+	/* Reset device */
+	gpiod_set_value_cansleep(rockchip->rst_gpio, 0);
+
+	rockchip_pcie_enable_ltssm(rockchip);
+
+	/*
+	 * PCIe requires the refclk to be stable for 100µs prior to releasing
+	 * PERST. See table 2-4 in section 2.6.2 AC Specifications of the PCI
+	 * Express Card Electromechanical Specification, 1.1. However, we don't
+	 * know if the refclk is coming from RC's PHY or external OSC. If it's
+	 * from RC, so enabling LTSSM is the just right place to release #PERST.
+	 * We need more extra time as before, rather than setting just
+	 * 100us as we don't know how long should the device need to reset.
+	 */
+	msleep(100);
Any rationale behind the time chosen ?
We found some device need about 30ms, so 100ms here just leave more room for
other devices.
Can you share information which PCIe card needs 30ms?

Last year I did tests with more WiFi AC cards and "the slowest" one was
Compex WLE1216 which needed about 11ms (more than 10ms). All other cards
were happy with just 1-2ms.

Re: [PATCH v9 2/2] PCI: rockchip: Add Rockchip RK356X host controller driver

From: xxm <hidden>
Date: 2021-06-24 08:58:50

Hi,

在 2021/6/24 16:23, Pali Rohár 写道:
On Thursday 24 June 2021 10:08:54 xxm wrote:
quoted
在 2021/6/23 22:33, Lorenzo Pieralisi 写道:
quoted
On Thu, May 06, 2021 at 10:35:44AM +0800, Simon Xue wrote:
quoted
+static int rockchip_pcie_start_link(struct dw_pcie *pci)
+{
+	struct rockchip_pcie *rockchip = to_rockchip_pcie(pci);
+
+	/* Reset device */
+	gpiod_set_value_cansleep(rockchip->rst_gpio, 0);
+
+	rockchip_pcie_enable_ltssm(rockchip);
+
+	/*
+	 * PCIe requires the refclk to be stable for 100µs prior to releasing
+	 * PERST. See table 2-4 in section 2.6.2 AC Specifications of the PCI
+	 * Express Card Electromechanical Specification, 1.1. However, we don't
+	 * know if the refclk is coming from RC's PHY or external OSC. If it's
+	 * from RC, so enabling LTSSM is the just right place to release #PERST.
+	 * We need more extra time as before, rather than setting just
+	 * 100us as we don't know how long should the device need to reset.
+	 */
+	msleep(100);
Any rationale behind the time chosen ?
We found some device need about 30ms, so 100ms here just leave more room for
other devices.
Can you share information which PCIe card needs 30ms?

Last year I did tests with more WiFi AC cards and "the slowest" one was
Compex WLE1216 which needed about 11ms (more than 10ms). All other cards
were happy with just 1-2ms.
Sorry, it was about 5 years ago, I can't find the specific card now.

By the way, we also take other's upstream code "msleep(100)" as a reference.

Re: [PATCH v9 1/2] dt-bindings: rockchip: Add DesignWare based PCIe controller

From: Rob Herring <robh@kernel.org>
Date: 2021-10-27 15:41:25

On Thu, May 06, 2021 at 10:34:48AM +0800, Simon Xue wrote:
Document DT bindings for PCIe controller found on Rockchip SoC.

Reviewed-by: Rob Herring <robh@kernel.org>
Reviewed-by: Kever Yang <redacted>
Signed-off-by: Simon Xue <redacted>
---
 .../bindings/pci/rockchip-dw-pcie.yaml        | 141 ++++++++++++++++++
 1 file changed, 141 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/pci/rockchip-dw-pcie.yaml
I don't know why this didn't get sent with subsequent versions, but it 
should have. I've applied it now.

Rob
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help