[PATCH v2 0/5] riscv: spacemit: Add PCIe RC controller support for K3

[PATCH v2 0/5] riscv: spacemit: Add PCIe RC controller support for K3

[Inochi Amaoto]

The PCIe controller on Spacemit K3 is almost a standard Synopsys
Designware PCIe IP with extra control and external MSI controller
(IMSIC).

Add binding and driver support for PCIe RC controller support on K3.

Changed from v1:
patch 1:
1. Remove post_init callback.
patch 2:
1. Fix devm_kmalloc_array arguments order and error check.
patch 3:
1. Apply Rob's tag.
2. Change check from oneOf to anyOf.
3. Fix spelling mistakes in commit message.
patch 4:
1. Add comment about PHY handle.
2. Remove interrupt "app".
3. Fix spelling mistakes in commit message.
patch 5:
1. Remove all logic related to link up interrupt.
2. Clean up all unused macros.
3. Fix devm_kmalloc_array arguments order and error check.
4. Remove unnecessary start_link logic.
5. Make all magic number as a macro.
6. Fix spelling mistakes in code and commit message.

Inochi Amaoto (5):
  PCI: spacemit-k1: Add device data support
  PCI: spacemit-k1: Add multiple PHY handles support
  dt-bindings: PCI: snps,dw-pcie: Add msi-parent for MSI handle check
  dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller
  PCI: spacemit-k1: Add Spacemit K3 PCIe host controller support

 .../devicetree/bindings/pci/snps,dw-pcie.yaml |   7 +-
 .../bindings/pci/spacemit,k3-pcie-host.yaml   | 135 +++++++++++
 drivers/pci/controller/dwc/Kconfig            |   4 +-
 drivers/pci/controller/dwc/pcie-spacemit-k1.c | 210 +++++++++++++++++-
 4 files changed, 344 insertions(+), 12 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml

--
2.54.0

[PATCH v2 1/5] PCI: spacemit-k1: Add device data support

[Inochi Amaoto]

To reuse the K1 PCIe driver logic for K3 PCIe controller, add device
data to handle the K1 specific logic and make room for the incoming
logic for K3.

Signed-off-by: Inochi Amaoto <inochiama@gmail.com>
---
 drivers/pci/controller/dwc/pcie-spacemit-k1.c | 25 ++++++++++++++++---
 1 file changed, 21 insertions(+), 4 deletions(-)

diff --git a/drivers/pci/controller/dwc/pcie-spacemit-k1.c b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
index be20a520255b..1b519d49dcc0 100644
--- a/drivers/pci/controller/dwc/pcie-spacemit-k1.c
+++ b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
@@ -57,6 +57,12 @@ struct k1_pcie {
 	u32 pmu_off;
 };
 
+struct k1_pcie_device_data {
+	const struct dw_pcie_host_ops *host_ops;
+	const struct dw_pcie_ops *ops;
+	int (*parse_port)(struct k1_pcie *k1);
+};
+
 #define to_k1_pcie(dw_pcie) \
 		platform_get_drvdata(to_platform_device((dw_pcie)->dev))
 
@@ -278,10 +284,15 @@ static int k1_pcie_parse_port(struct k1_pcie *k1)
 
 static int k1_pcie_probe(struct platform_device *pdev)
 {
+	const struct k1_pcie_device_data *data;
 	struct device *dev = &pdev->dev;
 	struct k1_pcie *k1;
 	int ret;
 
+	data = device_get_match_data(dev);
+	if (!data)
+		return -ENODEV;
+
 	k1 = devm_kzalloc(dev, sizeof(*k1), GFP_KERNEL);
 	if (!k1)
 		return -ENOMEM;
@@ -299,11 +310,11 @@ static int k1_pcie_probe(struct platform_device *pdev)
 				     "failed to map \"link\" registers\n");
 
 	k1->pci.dev = dev;
-	k1->pci.ops = &k1_pcie_ops;
+	k1->pci.ops = data->ops;
 	k1->pci.pp.num_vectors = MAX_MSI_IRQS;
 	dw_pcie_cap_set(&k1->pci, REQ_RES);
 
-	k1->pci.pp.ops = &k1_pcie_host_ops;
+	k1->pci.pp.ops = data->host_ops;
 
 	/* Hold the PHY in reset until we start the link */
 	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CLK_RESET_CONTROL,
@@ -320,7 +331,7 @@ static int k1_pcie_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, k1);
 
-	ret = k1_pcie_parse_port(k1);
+	ret = data->parse_port(k1);
 	if (ret)
 		return dev_err_probe(dev, ret, "failed to parse root port\n");
 
@@ -338,8 +349,14 @@ static void k1_pcie_remove(struct platform_device *pdev)
 	dw_pcie_host_deinit(&k1->pci.pp);
 }
 
+static const struct k1_pcie_device_data k1_pcie_device_data = {
+	.host_ops	= &k1_pcie_host_ops,
+	.ops		= &k1_pcie_ops,
+	.parse_port	= k1_pcie_parse_port,
+};
+
 static const struct of_device_id k1_pcie_of_match_table[] = {
-	{ .compatible = "spacemit,k1-pcie", },
+	{ .compatible = "spacemit,k1-pcie", .data = &k1_pcie_device_data},
 	{ }
 };
 
-- 
2.54.0

[PATCH v2 2/5] PCI: spacemit-k1: Add multiple PHY handles support

[Inochi Amaoto]

The PCIe controller on Spacemit K3 may use multiple PHYs at the
same time. The feature is not support by the current driver.
So extend the PHY definition to support multiple PHY handles.

Signed-off-by: Inochi Amaoto <inochiama@gmail.com>
---
 drivers/pci/controller/dwc/pcie-spacemit-k1.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/drivers/pci/controller/dwc/pcie-spacemit-k1.c b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
index 1b519d49dcc0..7f6f1df31cd8 100644
--- a/drivers/pci/controller/dwc/pcie-spacemit-k1.c
+++ b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
@@ -51,7 +51,8 @@
 
 struct k1_pcie {
 	struct dw_pcie pci;
-	struct phy *phy;
+	struct phy **phy;
+	int phy_count;
 	void __iomem *link;
 	struct regmap *pmu;	/* Errors ignored; MMIO-backed regmap */
 	u32 pmu_off;
@@ -171,7 +172,7 @@ static int k1_pcie_init(struct dw_pcie_rp *pp)
 	 */
 	regmap_set_bits(k1->pmu, reset_ctrl, DEVICE_TYPE_RC | PCIE_AUX_PWR_DET);
 
-	ret = phy_init(k1->phy);
+	ret = phy_init(k1->phy[0]);
 	if (ret) {
 		k1_pcie_disable_resources(k1);
 
@@ -191,12 +192,14 @@ static void k1_pcie_deinit(struct dw_pcie_rp *pp)
 {
 	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
 	struct k1_pcie *k1 = to_k1_pcie(pci);
+	int i;
 
 	/* Assert fundamental reset (drive PERST# low) */
 	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CLK_RESET_CONTROL,
 			PCIE_RC_PERST);
 
-	phy_exit(k1->phy);
+	for (i = 0; i < k1->phy_count; i++)
+		phy_exit(k1->phy[i]);
 
 	k1_pcie_disable_resources(k1);
 }
@@ -277,7 +280,12 @@ static int k1_pcie_parse_port(struct k1_pcie *k1)
 	if (IS_ERR(phy))
 		return PTR_ERR(phy);
 
-	k1->phy = phy;
+	k1->phy = devm_kmalloc_array(dev, 1, sizeof(*k1->phy), GFP_KERNEL);
+	if (!k1->phy)
+		return -ENOMEM;
+
+	k1->phy[0] = phy;
+	k1->phy_count = 1;
 
 	return 0;
 }
-- 
2.54.0

[PATCH v2 3/5] dt-bindings: PCI: snps,dw-pcie: Add msi-parent for MSI handle check

[Inochi Amaoto]

The IMSIC device on RISC-V based system does not require ID
remapping for MSI. So this device only needs "msi-parent"
property for IMSIC-based SoC, and the "msi-map" is not a
necessary property.

Add new condition for MSI handling on IMSIC based SoC.

Signed-off-by: Inochi Amaoto <inochiama@gmail.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
---
 Documentation/devicetree/bindings/pci/snps,dw-pcie.yaml | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/pci/snps,dw-pcie.yaml b/Documentation/devicetree/bindings/pci/snps,dw-pcie.yaml
index b3216141881c..91bbbc8924f6 100644
--- a/Documentation/devicetree/bindings/pci/snps,dw-pcie.yaml
+++ b/Documentation/devicetree/bindings/pci/snps,dw-pcie.yaml
@@ -27,8 +27,11 @@ allOf:
   - $ref: /schemas/pci/snps,dw-pcie-common.yaml#
   - if:
       not:
-        required:
-          - msi-map
+        anyOf:
+          - required:
+              - msi-map
+          - required:
+              - msi-parent
     then:
       properties:
         interrupt-names:
-- 
2.54.0

[PATCH v2 4/5] dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller

[Inochi Amaoto]

Add binding support for the PCIe controller on the SpacemiT K3 SoC.
This controller is almost a standard Synopsys DesignWare PCIe IP,
with some extra link and reset state control.

Signed-off-by: Inochi Amaoto <inochiama@gmail.com>
---
 .../bindings/pci/spacemit,k3-pcie-host.yaml   | 135 ++++++++++++++++++
 1 file changed, 135 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml

diff --git a/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
new file mode 100644
index 000000000000..46147a37a9ce
--- /dev/null
+++ b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
@@ -0,0 +1,135 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/pci/spacemit,k3-pcie-host.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: SpacemiT K3 PCI Express Host Controller
+
+maintainers:
+  - Inochi Amaoto <inochiama@gmail.com>
+
+description:
+  The SpacemiT K3 SoC PCIe host controller is based on the Synopsys
+  DesignWare PCIe IP. The controller uses the external MSI interrupt
+  controller.
+
+allOf:
+  - $ref: /schemas/pci/pci-host-bridge.yaml#
+  - $ref: /schemas/pci/snps,dw-pcie.yaml#
+
+properties:
+  compatible:
+    const: spacemit,k3-pcie
+
+  reg:
+    items:
+      - description: DesignWare PCIe registers
+      - description: Data Bus Interface (DBI) shadow registers
+      - description: ATU address space
+      - description: PCIe configuration space
+      - description: Link control registers
+
+  reg-names:
+    items:
+      - const: dbi
+      - const: dbi2
+      - const: atu
+      - const: config
+      - const: link
+
+  clocks:
+    items:
+      - description: DWC PCIe Data Bus Interface (DBI) clock
+      - description: DWC PCIe application AXI-bus master interface clock
+      - description: DWC PCIe application AXI-bus slave interface clock
+
+  clock-names:
+    items:
+      - const: dbi
+      - const: mstr
+      - const: slv
+
+  resets:
+    items:
+      - description: DWC PCIe Data Bus Interface (DBI) reset
+      - description: DWC PCIe application AXI-bus master interface reset
+      - description: DWC PCIe application AXI-bus slave interface reset
+
+  reset-names:
+    items:
+      - const: dbi
+      - const: mstr
+      - const: slv
+
+  msi-parent: true
+
+  phys:
+    description:
+      PHY phandle from the Combo PHY, the lane number does not depends
+      on this, since the number of lanes provided by Combo PHY can be
+      1 or 2.
+    minItems: 1
+    maxItems: 6
+
+  phy-names:
+    minItems: 1
+    maxItems: 6
+
+  spacemit,apmu:
+    $ref: /schemas/types.yaml#/definitions/phandle-array
+    description:
+      A phandle that refers to the APMU system controller, whose regmap is
+      used in managing resets and link state, along with and offset of its
+      reset control register.
+    items:
+      - items:
+          - description: phandle to APMU system controller
+          - description: register offset
+
+required:
+  - clocks
+  - clock-names
+  - resets
+  - reset-names
+  - msi-parent
+  - spacemit,apmu
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/irq.h>
+
+    soc {
+      #address-cells = <2>;
+      #size-cells = <2>;
+
+      pcie@80000000 {
+        compatible = "spacemit,k3-pcie";
+        reg = <0x0  0x80000000 0x0 0x00001000>,
+              <0x0  0x80100000 0x0 0x00001000>,
+              <0x0  0x80300000 0x0 0x00003f20>,
+              <0x11 0x00000000 0x0 0x00010000>,
+              <0x0  0x82900000 0x0 0x00001000>;
+        reg-names = "dbi", "dbi2", "atu", "config", "link";
+        device_type = "pci";
+        #address-cells = <3>;
+        #size-cells = <2>;
+        clocks = <&syscon_apmu 89>,
+                 <&syscon_apmu 56>,
+                 <&syscon_apmu 57>;
+        clock-names = "dbi", "mstr", "slv";
+        msi-parent = <&simsic>;
+        ranges = <0x01000000 0x00 0x00010000 0x11 0x00010000 0x0 0x00100000>,
+                 <0x02000000 0x0  0x00110000 0x11 0x00110000 0x0 0x7fef0000>,
+                 <0x43000000 0x18 0x00000000 0x18 0x00000000 0x1 0x00000000>;
+        resets = <&syscon_apmu 76>,
+                 <&syscon_apmu 78>,
+                 <&syscon_apmu 77>;
+        reset-names = "dbi", "mstr", "slv";
+        linux,pci-domain = <0>;
+        spacemit,apmu = <&syscon_apmu 0x1f0>;
+      };
+    };
+
-- 
2.54.0

[PATCH v2 5/5] PCI: spacemit-k1: Add Spacemit K3 PCIe host controller support

[Inochi Amaoto]

The PCIe controller on Spacemit K3 is almost a standard Synopsys
DesignWare PCIe IP with extra link and reset control. Unlike
the PCIe controller on K1, this controller supports external MSI
interrupt controller and can use multiple PHYs at the same time.

Add driver to support PCIe controller on Spacemit K3 PCIe.

Signed-off-by: Inochi Amaoto <inochiama@gmail.com>
---
 drivers/pci/controller/dwc/Kconfig            |   4 +-
 drivers/pci/controller/dwc/pcie-spacemit-k1.c | 169 ++++++++++++++++++
 2 files changed, 171 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/controller/dwc/Kconfig b/drivers/pci/controller/dwc/Kconfig
index f2fde13107f2..fae971ecd876 100644
--- a/drivers/pci/controller/dwc/Kconfig
+++ b/drivers/pci/controller/dwc/Kconfig
@@ -439,7 +439,7 @@ config PCIE_SOPHGO_DW
 	  Sophgo SoCs.
 
 config PCIE_SPACEMIT_K1
-	tristate "SpacemiT K1 PCIe controller (host mode)"
+	tristate "SpacemiT K1/K3 PCIe controller (host mode)"
 	depends on ARCH_SPACEMIT || COMPILE_TEST
 	depends on HAS_IOMEM
 	select PCIE_DW_HOST
@@ -447,7 +447,7 @@ config PCIE_SPACEMIT_K1
 	default ARCH_SPACEMIT
 	help
 	  Enables support for the DesignWare based PCIe controller in
-	  the SpacemiT K1 SoC operating in host mode.  Three controllers
+	  the SpacemiT K1/K3 SoC operating in host mode. Three controllers
 	  are available on the K1 SoC; the first of these shares a PHY
 	  with a USB 3.0 host controller (one or the other can be used).
 
diff --git a/drivers/pci/controller/dwc/pcie-spacemit-k1.c b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
index 7f6f1df31cd8..7854d26220a9 100644
--- a/drivers/pci/controller/dwc/pcie-spacemit-k1.c
+++ b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
@@ -23,6 +23,7 @@
 
 #define PCI_VENDOR_ID_SPACEMIT		0x201f
 #define PCI_DEVICE_ID_SPACEMIT_K1	0x0001
+#define PCI_DEVICE_ID_SPACEMIT_K3	0x0002
 
 /* Offsets and field definitions for link management registers */
 #define K1_PHY_AHB_IRQ_EN			0x0000
@@ -32,8 +33,20 @@
 #define SMLH_LINK_UP			BIT(1)
 #define RDLH_LINK_UP			BIT(12)
 
+#define INTR_STATUS				0x0010
+
 #define INTR_ENABLE				0x0014
 #define MSI_CTRL_INT			BIT(11)
+#define RDLH_LINK_UP_INT		BIT(20)
+
+#define K3_PHY_AHB_IRQSTATUS_INTX		0x0008
+
+#define K3_ADDR_INTR_STATUS1			0x0018
+
+#define K3_CACHE_MSTR_AWCACHE_MODE	GENMASK(14, 11)
+#define K3_CACHE_MSTR_AWCACHE_BEHAVIOR	0xf
+
+#define K3_MAX_PHY_NUMBER		6
 
 /* Some controls require APMU regmap access */
 #define SYSCON_APMU			"spacemit,apmu"
@@ -48,6 +61,9 @@
 
 #define PCIE_CONTROL_LOGIC			0x0004
 #define PCIE_SOFT_RESET			BIT(0)
+#define PCIE_PERSTN_OE			BIT(24)
+#define PCIE_PERSTN_OUT			BIT(25)
+#define PCIE_IGNORE_PERSTN		BIT(31)
 
 struct k1_pcie {
 	struct dw_pcie pci;
@@ -262,6 +278,152 @@ static const struct dw_pcie_ops k1_pcie_ops = {
 	.stop_link	= k1_pcie_stop_link,
 };
 
+static int k3_pcie_enable_phy(struct k1_pcie *pcie)
+{
+	int i, ret;
+
+	for (i = 0; i < pcie->phy_count; i++) {
+		ret = phy_init(pcie->phy[i]);
+		if (ret)
+			goto err_phy;
+	}
+
+	return 0;
+
+err_phy:
+	while (--i >= 0)
+		phy_exit(pcie->phy[i]);
+
+	return ret;
+}
+
+static int k3_pcie_init(struct dw_pcie_rp *pp)
+{
+	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
+	struct k1_pcie *k1 = to_k1_pcie(pci);
+	u32 reset_ctrl = k1->pmu_off + PCIE_CLK_RESET_CONTROL;
+	u32 val;
+	int ret;
+
+	regmap_clear_bits(k1->pmu, reset_ctrl, LTSSM_EN);
+
+	k1_pcie_toggle_soft_reset(k1);
+
+	ret = k1_pcie_enable_resources(k1);
+	if (ret)
+		return ret;
+
+	regmap_set_bits(k1->pmu, reset_ctrl, PCIE_AUX_PWR_DET);
+	regmap_clear_bits(k1->pmu, reset_ctrl, APP_HOLD_PHY_RST);
+
+	ret = k3_pcie_enable_phy(k1);
+	if (ret) {
+		k1_pcie_disable_resources(k1);
+		return ret;
+	}
+
+	/* K3: Set IGNORE_PERSTN and drive PERSTN_OE high (assert reset) */
+	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC,
+			PCIE_IGNORE_PERSTN | PCIE_PERSTN_OE | PCIE_PERSTN_OUT);
+	usleep_range(1000, 2000);
+	regmap_clear_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC, PCIE_PERSTN_OUT);
+
+	msleep(PCIE_T_PVPERL_MS);
+
+	/*
+	 * Put the controller in root complex mode, and indicate that
+	 * Vaux (3.3v) is present.
+	 */
+	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC,
+			PCIE_PERSTN_OUT | PCIE_PERSTN_OE);
+
+	val = dw_pcie_readl_dbi(pci, GEN3_EQ_CONTROL_OFF);
+	val = u32_replace_bits(val, GEN3_EQ_CONTROL_OFF_PHASE23_EXIT_MODE,
+			       GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC);
+	dw_pcie_writel_dbi(pci, GEN3_EQ_CONTROL_OFF, val);
+
+	dw_pcie_dbi_ro_wr_en(pci);
+	dw_pcie_writew_dbi(pci, PCI_VENDOR_ID, PCI_VENDOR_ID_SPACEMIT);
+	dw_pcie_writew_dbi(pci, PCI_DEVICE_ID, PCI_DEVICE_ID_SPACEMIT_K3);
+	dw_pcie_dbi_ro_wr_dis(pci);
+
+	/* Finally, as a workaround, disable ASPM L1 */
+	k1_pcie_disable_aspm_l1(k1);
+
+	return 0;
+}
+
+static int k3_pcie_msi_host_init(struct dw_pcie_rp *pp)
+{
+	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
+	u32 val;
+
+	dw_pcie_dbi_ro_wr_en(pci);
+
+	val = dw_pcie_readl_dbi(pci, COHERENCY_CONTROL_3_OFF);
+	val |= u32_replace_bits(val, K3_CACHE_MSTR_AWCACHE_BEHAVIOR,
+				K3_CACHE_MSTR_AWCACHE_MODE);
+	dw_pcie_writel_dbi(pci, COHERENCY_CONTROL_3_OFF, val);
+
+	dw_pcie_dbi_ro_wr_dis(pci);
+
+	return 0;
+}
+
+static const struct dw_pcie_host_ops k3_pcie_host_ops = {
+	.init		= k3_pcie_init,
+	.deinit		= k1_pcie_deinit,
+	.msi_init	= k3_pcie_msi_host_init,
+};
+
+static const struct dw_pcie_ops k3_pcie_ops = {
+	.link_up	= k1_pcie_link_up,
+	.start_link	= k1_pcie_start_link,
+	.stop_link	= k1_pcie_stop_link,
+};
+
+static void k3_pcie_clear_irq_status(struct k1_pcie *k1,
+				     u32 *status0, u32 *status1, u32 *status2)
+{
+	*status0 = readl_relaxed(k1->link + K3_PHY_AHB_IRQSTATUS_INTX);
+	*status1 = readl_relaxed(k1->link + INTR_STATUS);
+	*status2 = readl_relaxed(k1->link + K3_ADDR_INTR_STATUS1);
+
+	writel_relaxed(*status0, k1->link + K3_PHY_AHB_IRQSTATUS_INTX);
+	writel_relaxed(*status1, k1->link + INTR_STATUS);
+	writel_relaxed(*status2, k1->link + K3_ADDR_INTR_STATUS1);
+}
+
+static int k3_pcie_parse_port(struct k1_pcie *k1)
+{
+	struct device *dev = k1->pci.dev;
+	u32 status0, status1, status2;
+	int i;
+
+	k1->phy = devm_kmalloc_array(dev, K3_MAX_PHY_NUMBER, sizeof(*k1->phy),
+				     GFP_KERNEL);
+	if (!k1->phy)
+		return -ENOMEM;
+
+	for (i = 0; i < K3_MAX_PHY_NUMBER; i++) {
+		k1->phy[i] = devm_of_phy_get_by_index(dev, dev->of_node, i);
+		if (IS_ERR(k1->phy[i])) {
+			if (PTR_ERR(k1->phy[i]) == -ENODEV)
+				break;
+
+			return PTR_ERR(k1->phy[i]);
+		}
+	}
+
+	k1->phy_count = i;
+	if (k1->phy_count == 0)
+		return -EINVAL;
+
+	k3_pcie_clear_irq_status(k1, &status0, &status1, &status2);
+
+	return 0;
+}
+
 static int k1_pcie_parse_port(struct k1_pcie *k1)
 {
 	struct device *dev = k1->pci.dev;
@@ -363,8 +525,15 @@ static const struct k1_pcie_device_data k1_pcie_device_data = {
 	.parse_port	= k1_pcie_parse_port,
 };
 
+static const struct k1_pcie_device_data k3_pcie_device_data = {
+	.host_ops	= &k3_pcie_host_ops,
+	.ops		= &k3_pcie_ops,
+	.parse_port	= k3_pcie_parse_port,
+};
+
 static const struct of_device_id k1_pcie_of_match_table[] = {
 	{ .compatible = "spacemit,k1-pcie", .data = &k1_pcie_device_data},
+	{ .compatible = "spacemit,k3-pcie", .data = &k3_pcie_device_data},
 	{ }
 };
 
-- 
2.54.0

Re: [PATCH v2 4/5] dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [Medium] The binding completely omits the `interrupts` and `interrupt-names` properties for the host controller.
- [Medium] The schema omits standard legacy INTx interrupt routing properties (`#interrupt-cells`, `interrupt-map`, `interrupt-map-mask`).
- [Low] The `phy-names` property lacks an `items` definition to restrict valid string values.
--

commit 7b8ba34456e71d6a9cf1028caa5504d64e8f58ba
Author: Inochi Amaoto <inochiama@gmail.com>

dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller

This commit adds device tree binding support for the PCIe controller on the
SpacemiT K3 SoC. The controller is based on the Synopsys DesignWare PCIe IP
with some extra link and reset state control.

> diff --git a/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
> new file mode 100644
> index 0000000000000..46147a37a9ce6
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
[ ... ]
> +  reset-names:
> +    items:
> +      - const: dbi
> +      - const: mstr
> +      - const: slv
> +
> +  msi-parent: true
> +

The driver implementation reads and clears interrupt status registers for
host-level events, and references registers for receiving INTx events from
downstream devices.

Should this binding include interrupts and interrupt-names properties to
document the host's interrupt routing?

Additionally, should standard legacy INTx interrupt routing properties like
#interrupt-cells, interrupt-map, and interrupt-map-mask be included in the
schema and example to support downstream devices that do not use MSI?

> +  phys:
> +    description:
> +      PHY phandle from the Combo PHY, the lane number does not depends
> +      on this, since the number of lanes provided by Combo PHY can be
> +      1 or 2.
> +    minItems: 1
> +    maxItems: 6
> +
> +  phy-names:
> +    minItems: 1
> +    maxItems: 6

This isn't a bug, but the phy-names property lacks an items definition to
restrict valid string values. The schema allows any arbitrary string here.

Since the driver retrieves PHYs by index rather than name, is this property
necessary, or should the expected string items be defined?

[ ... ]

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260517014841.254085-1-inochiama@gmail.com?part=4

Re: [PATCH v2 5/5] PCI: spacemit-k1: Add Spacemit K3 PCIe host controller support

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 5 potential issue(s) to consider:
- [High] MMIO registers are accessed before the controller's clocks are enabled, leading to a synchronous external abort or system crash.
- [High] Incorrect use of `u32_replace_bits()` combined with bitwise OR (`|=`) fails to clear existing bits.
- [High] Missing Root Complex mode configuration for the K3 controller due to a copy-paste error.
- [Medium] A bitmask macro for a completely different field is incorrectly passed as the replacement value to `u32_replace_bits()`, writing unexpected values.
- [Medium] Asymmetric teardown: `k1_pcie_deinit()` fails to assert reset on the K3 controller because it uses a reset line the controller is explicitly configured to ignore.
--

commit 2826ac7fe9e0d0c0dd248a7fc412ce12f4d7b491
Author: Inochi Amaoto <inochiama@gmail.com>

PCI: spacemit-k1: Add Spacemit K3 PCIe host controller support

This commit adds host controller support for the Spacemit K3 PCIe IP. It
introduces K3-specific initialization, PHY enablement, and MSI host setup,
hooking these into the existing spacemit-k1 driver.

> diff --git a/drivers/pci/controller/dwc/pcie-spacemit-k1.c b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> index 7f6f1df31cd81..7854d26220a98 100644
> --- a/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> +++ b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
[ ... ]
> @@ -262,6 +278,152 @@ static const struct dw_pcie_ops k1_pcie_ops = {
[ ... ]
> +static int k3_pcie_init(struct dw_pcie_rp *pp)
> +{
[ ... ]
> +	/*
> +	 * Put the controller in root complex mode, and indicate that
> +	 * Vaux (3.3v) is present.
> +	 */
> +	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC,
> +			PCIE_PERSTN_OUT | PCIE_PERSTN_OE);

Does this code configure the controller as a Root Complex? 

The comment mentions putting the controller in root complex mode, but the 
register write only sets the PERSTN bits. In the K1 initialization, this is 
done by setting DEVICE_TYPE_RC, which appears to be missing here.

> +
> +	val = dw_pcie_readl_dbi(pci, GEN3_EQ_CONTROL_OFF);
> +	val = u32_replace_bits(val, GEN3_EQ_CONTROL_OFF_PHASE23_EXIT_MODE,
> +			       GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC);

Is the replacement value intended to be GEN3_EQ_CONTROL_OFF_PHASE23_EXIT_MODE? 

This passes a bitmask macro for a completely different field into 
u32_replace_bits(), which shifts the value into the target mask and writes an 
unexpected value into the register.

> +	dw_pcie_writel_dbi(pci, GEN3_EQ_CONTROL_OFF, val);
[ ... ]
> +static int k3_pcie_msi_host_init(struct dw_pcie_rp *pp)
> +{
> +	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
> +	u32 val;
> +
> +	dw_pcie_dbi_ro_wr_en(pci);
> +
> +	val = dw_pcie_readl_dbi(pci, COHERENCY_CONTROL_3_OFF);
> +	val |= u32_replace_bits(val, K3_CACHE_MSTR_AWCACHE_BEHAVIOR,
> +				K3_CACHE_MSTR_AWCACHE_MODE);

Will this clear the existing bits correctly? 

The u32_replace_bits() macro returns the fully modified 32-bit word. By using 
the bitwise OR assignment (|=), the original bits from val are ORed with the 
new bits, leaving any originally set bits as 1 instead of clearing them.

> +	dw_pcie_writel_dbi(pci, COHERENCY_CONTROL_3_OFF, val);
[ ... ]
> +static const struct dw_pcie_host_ops k3_pcie_host_ops = {
> +	.init		= k3_pcie_init,
> +	.deinit		= k1_pcie_deinit,

Can k1_pcie_deinit() successfully assert the reset for the K3 controller? 

During initialization, k3_pcie_init() sets PCIE_IGNORE_PERSTN in 
PCIE_CONTROL_LOGIC. If k1_pcie_deinit() attempts to assert reset using 
PCIE_RC_PERST, the hardware might ignore it, leaving the device improperly 
reset during teardown.

> +	.msi_init	= k3_pcie_msi_host_init,
> +};
[ ... ]
> +static void k3_pcie_clear_irq_status(struct k1_pcie *k1,
> +				     u32 *status0, u32 *status1, u32 *status2)
> +{
> +	*status0 = readl_relaxed(k1->link + K3_PHY_AHB_IRQSTATUS_INTX);
> +	*status1 = readl_relaxed(k1->link + INTR_STATUS);
> +	*status2 = readl_relaxed(k1->link + K3_ADDR_INTR_STATUS1);
> +
> +	writel_relaxed(*status0, k1->link + K3_PHY_AHB_IRQSTATUS_INTX);
> +	writel_relaxed(*status1, k1->link + INTR_STATUS);
> +	writel_relaxed(*status2, k1->link + K3_ADDR_INTR_STATUS1);
> +}
> +
> +static int k3_pcie_parse_port(struct k1_pcie *k1)
> +{
[ ... ]
> +	k1->phy_count = i;
> +	if (k1->phy_count == 0)
> +		return -EINVAL;
> +
> +	k3_pcie_clear_irq_status(k1, &status0, &status1, &status2);
> +
> +	return 0;
> +}

Could accessing these MMIO registers before the clocks are enabled cause a
synchronous external abort?

The probe function calls parse_port() before dw_pcie_host_init(). This means
k3_pcie_clear_irq_status() performs direct MMIO reads and writes to the 
k1->link registers before k1_pcie_enable_resources() is called later in 
k3_pcie_init(). 

Accessing an unclocked peripheral can cause a bus stall on ARM/RISC-V 
architectures.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260517014841.254085-1-inochiama@gmail.com?part=5

Re: [PATCH v2 4/5] dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller

[Inochi Amaoto]

On Sun, May 17, 2026 at 02:48:47AM +0000, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
> - [Medium] The binding completely omits the `interrupts` and `interrupt-names` properties for the host controller.
> - [Medium] The schema omits standard legacy INTx interrupt routing properties (`#interrupt-cells`, `interrupt-map`, `interrupt-map-mask`).
> - [Low] The `phy-names` property lacks an `items` definition to restrict valid string values.
> --
> 
> commit 7b8ba34456e71d6a9cf1028caa5504d64e8f58ba
> Author: Inochi Amaoto <inochiama@gmail.com>
> 
> dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller
> 
> This commit adds device tree binding support for the PCIe controller on the
> SpacemiT K3 SoC. The controller is based on the Synopsys DesignWare PCIe IP
> with some extra link and reset state control.
> 
> > diff --git a/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
> > new file mode 100644
> > index 0000000000000..46147a37a9ce6
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
> [ ... ]
> > +  reset-names:
> > +    items:
> > +      - const: dbi
> > +      - const: mstr
> > +      - const: slv
> > +
> > +  msi-parent: true
> > +
> 
> The driver implementation reads and clears interrupt status registers for
> host-level events, and references registers for receiving INTx events from
> downstream devices.
> 
> Should this binding include interrupts and interrupt-names properties to
> document the host's interrupt routing?
> 
> Additionally, should standard legacy INTx interrupt routing properties like
> #interrupt-cells, interrupt-map, and interrupt-map-mask be included in the
> schema and example to support downstream devices that do not use MSI?
> 

The driver just does the interrupt cleanup as it requires, and no
additional interrupt is needed

> > +  phys:
> > +    description:
> > +      PHY phandle from the Combo PHY, the lane number does not depends
> > +      on this, since the number of lanes provided by Combo PHY can be
> > +      1 or 2.
> > +    minItems: 1
> > +    maxItems: 6
> > +
> > +  phy-names:
> > +    minItems: 1
> > +    maxItems: 6
> 
> This isn't a bug, but the phy-names property lacks an items definition to
> restrict valid string values. The schema allows any arbitrary string here.
> 
> Since the driver retrieves PHYs by index rather than name, is this property
> necessary, or should the expected string items be defined?
> 

The driver just use index, and the name follow generic DesignWare
binding.

> [ ... ]
> 
> -- 
> Sashiko AI review · https://sashiko.dev/#/patchset/20260517014841.254085-1-inochiama@gmail.com?part=4

Regards,
Inochi

Re: [PATCH v2 5/5] PCI: spacemit-k1: Add Spacemit K3 PCIe host controller support

[Inochi Amaoto]

On Sun, May 17, 2026 at 03:16:20AM +0000, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 5 potential issue(s) to consider:
> - [High] MMIO registers are accessed before the controller's clocks are enabled, leading to a synchronous external abort or system crash.
> - [High] Incorrect use of `u32_replace_bits()` combined with bitwise OR (`|=`) fails to clear existing bits.
> - [High] Missing Root Complex mode configuration for the K3 controller due to a copy-paste error.
> - [Medium] A bitmask macro for a completely different field is incorrectly passed as the replacement value to `u32_replace_bits()`, writing unexpected values.
> - [Medium] Asymmetric teardown: `k1_pcie_deinit()` fails to assert reset on the K3 controller because it uses a reset line the controller is explicitly configured to ignore.
> --
> 
> commit 2826ac7fe9e0d0c0dd248a7fc412ce12f4d7b491
> Author: Inochi Amaoto <inochiama@gmail.com>
> 
> PCI: spacemit-k1: Add Spacemit K3 PCIe host controller support
> 
> This commit adds host controller support for the Spacemit K3 PCIe IP. It
> introduces K3-specific initialization, PHY enablement, and MSI host setup,
> hooking these into the existing spacemit-k1 driver.
> 
> > diff --git a/drivers/pci/controller/dwc/pcie-spacemit-k1.c b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> > index 7f6f1df31cd81..7854d26220a98 100644
> > --- a/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> > +++ b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> [ ... ]
> > @@ -262,6 +278,152 @@ static const struct dw_pcie_ops k1_pcie_ops = {
> [ ... ]
> > +static int k3_pcie_init(struct dw_pcie_rp *pp)
> > +{
> [ ... ]
> > +	/*
> > +	 * Put the controller in root complex mode, and indicate that
> > +	 * Vaux (3.3v) is present.
> > +	 */
> > +	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC,
> > +			PCIE_PERSTN_OUT | PCIE_PERSTN_OE);
> 
> Does this code configure the controller as a Root Complex? 
> 
> The comment mentions putting the controller in root complex mode, but the 
> register write only sets the PERSTN bits. In the K1 initialization, this is 
> done by setting DEVICE_TYPE_RC, which appears to be missing here.
> 

Unlike K1, K3 does not needs this.

> > +
> > +	val = dw_pcie_readl_dbi(pci, GEN3_EQ_CONTROL_OFF);
> > +	val = u32_replace_bits(val, GEN3_EQ_CONTROL_OFF_PHASE23_EXIT_MODE,
> > +			       GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC);
> 
> Is the replacement value intended to be GEN3_EQ_CONTROL_OFF_PHASE23_EXIT_MODE? 
> 
> This passes a bitmask macro for a completely different field into 
> u32_replace_bits(), which shifts the value into the target mask and writes an 
> unexpected value into the register.
> 

The GEN3_EQ_CONTROL_OFF_PHASE23_EXIT_MODE should be BIT(7),
it is a mistake.

> > +	dw_pcie_writel_dbi(pci, GEN3_EQ_CONTROL_OFF, val);
> [ ... ]
> > +static int k3_pcie_msi_host_init(struct dw_pcie_rp *pp)
> > +{
> > +	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
> > +	u32 val;
> > +
> > +	dw_pcie_dbi_ro_wr_en(pci);
> > +
> > +	val = dw_pcie_readl_dbi(pci, COHERENCY_CONTROL_3_OFF);
> > +	val |= u32_replace_bits(val, K3_CACHE_MSTR_AWCACHE_BEHAVIOR,
> > +				K3_CACHE_MSTR_AWCACHE_MODE);
> 
> Will this clear the existing bits correctly? 
> 
> The u32_replace_bits() macro returns the fully modified 32-bit word. By using 
> the bitwise OR assignment (|=), the original bits from val are ORed with the 
> new bits, leaving any originally set bits as 1 instead of clearing them.
> 

Yes, it is wrong, I will fix it.

> > +	dw_pcie_writel_dbi(pci, COHERENCY_CONTROL_3_OFF, val);
> [ ... ]
> > +static const struct dw_pcie_host_ops k3_pcie_host_ops = {
> > +	.init		= k3_pcie_init,
> > +	.deinit		= k1_pcie_deinit,
> 
> Can k1_pcie_deinit() successfully assert the reset for the K3 controller? 
> 
> During initialization, k3_pcie_init() sets PCIE_IGNORE_PERSTN in 
> PCIE_CONTROL_LOGIC. If k1_pcie_deinit() attempts to assert reset using 
> PCIE_RC_PERST, the hardware might ignore it, leaving the device improperly 
> reset during teardown.
> 
> > +	.msi_init	= k3_pcie_msi_host_init,
> > +};
> [ ... ]
> > +static void k3_pcie_clear_irq_status(struct k1_pcie *k1,
> > +				     u32 *status0, u32 *status1, u32 *status2)
> > +{
> > +	*status0 = readl_relaxed(k1->link + K3_PHY_AHB_IRQSTATUS_INTX);
> > +	*status1 = readl_relaxed(k1->link + INTR_STATUS);
> > +	*status2 = readl_relaxed(k1->link + K3_ADDR_INTR_STATUS1);
> > +
> > +	writel_relaxed(*status0, k1->link + K3_PHY_AHB_IRQSTATUS_INTX);
> > +	writel_relaxed(*status1, k1->link + INTR_STATUS);
> > +	writel_relaxed(*status2, k1->link + K3_ADDR_INTR_STATUS1);
> > +}
> > +
> > +static int k3_pcie_parse_port(struct k1_pcie *k1)
> > +{
> [ ... ]
> > +	k1->phy_count = i;
> > +	if (k1->phy_count == 0)
> > +		return -EINVAL;
> > +
> > +	k3_pcie_clear_irq_status(k1, &status0, &status1, &status2);
> > +
> > +	return 0;
> > +}
> 
> Could accessing these MMIO registers before the clocks are enabled cause a
> synchronous external abort?
> 

No problems, these register is always ready.

> The probe function calls parse_port() before dw_pcie_host_init(). This means
> k3_pcie_clear_irq_status() performs direct MMIO reads and writes to the 
> k1->link registers before k1_pcie_enable_resources() is called later in 
> k3_pcie_init(). 
> 
> Accessing an unclocked peripheral can cause a bus stall on ARM/RISC-V 
> architectures.
> 
> -- 
> Sashiko AI review · https://sashiko.dev/#/patchset/20260517014841.254085-1-inochiama@gmail.com?part=5

Re: [PATCH v2 2/5] PCI: spacemit-k1: Add multiple PHY handles support

[Andy Shevchenko]

On Sun, May 17, 2026 at 09:48:37AM +0800, Inochi Amaoto wrote:
> The PCIe controller on Spacemit K3 may use multiple PHYs at the
> same time. The feature is not support by the current driver.
> So extend the PHY definition to support multiple PHY handles.

...

> +	k1->phy = devm_kmalloc_array(dev, 1, sizeof(*k1->phy), GFP_KERNEL);

Hmm... What's the point in having _array(1) ? devm_kmalloc() should suffice.

> +	if (!k1->phy)
> +		return -ENOMEM;

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v2 2/5] PCI: spacemit-k1: Add multiple PHY handles support

[Inochi Amaoto]

On Sun, May 17, 2026 at 11:07:40AM +0300, Andy Shevchenko wrote:
> On Sun, May 17, 2026 at 09:48:37AM +0800, Inochi Amaoto wrote:
> > The PCIe controller on Spacemit K3 may use multiple PHYs at the
> > same time. The feature is not support by the current driver.
> > So extend the PHY definition to support multiple PHY handles.
> 
> ...
> 
> > +	k1->phy = devm_kmalloc_array(dev, 1, sizeof(*k1->phy), GFP_KERNEL);
> 
> Hmm... What's the point in having _array(1) ? devm_kmalloc() should suffice.
> 


Yes, you are right, I forgot it is "struct phy *", so the devm_kmalloc()
is enough, will fix in the next version.

Regards,
Inochi

> > +	if (!k1->phy)
> > +		return -ENOMEM;
> 
> -- 
> With Best Regards,
> Andy Shevchenko
> 
> 

Re: [PATCH v2 4/5] dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller

[Rob Herring]

On Sun, May 17, 2026 at 12:38:39PM +0800, Inochi Amaoto wrote:
> On Sun, May 17, 2026 at 02:48:47AM +0000, sashiko-bot@kernel.org wrote:
> > Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
> > - [Medium] The binding completely omits the `interrupts` and `interrupt-names` properties for the host controller.
> > - [Medium] The schema omits standard legacy INTx interrupt routing properties (`#interrupt-cells`, `interrupt-map`, `interrupt-map-mask`).
> > - [Low] The `phy-names` property lacks an `items` definition to restrict valid string values.
> > --
> > 
> > commit 7b8ba34456e71d6a9cf1028caa5504d64e8f58ba
> > Author: Inochi Amaoto <inochiama@gmail.com>
> > 
> > dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller
> > 
> > This commit adds device tree binding support for the PCIe controller on the
> > SpacemiT K3 SoC. The controller is based on the Synopsys DesignWare PCIe IP
> > with some extra link and reset state control.
> > 
> > > diff --git a/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
> > > new file mode 100644
> > > index 0000000000000..46147a37a9ce6
> > > --- /dev/null
> > > +++ b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
> > [ ... ]
> > > +  reset-names:
> > > +    items:
> > > +      - const: dbi
> > > +      - const: mstr
> > > +      - const: slv
> > > +
> > > +  msi-parent: true
> > > +
> > 
> > The driver implementation reads and clears interrupt status registers for
> > host-level events, and references registers for receiving INTx events from
> > downstream devices.
> > 
> > Should this binding include interrupts and interrupt-names properties to
> > document the host's interrupt routing?
> > 
> > Additionally, should standard legacy INTx interrupt routing properties like
> > #interrupt-cells, interrupt-map, and interrupt-map-mask be included in the
> > schema and example to support downstream devices that do not use MSI?
> > 
> 
> The driver just does the interrupt cleanup as it requires, and no
> additional interrupt is needed

Unless there are no interrupts and only MSIs are supported, then you 
need to define 'interrupts' property at least. The common schema only 
defines all possible names, but not which ones are used and the order.

Rob

Re: [PATCH v2 4/5] dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller

[Inochi Amaoto]

On Mon, Jun 01, 2026 at 05:42:24PM -0500, Rob Herring wrote:
> On Sun, May 17, 2026 at 12:38:39PM +0800, Inochi Amaoto wrote:
> > On Sun, May 17, 2026 at 02:48:47AM +0000, sashiko-bot@kernel.org wrote:
> > > Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
> > > - [Medium] The binding completely omits the `interrupts` and `interrupt-names` properties for the host controller.
> > > - [Medium] The schema omits standard legacy INTx interrupt routing properties (`#interrupt-cells`, `interrupt-map`, `interrupt-map-mask`).
> > > - [Low] The `phy-names` property lacks an `items` definition to restrict valid string values.
> > > --
> > > 
> > > commit 7b8ba34456e71d6a9cf1028caa5504d64e8f58ba
> > > Author: Inochi Amaoto <inochiama@gmail.com>
> > > 
> > > dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller
> > > 
> > > This commit adds device tree binding support for the PCIe controller on the
> > > SpacemiT K3 SoC. The controller is based on the Synopsys DesignWare PCIe IP
> > > with some extra link and reset state control.
> > > 
> > > > diff --git a/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
> > > > new file mode 100644
> > > > index 0000000000000..46147a37a9ce6
> > > > --- /dev/null
> > > > +++ b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
> > > [ ... ]
> > > > +  reset-names:
> > > > +    items:
> > > > +      - const: dbi
> > > > +      - const: mstr
> > > > +      - const: slv
> > > > +
> > > > +  msi-parent: true
> > > > +
> > > 
> > > The driver implementation reads and clears interrupt status registers for
> > > host-level events, and references registers for receiving INTx events from
> > > downstream devices.
> > > 
> > > Should this binding include interrupts and interrupt-names properties to
> > > document the host's interrupt routing?
> > > 
> > > Additionally, should standard legacy INTx interrupt routing properties like
> > > #interrupt-cells, interrupt-map, and interrupt-map-mask be included in the
> > > schema and example to support downstream devices that do not use MSI?
> > > 
> > 
> > The driver just does the interrupt cleanup as it requires, and no
> > additional interrupt is needed
> 
> Unless there are no interrupts and only MSIs are supported, then you 
> need to define 'interrupts' property at least. The common schema only 
> defines all possible names, but not which ones are used and the order.
> 
> Rob

Yes, currently there is no interrupts that should be enabled in Linux, 
and only MSI is supported without INTx. So I think it is fine to removed
the interrupt properity.

Regards,
Inochi

Re: [PATCH v2 4/5] dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller

[Manivannan Sadhasivam]

On Sun, May 17, 2026 at 09:48:39AM +0800, Inochi Amaoto wrote:
> Add binding support for the PCIe controller on the SpacemiT K3 SoC.
> This controller is almost a standard Synopsys DesignWare PCIe IP,
> with some extra link and reset state control.
> 
> Signed-off-by: Inochi Amaoto <inochiama@gmail.com>

Why can't you reuse the existing spacemit,k1-pcie-host.yaml binding? I see very
few differences which could be added using conditionals. Also, this binding
defines the PHY property in the controller node, which is a way backwards as we
now prefer to define these in Root Port node as spacemit,k1-pcie-host.yaml does.

- Mani

> ---
>  .../bindings/pci/spacemit,k3-pcie-host.yaml   | 135 ++++++++++++++++++
>  1 file changed, 135 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
> 
> diff --git a/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
> new file mode 100644
> index 000000000000..46147a37a9ce
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
> @@ -0,0 +1,135 @@
> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/pci/spacemit,k3-pcie-host.yaml#
> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: SpacemiT K3 PCI Express Host Controller
> +
> +maintainers:
> +  - Inochi Amaoto <inochiama@gmail.com>
> +
> +description:
> +  The SpacemiT K3 SoC PCIe host controller is based on the Synopsys
> +  DesignWare PCIe IP. The controller uses the external MSI interrupt
> +  controller.
> +
> +allOf:
> +  - $ref: /schemas/pci/pci-host-bridge.yaml#
> +  - $ref: /schemas/pci/snps,dw-pcie.yaml#
> +
> +properties:
> +  compatible:
> +    const: spacemit,k3-pcie
> +
> +  reg:
> +    items:
> +      - description: DesignWare PCIe registers
> +      - description: Data Bus Interface (DBI) shadow registers
> +      - description: ATU address space
> +      - description: PCIe configuration space
> +      - description: Link control registers
> +
> +  reg-names:
> +    items:
> +      - const: dbi
> +      - const: dbi2
> +      - const: atu
> +      - const: config
> +      - const: link
> +
> +  clocks:
> +    items:
> +      - description: DWC PCIe Data Bus Interface (DBI) clock
> +      - description: DWC PCIe application AXI-bus master interface clock
> +      - description: DWC PCIe application AXI-bus slave interface clock
> +
> +  clock-names:
> +    items:
> +      - const: dbi
> +      - const: mstr
> +      - const: slv
> +
> +  resets:
> +    items:
> +      - description: DWC PCIe Data Bus Interface (DBI) reset
> +      - description: DWC PCIe application AXI-bus master interface reset
> +      - description: DWC PCIe application AXI-bus slave interface reset
> +
> +  reset-names:
> +    items:
> +      - const: dbi
> +      - const: mstr
> +      - const: slv
> +
> +  msi-parent: true
> +
> +  phys:
> +    description:
> +      PHY phandle from the Combo PHY, the lane number does not depends
> +      on this, since the number of lanes provided by Combo PHY can be
> +      1 or 2.
> +    minItems: 1
> +    maxItems: 6
> +
> +  phy-names:
> +    minItems: 1
> +    maxItems: 6
> +
> +  spacemit,apmu:
> +    $ref: /schemas/types.yaml#/definitions/phandle-array
> +    description:
> +      A phandle that refers to the APMU system controller, whose regmap is
> +      used in managing resets and link state, along with and offset of its
> +      reset control register.
> +    items:
> +      - items:
> +          - description: phandle to APMU system controller
> +          - description: register offset
> +
> +required:
> +  - clocks
> +  - clock-names
> +  - resets
> +  - reset-names
> +  - msi-parent
> +  - spacemit,apmu
> +
> +unevaluatedProperties: false
> +
> +examples:
> +  - |
> +    #include <dt-bindings/interrupt-controller/irq.h>
> +
> +    soc {
> +      #address-cells = <2>;
> +      #size-cells = <2>;
> +
> +      pcie@80000000 {
> +        compatible = "spacemit,k3-pcie";
> +        reg = <0x0  0x80000000 0x0 0x00001000>,
> +              <0x0  0x80100000 0x0 0x00001000>,
> +              <0x0  0x80300000 0x0 0x00003f20>,
> +              <0x11 0x00000000 0x0 0x00010000>,
> +              <0x0  0x82900000 0x0 0x00001000>;
> +        reg-names = "dbi", "dbi2", "atu", "config", "link";
> +        device_type = "pci";
> +        #address-cells = <3>;
> +        #size-cells = <2>;
> +        clocks = <&syscon_apmu 89>,
> +                 <&syscon_apmu 56>,
> +                 <&syscon_apmu 57>;
> +        clock-names = "dbi", "mstr", "slv";
> +        msi-parent = <&simsic>;
> +        ranges = <0x01000000 0x00 0x00010000 0x11 0x00010000 0x0 0x00100000>,
> +                 <0x02000000 0x0  0x00110000 0x11 0x00110000 0x0 0x7fef0000>,
> +                 <0x43000000 0x18 0x00000000 0x18 0x00000000 0x1 0x00000000>;
> +        resets = <&syscon_apmu 76>,
> +                 <&syscon_apmu 78>,
> +                 <&syscon_apmu 77>;
> +        reset-names = "dbi", "mstr", "slv";
> +        linux,pci-domain = <0>;
> +        spacemit,apmu = <&syscon_apmu 0x1f0>;
> +      };
> +    };
> +
> -- 
> 2.54.0
> 

-- 
மணிவண்ணன் சதாசிவம்

Re: [PATCH v2 5/5] PCI: spacemit-k1: Add Spacemit K3 PCIe host controller support

[Manivannan Sadhasivam]

On Sun, May 17, 2026 at 09:48:40AM +0800, Inochi Amaoto wrote:
> The PCIe controller on Spacemit K3 is almost a standard Synopsys
> DesignWare PCIe IP with extra link and reset control. Unlike
> the PCIe controller on K1, this controller supports external MSI
> interrupt controller and can use multiple PHYs at the same time.
> 
> Add driver to support PCIe controller on Spacemit K3 PCIe.
> 
> Signed-off-by: Inochi Amaoto <inochiama@gmail.com>
> ---
>  drivers/pci/controller/dwc/Kconfig            |   4 +-
>  drivers/pci/controller/dwc/pcie-spacemit-k1.c | 169 ++++++++++++++++++
>  2 files changed, 171 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/pci/controller/dwc/Kconfig b/drivers/pci/controller/dwc/Kconfig
> index f2fde13107f2..fae971ecd876 100644
> --- a/drivers/pci/controller/dwc/Kconfig
> +++ b/drivers/pci/controller/dwc/Kconfig
> @@ -439,7 +439,7 @@ config PCIE_SOPHGO_DW
>  	  Sophgo SoCs.
>  
>  config PCIE_SPACEMIT_K1
> -	tristate "SpacemiT K1 PCIe controller (host mode)"
> +	tristate "SpacemiT K1/K3 PCIe controller (host mode)"

Can you just say 'SpacemiT PCIe controller (host mode)"? I believe I asked Alex
while adding K1 support and he said this driver might not support future IP
revisions, but here we are.

>  	depends on ARCH_SPACEMIT || COMPILE_TEST
>  	depends on HAS_IOMEM
>  	select PCIE_DW_HOST
> @@ -447,7 +447,7 @@ config PCIE_SPACEMIT_K1
>  	default ARCH_SPACEMIT
>  	help
>  	  Enables support for the DesignWare based PCIe controller in
> -	  the SpacemiT K1 SoC operating in host mode.  Three controllers
> +	  the SpacemiT K1/K3 SoC operating in host mode. Three controllers
>  	  are available on the K1 SoC; the first of these shares a PHY
>  	  with a USB 3.0 host controller (one or the other can be used).
>  
> diff --git a/drivers/pci/controller/dwc/pcie-spacemit-k1.c b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> index 7f6f1df31cd8..7854d26220a9 100644
> --- a/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> +++ b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> @@ -23,6 +23,7 @@
>  
>  #define PCI_VENDOR_ID_SPACEMIT		0x201f
>  #define PCI_DEVICE_ID_SPACEMIT_K1	0x0001
> +#define PCI_DEVICE_ID_SPACEMIT_K3	0x0002
>  
>  /* Offsets and field definitions for link management registers */
>  #define K1_PHY_AHB_IRQ_EN			0x0000
> @@ -32,8 +33,20 @@
>  #define SMLH_LINK_UP			BIT(1)
>  #define RDLH_LINK_UP			BIT(12)
>  
> +#define INTR_STATUS				0x0010
> +
>  #define INTR_ENABLE				0x0014
>  #define MSI_CTRL_INT			BIT(11)
> +#define RDLH_LINK_UP_INT		BIT(20)
> +
> +#define K3_PHY_AHB_IRQSTATUS_INTX		0x0008
> +
> +#define K3_ADDR_INTR_STATUS1			0x0018
> +
> +#define K3_CACHE_MSTR_AWCACHE_MODE	GENMASK(14, 11)
> +#define K3_CACHE_MSTR_AWCACHE_BEHAVIOR	0xf
> +
> +#define K3_MAX_PHY_NUMBER		6

What does this mean? 6 ports?

>  
>  /* Some controls require APMU regmap access */
>  #define SYSCON_APMU			"spacemit,apmu"
> @@ -48,6 +61,9 @@
>  
>  #define PCIE_CONTROL_LOGIC			0x0004
>  #define PCIE_SOFT_RESET			BIT(0)
> +#define PCIE_PERSTN_OE			BIT(24)
> +#define PCIE_PERSTN_OUT			BIT(25)
> +#define PCIE_IGNORE_PERSTN		BIT(31)
>  
>  struct k1_pcie {
>  	struct dw_pcie pci;
> @@ -262,6 +278,152 @@ static const struct dw_pcie_ops k1_pcie_ops = {
>  	.stop_link	= k1_pcie_stop_link,
>  };
>  
> +static int k3_pcie_enable_phy(struct k1_pcie *pcie)
> +{
> +	int i, ret;
> +
> +	for (i = 0; i < pcie->phy_count; i++) {
> +		ret = phy_init(pcie->phy[i]);
> +		if (ret)
> +			goto err_phy;
> +	}
> +
> +	return 0;
> +
> +err_phy:
> +	while (--i >= 0)
> +		phy_exit(pcie->phy[i]);
> +
> +	return ret;
> +}
> +
> +static int k3_pcie_init(struct dw_pcie_rp *pp)
> +{
> +	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
> +	struct k1_pcie *k1 = to_k1_pcie(pci);
> +	u32 reset_ctrl = k1->pmu_off + PCIE_CLK_RESET_CONTROL;
> +	u32 val;
> +	int ret;
> +
> +	regmap_clear_bits(k1->pmu, reset_ctrl, LTSSM_EN);
> +
> +	k1_pcie_toggle_soft_reset(k1);
> +
> +	ret = k1_pcie_enable_resources(k1);
> +	if (ret)
> +		return ret;
> +
> +	regmap_set_bits(k1->pmu, reset_ctrl, PCIE_AUX_PWR_DET);
> +	regmap_clear_bits(k1->pmu, reset_ctrl, APP_HOLD_PHY_RST);
> +
> +	ret = k3_pcie_enable_phy(k1);
> +	if (ret) {
> +		k1_pcie_disable_resources(k1);
> +		return ret;
> +	}
> +
> +	/* K3: Set IGNORE_PERSTN and drive PERSTN_OE high (assert reset) */

What does this mean?

> +	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC,
> +			PCIE_IGNORE_PERSTN | PCIE_PERSTN_OE | PCIE_PERSTN_OUT);
> +	usleep_range(1000, 2000);
> +	regmap_clear_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC, PCIE_PERSTN_OUT);
> +
> +	msleep(PCIE_T_PVPERL_MS);
> +
> +	/*
> +	 * Put the controller in root complex mode, and indicate that
> +	 * Vaux (3.3v) is present.
> +	 */

How can the driver confirm without checking DT for vpcie3v3aux-supply?

> +	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC,
> +			PCIE_PERSTN_OUT | PCIE_PERSTN_OE);
> +
> +	val = dw_pcie_readl_dbi(pci, GEN3_EQ_CONTROL_OFF);
> +	val = u32_replace_bits(val, GEN3_EQ_CONTROL_OFF_PHASE23_EXIT_MODE,
> +			       GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC);
> +	dw_pcie_writel_dbi(pci, GEN3_EQ_CONTROL_OFF, val);
> +
> +	dw_pcie_dbi_ro_wr_en(pci);
> +	dw_pcie_writew_dbi(pci, PCI_VENDOR_ID, PCI_VENDOR_ID_SPACEMIT);
> +	dw_pcie_writew_dbi(pci, PCI_DEVICE_ID, PCI_DEVICE_ID_SPACEMIT_K3);
> +	dw_pcie_dbi_ro_wr_dis(pci);
> +
> +	/* Finally, as a workaround, disable ASPM L1 */
> +	k1_pcie_disable_aspm_l1(k1);
> +
> +	return 0;
> +}
> +
> +static int k3_pcie_msi_host_init(struct dw_pcie_rp *pp)
> +{
> +	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
> +	u32 val;
> +
> +	dw_pcie_dbi_ro_wr_en(pci);
> +
> +	val = dw_pcie_readl_dbi(pci, COHERENCY_CONTROL_3_OFF);
> +	val |= u32_replace_bits(val, K3_CACHE_MSTR_AWCACHE_BEHAVIOR,
> +				K3_CACHE_MSTR_AWCACHE_MODE);
> +	dw_pcie_writel_dbi(pci, COHERENCY_CONTROL_3_OFF, val);
> +
> +	dw_pcie_dbi_ro_wr_dis(pci);
> +
> +	return 0;
> +}
> +
> +static const struct dw_pcie_host_ops k3_pcie_host_ops = {
> +	.init		= k3_pcie_init,
> +	.deinit		= k1_pcie_deinit,
> +	.msi_init	= k3_pcie_msi_host_init,
> +};
> +
> +static const struct dw_pcie_ops k3_pcie_ops = {
> +	.link_up	= k1_pcie_link_up,
> +	.start_link	= k1_pcie_start_link,
> +	.stop_link	= k1_pcie_stop_link,
> +};
> +
> +static void k3_pcie_clear_irq_status(struct k1_pcie *k1,
> +				     u32 *status0, u32 *status1, u32 *status2)
> +{
> +	*status0 = readl_relaxed(k1->link + K3_PHY_AHB_IRQSTATUS_INTX);
> +	*status1 = readl_relaxed(k1->link + INTR_STATUS);
> +	*status2 = readl_relaxed(k1->link + K3_ADDR_INTR_STATUS1);
> +
> +	writel_relaxed(*status0, k1->link + K3_PHY_AHB_IRQSTATUS_INTX);
> +	writel_relaxed(*status1, k1->link + INTR_STATUS);
> +	writel_relaxed(*status2, k1->link + K3_ADDR_INTR_STATUS1);
> +}
> +
> +static int k3_pcie_parse_port(struct k1_pcie *k1)
> +{
> +	struct device *dev = k1->pci.dev;
> +	u32 status0, status1, status2;
> +	int i;
> +
> +	k1->phy = devm_kmalloc_array(dev, K3_MAX_PHY_NUMBER, sizeof(*k1->phy),
> +				     GFP_KERNEL);
> +	if (!k1->phy)
> +		return -ENOMEM;
> +
> +	for (i = 0; i < K3_MAX_PHY_NUMBER; i++) {
> +		k1->phy[i] = devm_of_phy_get_by_index(dev, dev->of_node, i);
> +		if (IS_ERR(k1->phy[i])) {
> +			if (PTR_ERR(k1->phy[i]) == -ENODEV)
> +				break;
> +
> +			return PTR_ERR(k1->phy[i]);
> +		}
> +	}
> +
> +	k1->phy_count = i;
> +	if (k1->phy_count == 0)
> +		return -EINVAL;
> +
> +	k3_pcie_clear_irq_status(k1, &status0, &status1, &status2);
> +
> +	return 0;
> +}

This function should iterate over the Root Port nodes defined in DT.

- Mani

-- 
மணிவண்ணன் சதாசிவம்

Re: [PATCH v2 5/5] PCI: spacemit-k1: Add Spacemit K3 PCIe host controller support

[Andy Shevchenko]

On Tue, Jun 09, 2026 at 07:48:12PM +0530, Manivannan Sadhasivam wrote:
> On Sun, May 17, 2026 at 09:48:40AM +0800, Inochi Amaoto wrote:
> > The PCIe controller on Spacemit K3 is almost a standard Synopsys
> > DesignWare PCIe IP with extra link and reset control. Unlike
> > the PCIe controller on K1, this controller supports external MSI
> > interrupt controller and can use multiple PHYs at the same time.
> > 
> > Add driver to support PCIe controller on Spacemit K3 PCIe.

...

> > +static int k3_pcie_enable_phy(struct k1_pcie *pcie)
> > +{
> > +	int i, ret;

No need to have i signed.

> > +
> > +	for (i = 0; i < pcie->phy_count; i++) {
> > +		ret = phy_init(pcie->phy[i]);
> > +		if (ret)
> > +			goto err_phy;
> > +	}
> > +
> > +	return 0;
> > +
> > +err_phy:
> > +	while (--i >= 0)

	while (i--)

is shorter form of the same.

> > +		phy_exit(pcie->phy[i]);
> > +

...

> > +	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC,
> > +			PCIE_IGNORE_PERSTN | PCIE_PERSTN_OE | PCIE_PERSTN_OUT);
> > +	usleep_range(1000, 2000);

fsleep(1 * USEC_PER_MSEC)

> > +	regmap_clear_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC, PCIE_PERSTN_OUT);
> > +
> > +	msleep(PCIE_T_PVPERL_MS);

...

> > +	int i;

Why is 'i' signed?

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v2 2/5] PCI: spacemit-k1: Add multiple PHY handles support

[Alex Elder]

On 5/16/26 8:48 PM, Inochi Amaoto wrote:
> The PCIe controller on Spacemit K3 may use multiple PHYs at the
> same time. The feature is not support by the current driver.
> So extend the PHY definition to support multiple PHY handles.
> 
> Signed-off-by: Inochi Amaoto <inochiama@gmail.com>
> ---
>   drivers/pci/controller/dwc/pcie-spacemit-k1.c | 16 ++++++++++++----
>   1 file changed, 12 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/pci/controller/dwc/pcie-spacemit-k1.c b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> index 1b519d49dcc0..7f6f1df31cd8 100644
> --- a/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> +++ b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> @@ -51,7 +51,8 @@
>   
>   struct k1_pcie {
>   	struct dw_pcie pci;
> -	struct phy *phy;
> +	struct phy **phy;

I haven't looked further at this series yet, but do you have
any interest in making this a flexible array of pointers
(counted_by(phy_count))?

> +	int phy_count;
>   	void __iomem *link;
>   	struct regmap *pmu;	/* Errors ignored; MMIO-backed regmap */
>   	u32 pmu_off;
> @@ -171,7 +172,7 @@ static int k1_pcie_init(struct dw_pcie_rp *pp)
>   	 */
>   	regmap_set_bits(k1->pmu, reset_ctrl, DEVICE_TYPE_RC | PCIE_AUX_PWR_DET);
>   
> -	ret = phy_init(k1->phy);
> +	ret = phy_init(k1->phy[0]);

If you're going to have an array you should probably put in place
a loop to initialize all phy_count elements of the array here
(as is done for phy_exit(), below).

>   	if (ret) {
>   		k1_pcie_disable_resources(k1);
>   
> @@ -191,12 +192,14 @@ static void k1_pcie_deinit(struct dw_pcie_rp *pp)
>   {
>   	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
>   	struct k1_pcie *k1 = to_k1_pcie(pci);
> +	int i;
>   
>   	/* Assert fundamental reset (drive PERST# low) */
>   	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CLK_RESET_CONTROL,
>   			PCIE_RC_PERST);
>   
> -	phy_exit(k1->phy);
> +	for (i = 0; i < k1->phy_count; i++)
> +		phy_exit(k1->phy[i]);
>   
>   	k1_pcie_disable_resources(k1);
>   }
> @@ -277,7 +280,12 @@ static int k1_pcie_parse_port(struct k1_pcie *k1)
>   	if (IS_ERR(phy))
>   		return PTR_ERR(phy);

Andy said something different about this.

But it seems to me you are intentionally allocating an
*array* of PHY pointers, which happens to have only one entry.
But your objective is that you want to allocate an array of
them so you can support more than just one PHY, right?

> -	k1->phy = phy;
> +	k1->phy = devm_kmalloc_array(dev, 1, sizeof(*k1->phy), GFP_KERNEL);

Probably should use the kzalloc variant.

> +	if (!k1->phy)
> +		return -ENOMEM;
> +
> +	k1->phy[0] = phy;

I think what's wrong is the above assignment.

If you are truly allocating an array of PHY *pointers*,
then you need to also allocate the phy structure that
each entry in the array points to.

The above assignment is erroneously assigning the first entry in the
array to point to the array, and not a new entry.

					-Alex


> +	k1->phy_count = 1;
>   
>   	return 0;
>   }

Re: [PATCH v2 4/5] dt-bindings: PCI: spacemit: Introduce Spacemit K3 PCIe host controller

[Alex Elder]

On 6/9/26 9:03 AM, Manivannan Sadhasivam wrote:
> On Sun, May 17, 2026 at 09:48:39AM +0800, Inochi Amaoto wrote:
>> Add binding support for the PCIe controller on the SpacemiT K3 SoC.
>> This controller is almost a standard Synopsys DesignWare PCIe IP,
>> with some extra link and reset state control.
>>
>> Signed-off-by: Inochi Amaoto <inochiama@gmail.com>
> 
> Why can't you reuse the existing spacemit,k1-pcie-host.yaml binding? I see very
> few differences which could be added using conditionals. Also, this binding
> defines the PHY property in the controller node, which is a way backwards as we
> now prefer to define these in Root Port node as spacemit,k1-pcie-host.yaml does.

I agree.  I have another couple of comments below, including several
things that just point out what's the same and what's different.

					-Alex

> 
> - Mani
> 
>> ---
>>   .../bindings/pci/spacemit,k3-pcie-host.yaml   | 135 ++++++++++++++++++
>>   1 file changed, 135 insertions(+)
>>   create mode 100644 Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
>>
>> diff --git a/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
>> new file mode 100644
>> index 000000000000..46147a37a9ce
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/pci/spacemit,k3-pcie-host.yaml
>> @@ -0,0 +1,135 @@
>> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
>> +%YAML 1.2
>> +---
>> +$id: http://devicetree.org/schemas/pci/spacemit,k3-pcie-host.yaml#
>> +$schema: http://devicetree.org/meta-schemas/core.yaml#
>> +
>> +title: SpacemiT K3 PCI Express Host Controller
>> +
>> +maintainers:
>> +  - Inochi Amaoto <inochiama@gmail.com>

If you would like to maintain the (common) SpacemiT K1 host
controller binding I'd be OK with that.

>> +
>> +description:
>> +  The SpacemiT K3 SoC PCIe host controller is based on the Synopsys
>> +  DesignWare PCIe IP. The controller uses the external MSI interrupt
>> +  controller.
>> +
>> +allOf:
>> +  - $ref: /schemas/pci/pci-host-bridge.yaml#
>> +  - $ref: /schemas/pci/snps,dw-pcie.yaml#
>> +
>> +properties:
>> +  compatible:
>> +    const: spacemit,k3-pcie
>> +
>> +  reg:
>> +    items:
>> +      - description: DesignWare PCIe registers
>> +      - description: Data Bus Interface (DBI) shadow registers

The above is new (different from K1), as is "dbi2" as its name.

>> +      - description: ATU address space
>> +      - description: PCIe configuration space
>> +      - description: Link control registers
>> +
>> +  reg-names:
>> +    items:
>> +      - const: dbi
>> +      - const: dbi2
>> +      - const: atu
>> +      - const: config
>> +      - const: link
>> +
>> +  clocks:

Clocks and resets are the same as K1.

>> +    items:
>> +      - description: DWC PCIe Data Bus Interface (DBI) clock
>> +      - description: DWC PCIe application AXI-bus master interface clock
>> +      - description: DWC PCIe application AXI-bus slave interface clock
>> +
>> +  clock-names:
>> +    items:
>> +      - const: dbi
>> +      - const: mstr
>> +      - const: slv
>> +
>> +  resets:
>> +    items:
>> +      - description: DWC PCIe Data Bus Interface (DBI) reset
>> +      - description: DWC PCIe application AXI-bus master interface reset
>> +      - description: DWC PCIe application AXI-bus slave interface reset
>> +
>> +  reset-names:
>> +    items:
>> +      - const: dbi
>> +      - const: mstr
>> +      - const: slv
>> +
>> +  msi-parent: true

msi-parent and phys are not present in the K1 binding.

>> +
>> +  phys:
>> +    description:
>> +      PHY phandle from the Combo PHY, the lane number does not depends
>> +      on this, since the number of lanes provided by Combo PHY can be
>> +      1 or 2.
>> +    minItems: 1
>> +    maxItems: 6
>> +
>> +  phy-names:
>> +    minItems: 1
>> +    maxItems: 6
>> +

The following property is the same as K1.

>> +  spacemit,apmu:
>> +    $ref: /schemas/types.yaml#/definitions/phandle-array
>> +    description:
>> +      A phandle that refers to the APMU system controller, whose regmap is
>> +      used in managing resets and link state, along with and offset of its
>> +      reset control register.
>> +    items:
>> +      - items:
>> +          - description: phandle to APMU system controller
>> +          - description: register offset
>> +
>> +required:
>> +  - clocks
>> +  - clock-names
>> +  - resets
>> +  - reset-names
>> +  - msi-parent
>> +  - spacemit,apmu
>> +
>> +unevaluatedProperties: false
>> +
>> +examples:
>> +  - |
>> +    #include <dt-bindings/interrupt-controller/irq.h>
>> +
>> +    soc {
>> +      #address-cells = <2>;
>> +      #size-cells = <2>;
>> +
>> +      pcie@80000000 {
>> +        compatible = "spacemit,k3-pcie";
>> +        reg = <0x0  0x80000000 0x0 0x00001000>,
>> +              <0x0  0x80100000 0x0 0x00001000>,
>> +              <0x0  0x80300000 0x0 0x00003f20>,
>> +              <0x11 0x00000000 0x0 0x00010000>,
>> +              <0x0  0x82900000 0x0 0x00001000>;
>> +        reg-names = "dbi", "dbi2", "atu", "config", "link";
>> +        device_type = "pci";
>> +        #address-cells = <3>;
>> +        #size-cells = <2>;
>> +        clocks = <&syscon_apmu 89>,
>> +                 <&syscon_apmu 56>,
>> +                 <&syscon_apmu 57>;
>> +        clock-names = "dbi", "mstr", "slv";
>> +        msi-parent = <&simsic>;
>> +        ranges = <0x01000000 0x00 0x00010000 0x11 0x00010000 0x0 0x00100000>,
>> +                 <0x02000000 0x0  0x00110000 0x11 0x00110000 0x0 0x7fef0000>,
>> +                 <0x43000000 0x18 0x00000000 0x18 0x00000000 0x1 0x00000000>;
>> +        resets = <&syscon_apmu 76>,
>> +                 <&syscon_apmu 78>,
>> +                 <&syscon_apmu 77>;
>> +        reset-names = "dbi", "mstr", "slv";
>> +        linux,pci-domain = <0>;
>> +        spacemit,apmu = <&syscon_apmu 0x1f0>;
>> +      };
>> +    };
>> +
>> -- 
>> 2.54.0
>>
> 

Re: [PATCH v2 5/5] PCI: spacemit-k1: Add Spacemit K3 PCIe host controller support

[Alex Elder]

On 5/16/26 8:48 PM, Inochi Amaoto wrote:
> The PCIe controller on Spacemit K3 is almost a standard Synopsys
> DesignWare PCIe IP with extra link and reset control. Unlike
> the PCIe controller on K1, this controller supports external MSI
> interrupt controller and can use multiple PHYs at the same time.
> 
> Add driver to support PCIe controller on Spacemit K3 PCIe.

It seems like you're creating a lot of new code for K3.  In
some cases it's very similar to K1, and it's not clear why it
needs to be different.

I'd much rather see a patch that prepares for K3 support by
doing minor refactoring of the existing code to support K1
in a way that will make adding K3 support more natural.
Then a patch to enable K3 should be simpler (and can focus
on what is truly different).

> 
> Signed-off-by: Inochi Amaoto <inochiama@gmail.com>
> ---
>   drivers/pci/controller/dwc/Kconfig            |   4 +-
>   drivers/pci/controller/dwc/pcie-spacemit-k1.c | 169 ++++++++++++++++++
>   2 files changed, 171 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/pci/controller/dwc/Kconfig b/drivers/pci/controller/dwc/Kconfig
> index f2fde13107f2..fae971ecd876 100644
> --- a/drivers/pci/controller/dwc/Kconfig
> +++ b/drivers/pci/controller/dwc/Kconfig
> @@ -439,7 +439,7 @@ config PCIE_SOPHGO_DW
>   	  Sophgo SoCs.
>   
>   config PCIE_SPACEMIT_K1
> -	tristate "SpacemiT K1 PCIe controller (host mode)"
> +	tristate "SpacemiT K1/K3 PCIe controller (host mode)"
>   	depends on ARCH_SPACEMIT || COMPILE_TEST
>   	depends on HAS_IOMEM
>   	select PCIE_DW_HOST
> @@ -447,7 +447,7 @@ config PCIE_SPACEMIT_K1
>   	default ARCH_SPACEMIT
>   	help
>   	  Enables support for the DesignWare based PCIe controller in
> -	  the SpacemiT K1 SoC operating in host mode.  Three controllers
> +	  the SpacemiT K1/K3 SoC operating in host mode. Three controllers
>   	  are available on the K1 SoC; the first of these shares a PHY
>   	  with a USB 3.0 host controller (one or the other can be used).
>   
> diff --git a/drivers/pci/controller/dwc/pcie-spacemit-k1.c b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> index 7f6f1df31cd8..7854d26220a9 100644
> --- a/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> +++ b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
> @@ -23,6 +23,7 @@
>   
>   #define PCI_VENDOR_ID_SPACEMIT		0x201f
>   #define PCI_DEVICE_ID_SPACEMIT_K1	0x0001
> +#define PCI_DEVICE_ID_SPACEMIT_K3	0x0002
>   
>   /* Offsets and field definitions for link management registers */
>   #define K1_PHY_AHB_IRQ_EN			0x0000
> @@ -32,8 +33,20 @@
>   #define SMLH_LINK_UP			BIT(1)
>   #define RDLH_LINK_UP			BIT(12)
>   
> +#define INTR_STATUS				0x0010
> +
>   #define INTR_ENABLE				0x0014
>   #define MSI_CTRL_INT			BIT(11)
> +#define RDLH_LINK_UP_INT		BIT(20)
> +
> +#define K3_PHY_AHB_IRQSTATUS_INTX		0x0008

Can you add INTX support for K1 as well (perhaps in a separate
patch)?

> +#define K3_ADDR_INTR_STATUS1			0x0018
> +
> +#define K3_CACHE_MSTR_AWCACHE_MODE	GENMASK(14, 11)
> +#define K3_CACHE_MSTR_AWCACHE_BEHAVIOR	0xf
> +
> +#define K3_MAX_PHY_NUMBER		6

You used "count" in patch 2 as the field name.

>   
>   /* Some controls require APMU regmap access */
>   #define SYSCON_APMU			"spacemit,apmu"
> @@ -48,6 +61,9 @@
>   
>   #define PCIE_CONTROL_LOGIC			0x0004
>   #define PCIE_SOFT_RESET			BIT(0)
> +#define PCIE_PERSTN_OE			BIT(24)
> +#define PCIE_PERSTN_OUT			BIT(25)
> +#define PCIE_IGNORE_PERSTN		BIT(31)
>   
>   struct k1_pcie {
>   	struct dw_pcie pci;
> @@ -262,6 +278,152 @@ static const struct dw_pcie_ops k1_pcie_ops = {
>   	.stop_link	= k1_pcie_stop_link,
>   };
>   
> +static int k3_pcie_enable_phy(struct k1_pcie *pcie)

Can you just make K1's single PHY be a special case of
having "N" PHYs?  I.e., just set the phy_count for
K1 to be 1, so this loop would work for both K1 and K3?

> +{
> +	int i, ret;
> +
> +	for (i = 0; i < pcie->phy_count; i++) {
> +		ret = phy_init(pcie->phy[i]);
> +		if (ret)
> +			goto err_phy;
> +	}
> +
> +	return 0;
> +
> +err_phy:
> +	while (--i >= 0)
> +		phy_exit(pcie->phy[i]);
> +
> +	return ret;
> +}
> +
> +static int k3_pcie_init(struct dw_pcie_rp *pp)
> +{
> +	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
> +	struct k1_pcie *k1 = to_k1_pcie(pci);
> +	u32 reset_ctrl = k1->pmu_off + PCIE_CLK_RESET_CONTROL;
> +	u32 val;
> +	int ret;
> +
> +	regmap_clear_bits(k1->pmu, reset_ctrl, LTSSM_EN);

Should the above be done for K1?  Would it hurt?  Handle
both K1 and K3 the same way if possible.

The next two things are identical to k1_pcie_init().  Make
the code common if possible, so it's very obvious what
really needs to be different between the two.

> +
> +	k1_pcie_toggle_soft_reset(k1);

The "k1" prefix is fine for now, but if this driver gets
used for more devices in the future, it might be worth
renaming things to emphasize that it's not K1-specific.

> +	ret = k1_pcie_enable_resources(k1);
> +	if (ret)
> +		return ret;
> +
> +	regmap_set_bits(k1->pmu, reset_ctrl, PCIE_AUX_PWR_DET);
> +	regmap_clear_bits(k1->pmu, reset_ctrl, APP_HOLD_PHY_RST);
> +

You enable the PHY here much earlier than what's done in
the K1 code.  Should the K1 PHY be enabled earlier?
Also, I don't really think there needs to be separate
versions of the code that enables PHYs for K1 and K3.

> +	ret = k3_pcie_enable_phy(k1);
> +	if (ret) {
> +		k1_pcie_disable_resources(k1);
> +		return ret;
> +	}
> +
> +	/* K3: Set IGNORE_PERSTN and drive PERSTN_OE high (assert reset) */
> +	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC,
> +			PCIE_IGNORE_PERSTN | PCIE_PERSTN_OE | PCIE_PERSTN_OUT);
> +	usleep_range(1000, 2000);
> +	regmap_clear_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC, PCIE_PERSTN_OUT);
> +
> +	msleep(PCIE_T_PVPERL_MS);
> +
> +	/*
> +	 * Put the controller in root complex mode, and indicate that
> +	 * Vaux (3.3v) is present.
> +	 */
> +	regmap_set_bits(k1->pmu, k1->pmu_off + PCIE_CONTROL_LOGIC,
> +			PCIE_PERSTN_OUT | PCIE_PERSTN_OE);
> +
> +	val = dw_pcie_readl_dbi(pci, GEN3_EQ_CONTROL_OFF);
> +	val = u32_replace_bits(val, GEN3_EQ_CONTROL_OFF_PHASE23_EXIT_MODE,
> +			       GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC);
> +	dw_pcie_writel_dbi(pci, GEN3_EQ_CONTROL_OFF, val);
> +

The following block of code (roughly) is done right after
enabling resources in the K1 version of this function.

Maybe the order you do it is better, but in that case,
change the (existing, and soon, common) code to do it
however is best if that's the case.

You should try to factor out the common parts and minimize
what's actually different between the two.

I would also expect that the device ID would be stored in the
platform data rather than having both init functions hard-code
the value here.

I'm going to leave it at that for now.

					-Alex

> +	dw_pcie_dbi_ro_wr_en(pci);
> +	dw_pcie_writew_dbi(pci, PCI_VENDOR_ID, PCI_VENDOR_ID_SPACEMIT);
> +	dw_pcie_writew_dbi(pci, PCI_DEVICE_ID, PCI_DEVICE_ID_SPACEMIT_K3);
> +	dw_pcie_dbi_ro_wr_dis(pci);
> +
> +	/* Finally, as a workaround, disable ASPM L1 */
> +	k1_pcie_disable_aspm_l1(k1);
> +
> +	return 0;
> +}
> +
> +static int k3_pcie_msi_host_init(struct dw_pcie_rp *pp)
> +{
> +	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
> +	u32 val;
> +
> +	dw_pcie_dbi_ro_wr_en(pci);
> +
> +	val = dw_pcie_readl_dbi(pci, COHERENCY_CONTROL_3_OFF);
> +	val |= u32_replace_bits(val, K3_CACHE_MSTR_AWCACHE_BEHAVIOR,
> +				K3_CACHE_MSTR_AWCACHE_MODE);
> +	dw_pcie_writel_dbi(pci, COHERENCY_CONTROL_3_OFF, val);
> +
> +	dw_pcie_dbi_ro_wr_dis(pci);
> +
> +	return 0;
> +}
> +
> +static const struct dw_pcie_host_ops k3_pcie_host_ops = {
> +	.init		= k3_pcie_init,
> +	.deinit		= k1_pcie_deinit,
> +	.msi_init	= k3_pcie_msi_host_init,
> +};
> +
> +static const struct dw_pcie_ops k3_pcie_ops = {
> +	.link_up	= k1_pcie_link_up,
> +	.start_link	= k1_pcie_start_link,
> +	.stop_link	= k1_pcie_stop_link,
> +};
> +
> +static void k3_pcie_clear_irq_status(struct k1_pcie *k1,
> +				     u32 *status0, u32 *status1, u32 *status2)
> +{
> +	*status0 = readl_relaxed(k1->link + K3_PHY_AHB_IRQSTATUS_INTX);
> +	*status1 = readl_relaxed(k1->link + INTR_STATUS);
> +	*status2 = readl_relaxed(k1->link + K3_ADDR_INTR_STATUS1);
> +
> +	writel_relaxed(*status0, k1->link + K3_PHY_AHB_IRQSTATUS_INTX);
> +	writel_relaxed(*status1, k1->link + INTR_STATUS);
> +	writel_relaxed(*status2, k1->link + K3_ADDR_INTR_STATUS1);
> +}
> +
> +static int k3_pcie_parse_port(struct k1_pcie *k1)
> +{
> +	struct device *dev = k1->pci.dev;
> +	u32 status0, status1, status2;
> +	int i;
> +
> +	k1->phy = devm_kmalloc_array(dev, K3_MAX_PHY_NUMBER, sizeof(*k1->phy),
> +				     GFP_KERNEL);
> +	if (!k1->phy)
> +		return -ENOMEM;
> +
> +	for (i = 0; i < K3_MAX_PHY_NUMBER; i++) {
> +		k1->phy[i] = devm_of_phy_get_by_index(dev, dev->of_node, i);
> +		if (IS_ERR(k1->phy[i])) {
> +			if (PTR_ERR(k1->phy[i]) == -ENODEV)
> +				break;
> +
> +			return PTR_ERR(k1->phy[i]);
> +		}
> +	}
> +
> +	k1->phy_count = i;
> +	if (k1->phy_count == 0)
> +		return -EINVAL;
> +
> +	k3_pcie_clear_irq_status(k1, &status0, &status1, &status2);
> +
> +	return 0;
> +}
> +
>   static int k1_pcie_parse_port(struct k1_pcie *k1)
>   {
>   	struct device *dev = k1->pci.dev;
> @@ -363,8 +525,15 @@ static const struct k1_pcie_device_data k1_pcie_device_data = {
>   	.parse_port	= k1_pcie_parse_port,
>   };
>   
> +static const struct k1_pcie_device_data k3_pcie_device_data = {
> +	.host_ops	= &k3_pcie_host_ops,
> +	.ops		= &k3_pcie_ops,
> +	.parse_port	= k3_pcie_parse_port,
> +};
> +
>   static const struct of_device_id k1_pcie_of_match_table[] = {
>   	{ .compatible = "spacemit,k1-pcie", .data = &k1_pcie_device_data},
> +	{ .compatible = "spacemit,k3-pcie", .data = &k3_pcie_device_data},
>   	{ }
>   };
>   

Re: [PATCH v2 5/5] PCI: spacemit-k1: Add Spacemit K3 PCIe host controller support

[Alex Elder]

On 6/9/26 9:18 AM, Manivannan Sadhasivam wrote:
> On Sun, May 17, 2026 at 09:48:40AM +0800, Inochi Amaoto wrote:
>> The PCIe controller on Spacemit K3 is almost a standard Synopsys
>> DesignWare PCIe IP with extra link and reset control. Unlike
>> the PCIe controller on K1, this controller supports external MSI
>> interrupt controller and can use multiple PHYs at the same time.
>>
>> Add driver to support PCIe controller on Spacemit K3 PCIe.
>>
>> Signed-off-by: Inochi Amaoto <inochiama@gmail.com>
>> ---
>>   drivers/pci/controller/dwc/Kconfig            |   4 +-
>>   drivers/pci/controller/dwc/pcie-spacemit-k1.c | 169 ++++++++++++++++++
>>   2 files changed, 171 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/pci/controller/dwc/Kconfig b/drivers/pci/controller/dwc/Kconfig
>> index f2fde13107f2..fae971ecd876 100644
>> --- a/drivers/pci/controller/dwc/Kconfig
>> +++ b/drivers/pci/controller/dwc/Kconfig
>> @@ -439,7 +439,7 @@ config PCIE_SOPHGO_DW
>>   	  Sophgo SoCs.
>>   
>>   config PCIE_SPACEMIT_K1
>> -	tristate "SpacemiT K1 PCIe controller (host mode)"
>> +	tristate "SpacemiT K1/K3 PCIe controller (host mode)"
> 
> Can you just say 'SpacemiT PCIe controller (host mode)"? I believe I asked Alex
> while adding K1 support and he said this driver might not support future IP
> revisions, but here we are.

(Sorry if I said that in error.)  Will this be K1 and K3 only?
If so what you say is fine, but I agree with Mani, if it's going
to also form the basis of K5 (or whatever) just make it "SpacemiT".
And probably change the Kconfig symbol to PCIE_SPACEMIT.

					-Alex

>>   	depends on ARCH_SPACEMIT || COMPILE_TEST
>>   	depends on HAS_IOMEM
>>   	select PCIE_DW_HOST
>> @@ -447,7 +447,7 @@ config PCIE_SPACEMIT_K1
>>   	default ARCH_SPACEMIT
>>   	help
>>   	  Enables support for the DesignWare based PCIe controller in
>> -	  the SpacemiT K1 SoC operating in host mode.  Three controllers
>> +	  the SpacemiT K1/K3 SoC operating in host mode. Three controllers
>>   	  are available on the K1 SoC; the first of these shares a PHY
>>   	  with a USB 3.0 host controller (one or the other can be used).
>>   . . .