* [PATCH v10 0/3] Add support for AMD MDB IP as Root Port
@ 2025-02-11 6:38 Thippeswamy Havalige
2025-02-11 6:38 ` [PATCH v10 1/3] dt-bindings: PCI: dwc: Add AMD Versal2 mdb slcr support Thippeswamy Havalige
` (2 more replies)
0 siblings, 3 replies; 6+ messages in thread
From: Thippeswamy Havalige @ 2025-02-11 6:38 UTC (permalink / raw)
To: bhelgaas, lpieralisi, kw, manivannan.sadhasivam, robh, krzk+dt,
conor+dt
Cc: linux-pci, devicetree, linux-kernel, jingoohan1, michal.simek,
bharat.kumar.gogada, Thippeswamy Havalige
This series of patch add support for AMD MDB IP as Root Port.
The AMD MDB IP support's 32 bit and 64bit BAR's at Gen5 speed.
As Root Port it supports MSI and legacy interrupts.
Thippeswamy Havalige (3):
dt-bindings: PCI: dwc: Add AMD Versal2 mdb slcr support
dt-bindings: PCI: amd-mdb: Add AMD Versal2 MDB PCIe Root Port Bridge
PCI: amd-mdb: Add AMD MDB Root Port driver
.../bindings/pci/amd,versal2-mdb-host.yaml | 121 +++++
.../devicetree/bindings/pci/snps,dw-pcie.yaml | 2 +
drivers/pci/controller/dwc/Kconfig | 11 +
drivers/pci/controller/dwc/Makefile | 1 +
drivers/pci/controller/dwc/pcie-amd-mdb.c | 477 ++++++++++++++++++
5 files changed, 612 insertions(+)
create mode 100644 Documentation/devicetree/bindings/pci/amd,versal2-mdb-host.yaml
create mode 100644 drivers/pci/controller/dwc/pcie-amd-mdb.c
--
2.43.0
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH v10 1/3] dt-bindings: PCI: dwc: Add AMD Versal2 mdb slcr support
2025-02-11 6:38 [PATCH v10 0/3] Add support for AMD MDB IP as Root Port Thippeswamy Havalige
@ 2025-02-11 6:38 ` Thippeswamy Havalige
2025-02-11 6:38 ` [PATCH v10 2/3] dt-bindings: PCI: amd-mdb: Add AMD Versal2 MDB PCIe Root Port Bridge Thippeswamy Havalige
2025-02-11 6:38 ` [PATCH v10 3/3] PCI: amd-mdb: Add AMD MDB Root Port driver Thippeswamy Havalige
2 siblings, 0 replies; 6+ messages in thread
From: Thippeswamy Havalige @ 2025-02-11 6:38 UTC (permalink / raw)
To: bhelgaas, lpieralisi, kw, manivannan.sadhasivam, robh, krzk+dt,
conor+dt
Cc: linux-pci, devicetree, linux-kernel, jingoohan1, michal.simek,
bharat.kumar.gogada, Thippeswamy Havalige
Add support for mdb slcr aperture that is only supported for AMD Versal2
devices.
Signed-off-by: Thippeswamy Havalige <thippeswamy.havalige@amd.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Acked-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
---
Changes in v3:
-------------
- Introduced below changes in dwc yaml schema.
Changes in v5:
-------------
- Modify mdb_pcie_slcr as constant.
Changes in v6:
-------------
-Modify slcr constant
---
Documentation/devicetree/bindings/pci/snps,dw-pcie.yaml | 2 ++
1 file changed, 2 insertions(+)
diff --git a/Documentation/devicetree/bindings/pci/snps,dw-pcie.yaml b/Documentation/devicetree/bindings/pci/snps,dw-pcie.yaml
index 205326fb2d75..fdecfe6ad5f1 100644
--- a/Documentation/devicetree/bindings/pci/snps,dw-pcie.yaml
+++ b/Documentation/devicetree/bindings/pci/snps,dw-pcie.yaml
@@ -113,6 +113,8 @@ properties:
enum: [ smu, mpu ]
- description: Tegra234 aperture
enum: [ ecam ]
+ - description: AMD MDB PCIe slcr region
+ const: slcr
allOf:
- contains:
const: dbi
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH v10 2/3] dt-bindings: PCI: amd-mdb: Add AMD Versal2 MDB PCIe Root Port Bridge
2025-02-11 6:38 [PATCH v10 0/3] Add support for AMD MDB IP as Root Port Thippeswamy Havalige
2025-02-11 6:38 ` [PATCH v10 1/3] dt-bindings: PCI: dwc: Add AMD Versal2 mdb slcr support Thippeswamy Havalige
@ 2025-02-11 6:38 ` Thippeswamy Havalige
2025-02-11 6:38 ` [PATCH v10 3/3] PCI: amd-mdb: Add AMD MDB Root Port driver Thippeswamy Havalige
2 siblings, 0 replies; 6+ messages in thread
From: Thippeswamy Havalige @ 2025-02-11 6:38 UTC (permalink / raw)
To: bhelgaas, lpieralisi, kw, manivannan.sadhasivam, robh, krzk+dt,
conor+dt
Cc: linux-pci, devicetree, linux-kernel, jingoohan1, michal.simek,
bharat.kumar.gogada, Thippeswamy Havalige
Add AMD Versal2 MDB (Multimedia DMA Bridge) PCIe Root Port Bridge.
Signed-off-by: Thippeswamy Havalige <thippeswamy.havalige@amd.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Acked-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
---
.../bindings/pci/amd,versal2-mdb-host.yaml | 121 ++++++++++++++++++
Changes in v2:
-------------
- Modify patch subject.
- Add pcie host bridge reference.
- Modify filename as per compatible string.
- Remove standard PCI properties.
- Modify interrupt controller description.
- Indentation
Changes in v3:
-------------
- Modified SLCR to lower case.
- Add dwc schemas.
- Remove common properties.
- Move additionalProperties below properties.
- Remove ranges property from required properties.
- Drop blank line.
- Modify pci@ to pcie@
Changes in v4:
-------------
- None.
Changes in v5:
-------------
- None.
Changes in v6:
--------------
- Reduce dbi size to 4k.
- update register name to slcr.
Changes in v7:
--------------
- Add reviewed-by
Changes in v10:
--------------
- None
---
1 file changed, 121 insertions(+)
create mode 100644 Documentation/devicetree/bindings/pci/amd,versal2-mdb-host.yaml
diff --git a/Documentation/devicetree/bindings/pci/amd,versal2-mdb-host.yaml b/Documentation/devicetree/bindings/pci/amd,versal2-mdb-host.yaml
new file mode 100644
index 000000000000..43dc2585c237
--- /dev/null
+++ b/Documentation/devicetree/bindings/pci/amd,versal2-mdb-host.yaml
@@ -0,0 +1,121 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/pci/amd,versal2-mdb-host.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: AMD Versal2 MDB(Multimedia DMA Bridge) Host Controller
+
+maintainers:
+ - Thippeswamy Havalige <thippeswamy.havalige@amd.com>
+
+allOf:
+ - $ref: /schemas/pci/pci-host-bridge.yaml#
+ - $ref: /schemas/pci/snps,dw-pcie.yaml#
+
+properties:
+ compatible:
+ const: amd,versal2-mdb-host
+
+ reg:
+ items:
+ - description: MDB System Level Control and Status Register (SLCR) Base
+ - description: configuration region
+ - description: data bus interface
+ - description: address translation unit register
+
+ reg-names:
+ items:
+ - const: slcr
+ - const: config
+ - const: dbi
+ - const: atu
+
+ ranges:
+ maxItems: 2
+
+ msi-map:
+ maxItems: 1
+
+ interrupts:
+ maxItems: 1
+
+ interrupt-map-mask:
+ items:
+ - const: 0
+ - const: 0
+ - const: 0
+ - const: 7
+
+ interrupt-map:
+ maxItems: 4
+
+ "#interrupt-cells":
+ const: 1
+
+ interrupt-controller:
+ description: identifies the node as an interrupt controller
+ type: object
+ additionalProperties: false
+ properties:
+ interrupt-controller: true
+
+ "#address-cells":
+ const: 0
+
+ "#interrupt-cells":
+ const: 1
+
+ required:
+ - interrupt-controller
+ - "#address-cells"
+ - "#interrupt-cells"
+
+required:
+ - reg
+ - reg-names
+ - interrupts
+ - interrupt-map
+ - interrupt-map-mask
+ - msi-map
+ - "#interrupt-cells"
+ - interrupt-controller
+
+unevaluatedProperties: false
+
+examples:
+ - |
+ #include <dt-bindings/interrupt-controller/arm-gic.h>
+ #include <dt-bindings/interrupt-controller/irq.h>
+
+ soc {
+ #address-cells = <2>;
+ #size-cells = <2>;
+ pcie@ed931000 {
+ compatible = "amd,versal2-mdb-host";
+ reg = <0x0 0xed931000 0x0 0x2000>,
+ <0x1000 0x100000 0x0 0xff00000>,
+ <0x1000 0x0 0x0 0x1000>,
+ <0x0 0xed860000 0x0 0x2000>;
+ reg-names = "slcr", "config", "dbi", "atu";
+ ranges = <0x2000000 0x00 0xa0000000 0x00 0xa0000000 0x00 0x10000000>,
+ <0x43000000 0x1100 0x00 0x1100 0x00 0x00 0x1000000>;
+ interrupts = <GIC_SPI 165 IRQ_TYPE_LEVEL_HIGH>;
+ interrupt-parent = <&gic>;
+ interrupt-map-mask = <0 0 0 7>;
+ interrupt-map = <0 0 0 1 &pcie_intc_0 0>,
+ <0 0 0 2 &pcie_intc_0 1>,
+ <0 0 0 3 &pcie_intc_0 2>,
+ <0 0 0 4 &pcie_intc_0 3>;
+ msi-map = <0x0 &gic_its 0x00 0x10000>;
+ #address-cells = <3>;
+ #size-cells = <2>;
+ #interrupt-cells = <1>;
+ device_type = "pci";
+ pcie_intc_0: interrupt-controller {
+ #address-cells = <0>;
+ #interrupt-cells = <1>;
+ interrupt-controller;
+ };
+ };
+ };
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH v10 3/3] PCI: amd-mdb: Add AMD MDB Root Port driver
2025-02-11 6:38 [PATCH v10 0/3] Add support for AMD MDB IP as Root Port Thippeswamy Havalige
2025-02-11 6:38 ` [PATCH v10 1/3] dt-bindings: PCI: dwc: Add AMD Versal2 mdb slcr support Thippeswamy Havalige
2025-02-11 6:38 ` [PATCH v10 2/3] dt-bindings: PCI: amd-mdb: Add AMD Versal2 MDB PCIe Root Port Bridge Thippeswamy Havalige
@ 2025-02-11 6:38 ` Thippeswamy Havalige
2025-02-11 18:33 ` Bjorn Helgaas
2 siblings, 1 reply; 6+ messages in thread
From: Thippeswamy Havalige @ 2025-02-11 6:38 UTC (permalink / raw)
To: bhelgaas, lpieralisi, kw, manivannan.sadhasivam, robh, krzk+dt,
conor+dt
Cc: linux-pci, devicetree, linux-kernel, jingoohan1, michal.simek,
bharat.kumar.gogada, Thippeswamy Havalige
Add support for AMD MDB (Multimedia DMA Bridge) IP core as Root Port.
The Versal2 devices include MDB Module. The integrated block for MDB along
with the integrated bridge can function as PCIe Root Port controller at
Gen5 32-Gb/s operation per lane.
Bridge supports error and legacy interrupts and are handled using platform
specific interrupt line in Versal2.
Signed-off-by: Thippeswamy Havalige <thippeswamy.havalige@amd.com>
---
changes in v2:
-------------
- Update Gen5 speed in the patch description.
- Modify Kconfig file.
- Update string _leg_ to intx.
- Get platform structure through automic variables.
- Remove _rp_ in function.
Changes in v3:
--------------
-None.
Changes in v4:
--------------
-None.
Changes in v5:
--------------
-None.
Changes in v6:
--------------
- Remove pdev automatic variable.
- Update register name to slcr.
- Fix whitespace.
- remove Spurious extra line.
- Update Legacy to INTx.
- Add space before (SLCR).
- Update menuconfig description.
Changes in v7:
--------------
- None.
Changes in v8:
--------------
- Remove inline keyword.
- Fix indentations.
- Add AMD MDB prefix to interrupt names.
- Remove Kernel doc.
- Fix return types.
- Modify dev_warn to dev_warn_once.
- Add Intx handler & callbacks.
Changes in v10:
---------------
- Add intx assert & deassert macros.
- Move amd_mdb_pcie_init_port function.
- Add kernel doc for error warning messages.
---
drivers/pci/controller/dwc/Kconfig | 11 +
drivers/pci/controller/dwc/Makefile | 1 +
drivers/pci/controller/dwc/pcie-amd-mdb.c | 477 ++++++++++++++++++++++
3 files changed, 489 insertions(+)
create mode 100644 drivers/pci/controller/dwc/pcie-amd-mdb.c
diff --git a/drivers/pci/controller/dwc/Kconfig b/drivers/pci/controller/dwc/Kconfig
index b6d6778b0698..61d119646749 100644
--- a/drivers/pci/controller/dwc/Kconfig
+++ b/drivers/pci/controller/dwc/Kconfig
@@ -27,6 +27,17 @@ config PCIE_AL
required only for DT-based platforms. ACPI platforms with the
Annapurna Labs PCIe controller don't need to enable this.
+config PCIE_AMD_MDB
+ bool "AMD MDB Versal2 PCIe Host controller"
+ depends on OF || COMPILE_TEST
+ depends on PCI && PCI_MSI
+ select PCIE_DW_HOST
+ help
+ Say Y here if you want to enable PCIe controller support on AMD
+ Versal2 SoCs. The AMD MDB Versal2 PCIe controller is based on DesignWare
+ IP and therefore the driver re-uses the Designware core functions to
+ implement the driver.
+
config PCI_MESON
tristate "Amlogic Meson PCIe controller"
default m if ARCH_MESON
diff --git a/drivers/pci/controller/dwc/Makefile b/drivers/pci/controller/dwc/Makefile
index a8308d9ea986..ae27eda6ec5e 100644
--- a/drivers/pci/controller/dwc/Makefile
+++ b/drivers/pci/controller/dwc/Makefile
@@ -3,6 +3,7 @@ obj-$(CONFIG_PCIE_DW) += pcie-designware.o
obj-$(CONFIG_PCIE_DW_HOST) += pcie-designware-host.o
obj-$(CONFIG_PCIE_DW_EP) += pcie-designware-ep.o
obj-$(CONFIG_PCIE_DW_PLAT) += pcie-designware-plat.o
+obj-$(CONFIG_PCIE_AMD_MDB) += pcie-amd-mdb.o
obj-$(CONFIG_PCIE_BT1) += pcie-bt1.o
obj-$(CONFIG_PCI_DRA7XX) += pci-dra7xx.o
obj-$(CONFIG_PCI_EXYNOS) += pci-exynos.o
diff --git a/drivers/pci/controller/dwc/pcie-amd-mdb.c b/drivers/pci/controller/dwc/pcie-amd-mdb.c
new file mode 100644
index 000000000000..eabaf943679d
--- /dev/null
+++ b/drivers/pci/controller/dwc/pcie-amd-mdb.c
@@ -0,0 +1,477 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * PCIe host controller driver for AMD MDB PCIe Bridge
+ *
+ * Copyright (C) 2024-2025, Advanced Micro Devices, Inc.
+ */
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/gpio.h>
+#include <linux/interrupt.h>
+#include <linux/irqdomain.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/of_device.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+#include <linux/resource.h>
+#include <linux/types.h>
+
+#include "pcie-designware.h"
+
+#define AMD_MDB_TLP_IR_STATUS_MISC 0x4C0
+#define AMD_MDB_TLP_IR_MASK_MISC 0x4C4
+#define AMD_MDB_TLP_IR_ENABLE_MISC 0x4C8
+
+#define AMD_MDB_TLP_PCIE_INTX_MASK GENMASK(23, 16)
+
+#define AMD_MDB_PCIE_INTX_BIT(x) FIELD_PREP(AMD_MDB_TLP_PCIE_INTX_MASK, BIT(x))
+
+#define AMD_MDB_PCIE_INTR_INTX_ASSERT(x) BIT((x) * 2)
+#define AMD_MDB_PCIE_INTR_INTX_DEASSERT(x) BIT(((x) * 2) + 1)
+
+/* Interrupt registers definitions */
+#define AMD_MDB_PCIE_INTR_CMPL_TIMEOUT 15
+#define AMD_MDB_PCIE_INTR_INTX 16
+#define AMD_MDB_PCIE_INTR_PM_PME_RCVD 24
+#define AMD_MDB_PCIE_INTR_PME_TO_ACK_RCVD 25
+#define AMD_MDB_PCIE_INTR_MISC_CORRECTABLE 26
+#define AMD_MDB_PCIE_INTR_NONFATAL 27
+#define AMD_MDB_PCIE_INTR_FATAL 28
+
+#define IMR(x) BIT(AMD_MDB_PCIE_INTR_ ##x)
+#define AMD_MDB_PCIE_IMR_ALL_MASK \
+ ( \
+ IMR(CMPL_TIMEOUT) | \
+ IMR(PM_PME_RCVD) | \
+ IMR(PME_TO_ACK_RCVD) | \
+ IMR(MISC_CORRECTABLE) | \
+ IMR(NONFATAL) | \
+ IMR(FATAL) | \
+ AMD_MDB_TLP_PCIE_INTX_MASK \
+ )
+
+/**
+ * struct amd_mdb_pcie - PCIe port information
+ * @pci: DesignWare PCIe controller structure
+ * @slcr: MDB System Level Control and Status Register (SLCR) Base
+ * @intx_domain: INTx IRQ domain pointer
+ * @mdb_domain: MDB IRQ domain pointer
+ */
+struct amd_mdb_pcie {
+ struct dw_pcie pci;
+ void __iomem *slcr;
+ struct irq_domain *intx_domain;
+ struct irq_domain *mdb_domain;
+ int intx_irq;
+};
+
+static const struct dw_pcie_host_ops amd_mdb_pcie_host_ops = {
+};
+
+static inline u32 pcie_read(struct amd_mdb_pcie *pcie, u32 reg)
+{
+ return readl_relaxed(pcie->slcr + reg);
+}
+
+static inline void pcie_write(struct amd_mdb_pcie *pcie,
+ u32 val, u32 reg)
+{
+ writel_relaxed(val, pcie->slcr + reg);
+}
+
+static void amd_mdb_intx_irq_mask(struct irq_data *data)
+{
+ struct amd_mdb_pcie *pcie = irq_data_get_irq_chip_data(data);
+ struct dw_pcie *pci = &pcie->pci;
+ struct dw_pcie_rp *port = &pci->pp;
+ unsigned long flags;
+ u32 val;
+
+ raw_spin_lock_irqsave(&port->lock, flags);
+ val = pcie_read(pcie, AMD_MDB_TLP_IR_MASK_MISC);
+ val &= ~AMD_MDB_PCIE_INTX_BIT(data->hwirq);
+ pcie_write(pcie, val, AMD_MDB_TLP_IR_ENABLE_MISC);
+ raw_spin_unlock_irqrestore(&port->lock, flags);
+}
+
+static void amd_mdb_intx_irq_unmask(struct irq_data *data)
+{
+ struct amd_mdb_pcie *pcie = irq_data_get_irq_chip_data(data);
+ struct dw_pcie *pci = &pcie->pci;
+ struct dw_pcie_rp *port = &pci->pp;
+ unsigned long flags;
+ u32 val;
+
+ raw_spin_lock_irqsave(&port->lock, flags);
+ val = pcie_read(pcie, AMD_MDB_TLP_IR_MASK_MISC);
+ val |= AMD_MDB_PCIE_INTX_BIT(data->hwirq);
+ pcie_write(pcie, val, AMD_MDB_TLP_IR_ENABLE_MISC);
+ raw_spin_unlock_irqrestore(&port->lock, flags);
+}
+
+static struct irq_chip amd_mdb_intx_irq_chip = {
+ .name = "AMD MDB INTx",
+ .irq_mask = amd_mdb_intx_irq_mask,
+ .irq_unmask = amd_mdb_intx_irq_unmask,
+};
+
+/**
+ * amd_mdb_pcie_intx_map - Set the handler for the INTx and mark IRQ
+ * as valid
+ * @domain: IRQ domain
+ * @irq: Virtual IRQ number
+ * @hwirq: HW interrupt number
+ *
+ * Return: Always returns 0.
+ */
+static int amd_mdb_pcie_intx_map(struct irq_domain *domain,
+ unsigned int irq, irq_hw_number_t hwirq)
+{
+ irq_set_chip_and_handler(irq, &amd_mdb_intx_irq_chip,
+ handle_level_irq);
+ irq_set_chip_data(irq, domain->host_data);
+ irq_set_status_flags(irq, IRQ_LEVEL);
+
+ return 0;
+}
+
+/* INTx IRQ Domain operations */
+static const struct irq_domain_ops amd_intx_domain_ops = {
+ .map = amd_mdb_pcie_intx_map,
+};
+
+static int amd_mdb_pcie_init_port(struct amd_mdb_pcie *pcie)
+{
+ int val;
+
+ /* Disable all TLP Interrupts */
+ pcie_write(pcie, pcie_read(pcie, AMD_MDB_TLP_IR_ENABLE_MISC) &
+ ~AMD_MDB_PCIE_IMR_ALL_MASK,
+ AMD_MDB_TLP_IR_ENABLE_MISC);
+
+ /* Clear pending TLP interrupts */
+ pcie_write(pcie, pcie_read(pcie, AMD_MDB_TLP_IR_STATUS_MISC) &
+ AMD_MDB_PCIE_IMR_ALL_MASK,
+ AMD_MDB_TLP_IR_STATUS_MISC);
+
+ /* Enable all TLP Interrupts */
+ val = pcie_read(pcie, AMD_MDB_TLP_IR_ENABLE_MISC);
+ pcie_write(pcie, (val | AMD_MDB_PCIE_IMR_ALL_MASK),
+ AMD_MDB_TLP_IR_ENABLE_MISC);
+
+ return 0;
+}
+
+static irqreturn_t amd_mdb_pcie_event_flow(int irq, void *args)
+{
+ struct amd_mdb_pcie *pcie = args;
+ unsigned long val;
+ int i;
+
+ val = pcie_read(pcie, AMD_MDB_TLP_IR_STATUS_MISC);
+ val &= ~pcie_read(pcie, AMD_MDB_TLP_IR_MASK_MISC);
+ for_each_set_bit(i, &val, 32)
+ generic_handle_domain_irq(pcie->mdb_domain, i);
+ pcie_write(pcie, val, AMD_MDB_TLP_IR_STATUS_MISC);
+
+ return IRQ_HANDLED;
+}
+
+#define _IC(x, s)[AMD_MDB_PCIE_INTR_ ## x] = { __stringify(x), s }
+
+static const struct {
+ const char *sym;
+ const char *str;
+} intr_cause[32] = {
+ _IC(CMPL_TIMEOUT, "completion timeout"),
+ _IC(PM_PME_RCVD, "PM_PME message received"),
+ _IC(PME_TO_ACK_RCVD, "PME_TO_ACK message received"),
+ _IC(MISC_CORRECTABLE, "Correctable error message"),
+ _IC(NONFATAL, "Non fatal error message"),
+ _IC(FATAL, "Fatal error message"),
+};
+
+static void amd_mdb_event_irq_mask(struct irq_data *d)
+{
+ struct amd_mdb_pcie *pcie = irq_data_get_irq_chip_data(d);
+ struct dw_pcie *pci = &pcie->pci;
+ struct dw_pcie_rp *port = &pci->pp;
+ u32 val;
+
+ raw_spin_lock(&port->lock);
+ val = pcie_read(pcie, AMD_MDB_TLP_IR_STATUS_MISC);
+ val &= ~BIT(d->hwirq);
+ pcie_write(pcie, val, AMD_MDB_TLP_IR_STATUS_MISC);
+ raw_spin_unlock(&port->lock);
+}
+
+static void amd_mdb_event_irq_unmask(struct irq_data *d)
+{
+ struct amd_mdb_pcie *pcie = irq_data_get_irq_chip_data(d);
+ struct dw_pcie *pci = &pcie->pci;
+ struct dw_pcie_rp *port = &pci->pp;
+ u32 val;
+
+ raw_spin_lock(&port->lock);
+ val = pcie_read(pcie, AMD_MDB_TLP_IR_STATUS_MISC);
+ val |= BIT(d->hwirq);
+ pcie_write(pcie, val, AMD_MDB_TLP_IR_STATUS_MISC);
+ raw_spin_unlock(&port->lock);
+}
+
+static struct irq_chip amd_mdb_event_irq_chip = {
+ .name = "AMD MDB RC-Event",
+ .irq_mask = amd_mdb_event_irq_mask,
+ .irq_unmask = amd_mdb_event_irq_unmask,
+};
+
+static int amd_mdb_pcie_event_map(struct irq_domain *domain,
+ unsigned int irq, irq_hw_number_t hwirq)
+{
+ irq_set_chip_and_handler(irq, &amd_mdb_event_irq_chip,
+ handle_level_irq);
+ irq_set_chip_data(irq, domain->host_data);
+ irq_set_status_flags(irq, IRQ_LEVEL);
+
+ return 0;
+}
+
+static const struct irq_domain_ops event_domain_ops = {
+ .map = amd_mdb_pcie_event_map,
+};
+
+static void amd_mdb_pcie_free_irq_domains(struct amd_mdb_pcie *pcie)
+{
+ if (pcie->intx_domain) {
+ irq_domain_remove(pcie->intx_domain);
+ pcie->intx_domain = NULL;
+ }
+
+ if (pcie->mdb_domain) {
+ irq_domain_remove(pcie->mdb_domain);
+ pcie->mdb_domain = NULL;
+ }
+}
+
+/**
+ * amd_mdb_pcie_init_irq_domains - Initialize IRQ domain
+ * @pcie: PCIe port information
+ * @pdev: platform device
+ * Return: '0' on success and error value on failure
+ */
+static int amd_mdb_pcie_init_irq_domains(struct amd_mdb_pcie *pcie,
+ struct platform_device *pdev)
+{
+ struct dw_pcie *pci = &pcie->pci;
+ struct dw_pcie_rp *pp = &pci->pp;
+ struct device *dev = &pdev->dev;
+ struct device_node *node = dev->of_node;
+ struct device_node *pcie_intc_node;
+
+ /* Setup INTx */
+ pcie_intc_node = of_get_next_child(node, NULL);
+ if (!pcie_intc_node) {
+ dev_err(dev, "No PCIe Intc node found\n");
+ return -ENODATA;
+ }
+
+ pcie->mdb_domain = irq_domain_add_linear(pcie_intc_node, 32,
+ &event_domain_ops, pcie);
+ if (!pcie->mdb_domain) {
+ dev_err(dev, "Failed to add mdb_domain\n");
+ goto out;
+ }
+
+ irq_domain_update_bus_token(pcie->mdb_domain, DOMAIN_BUS_NEXUS);
+
+ pcie->intx_domain = irq_domain_add_linear(pcie_intc_node, PCI_NUM_INTX,
+ &amd_intx_domain_ops, pcie);
+ if (!pcie->intx_domain) {
+ dev_err(dev, "Failed to add intx_domain\n");
+ goto mdb_out;
+ }
+
+ of_node_put(pcie_intc_node);
+ irq_domain_update_bus_token(pcie->intx_domain, DOMAIN_BUS_WIRED);
+
+ raw_spin_lock_init(&pp->lock);
+
+ return 0;
+mdb_out:
+ amd_mdb_pcie_free_irq_domains(pcie);
+out:
+ of_node_put(pcie_intc_node);
+
+ return -ENOMEM;
+}
+
+static irqreturn_t dw_pcie_rp_intx_flow(int irq, void *args)
+{
+ struct amd_mdb_pcie *pcie = args;
+ unsigned long val;
+ int i;
+
+ val = FIELD_GET(AMD_MDB_TLP_PCIE_INTX_MASK,
+ pcie_read(pcie, AMD_MDB_TLP_IR_STATUS_MISC));
+
+ for (i = 0; i < PCI_NUM_INTX; i++) {
+ if (val & AMD_MDB_PCIE_INTR_INTX_ASSERT(i))
+ generic_handle_domain_irq(pcie->intx_domain, i);
+ if (val & AMD_MDB_PCIE_INTR_INTX_DEASSERT(i)
+ generic_handle_domain_irq(pcie->intx_domain, (i * 2) + 1);
+ }
+
+ return IRQ_HANDLED;
+}
+
+static irqreturn_t amd_mdb_pcie_intr_handler(int irq, void *args)
+{
+ struct amd_mdb_pcie *pcie = args;
+ struct device *dev;
+ struct irq_data *d;
+
+ dev = pcie->pci.dev;
+
+ /**
+ * In future, error reporting will be hooked to the AER subsystem.
+ * Currently, the driver prints a warning message to the user.
+ */
+ d = irq_domain_get_irq_data(pcie->mdb_domain, irq);
+ if (intr_cause[d->hwirq].str)
+ dev_warn(dev, "%s\n", intr_cause[d->hwirq].str);
+ else
+ dev_warn_once(dev, "Unknown IRQ %ld\n", d->hwirq);
+
+ return IRQ_HANDLED;
+}
+
+static int amd_mdb_setup_irq(struct amd_mdb_pcie *pcie,
+ struct platform_device *pdev)
+{
+ struct dw_pcie *pci = &pcie->pci;
+ struct dw_pcie_rp *pp = &pci->pp;
+ struct device *dev = &pdev->dev;
+ int i, irq, err;
+
+ amd_mdb_pcie_init_port(pcie);
+
+ pp->irq = platform_get_irq(pdev, 0);
+ if (pp->irq < 0)
+ return pp->irq;
+
+ for (i = 0; i < ARRAY_SIZE(intr_cause); i++) {
+ if (!intr_cause[i].str)
+ continue;
+ irq = irq_create_mapping(pcie->mdb_domain, i);
+ if (!irq) {
+ dev_err(dev, "Failed to map mdb domain interrupt\n");
+ return -ENOMEM;
+ }
+ err = devm_request_irq(dev, irq, amd_mdb_pcie_intr_handler,
+ IRQF_SHARED | IRQF_NO_THREAD,
+ intr_cause[i].sym, pcie);
+ if (err) {
+ dev_err(dev, "Failed to request IRQ %d\n", irq);
+ return err;
+ }
+ }
+
+ pcie->intx_irq = irq_create_mapping(pcie->mdb_domain,
+ AMD_MDB_PCIE_INTR_INTX);
+ if (!pcie->intx_irq) {
+ dev_err(dev, "Failed to map INTx interrupt\n");
+ return -ENXIO;
+ }
+
+ err = devm_request_irq(dev, pcie->intx_irq,
+ dw_pcie_rp_intx_flow,
+ IRQF_SHARED | IRQF_NO_THREAD, NULL, pcie);
+ if (err) {
+ dev_err(dev, "Failed to request INTx IRQ %d\n", irq);
+ return err;
+ }
+
+ /* Plug the main event handler */
+ err = devm_request_irq(dev, pp->irq, amd_mdb_pcie_event_flow,
+ IRQF_SHARED | IRQF_NO_THREAD, "amd_mdb pcie_irq", pcie);
+ if (err) {
+ dev_err(dev, "Failed to request event IRQ %d\n", pp->irq);
+ return err;
+ }
+
+ return 0;
+}
+
+static int amd_mdb_add_pcie_port(struct amd_mdb_pcie *pcie,
+ struct platform_device *pdev)
+{
+ struct dw_pcie *pci = &pcie->pci;
+ struct dw_pcie_rp *pp = &pci->pp;
+ struct device *dev = &pdev->dev;
+ int ret;
+
+ pcie->slcr = devm_platform_ioremap_resource_byname(pdev, "slcr");
+ if (IS_ERR(pcie->slcr))
+ return PTR_ERR(pcie->slcr);
+
+ ret = amd_mdb_pcie_init_irq_domains(pcie, pdev);
+ if (ret)
+ return ret;
+
+ ret = amd_mdb_setup_irq(pcie, pdev);
+ if (ret) {
+ dev_err(dev, "Failed to set up interrupts\n");
+ goto out;
+ }
+
+ pp->ops = &amd_mdb_pcie_host_ops;
+
+ ret = dw_pcie_host_init(pp);
+ if (ret) {
+ dev_err(dev, "Failed to initialize host\n");
+ goto out;
+ }
+
+ return 0;
+
+out:
+ amd_mdb_pcie_free_irq_domains(pcie);
+ return ret;
+}
+
+static int amd_mdb_pcie_probe(struct platform_device *pdev)
+{
+ struct device *dev = &pdev->dev;
+ struct amd_mdb_pcie *pcie;
+ struct dw_pcie *pci;
+
+ pcie = devm_kzalloc(dev, sizeof(*pcie), GFP_KERNEL);
+ if (!pcie)
+ return -ENOMEM;
+
+ pci = &pcie->pci;
+ pci->dev = dev;
+
+ platform_set_drvdata(pdev, pcie);
+
+ return amd_mdb_add_pcie_port(pcie, pdev);
+}
+
+static const struct of_device_id amd_mdb_pcie_of_match[] = {
+ {
+ .compatible = "amd,versal2-mdb-host",
+ },
+ {},
+};
+
+static struct platform_driver amd_mdb_pcie_driver = {
+ .driver = {
+ .name = "amd-mdb-pcie",
+ .of_match_table = amd_mdb_pcie_of_match,
+ .suppress_bind_attrs = true,
+ },
+ .probe = amd_mdb_pcie_probe,
+};
+builtin_platform_driver(amd_mdb_pcie_driver);
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH v10 3/3] PCI: amd-mdb: Add AMD MDB Root Port driver
2025-02-11 6:38 ` [PATCH v10 3/3] PCI: amd-mdb: Add AMD MDB Root Port driver Thippeswamy Havalige
@ 2025-02-11 18:33 ` Bjorn Helgaas
2025-02-17 9:59 ` Havalige, Thippeswamy
0 siblings, 1 reply; 6+ messages in thread
From: Bjorn Helgaas @ 2025-02-11 18:33 UTC (permalink / raw)
To: Thippeswamy Havalige
Cc: bhelgaas, lpieralisi, kw, manivannan.sadhasivam, robh, krzk+dt,
conor+dt, linux-pci, devicetree, linux-kernel, jingoohan1,
michal.simek, bharat.kumar.gogada
On Tue, Feb 11, 2025 at 12:08:51PM +0530, Thippeswamy Havalige wrote:
> Add support for AMD MDB (Multimedia DMA Bridge) IP core as Root Port.
> +++ b/drivers/pci/controller/dwc/Kconfig
> @@ -27,6 +27,17 @@ config PCIE_AL
> required only for DT-based platforms. ACPI platforms with the
> Annapurna Labs PCIe controller don't need to enable this.
>
> +config PCIE_AMD_MDB
> + bool "AMD MDB Versal2 PCIe Host controller"
> + depends on OF || COMPILE_TEST
> + depends on PCI && PCI_MSI
> + select PCIE_DW_HOST
> + help
> + Say Y here if you want to enable PCIe controller support on AMD
> + Versal2 SoCs. The AMD MDB Versal2 PCIe controller is based on DesignWare
> + IP and therefore the driver re-uses the Designware core functions to
> + implement the driver.
Wrap to fit in 75-78 columns like the rest of the file. This gets
chopped off in an 80 column menuconfig window.
> +++ b/drivers/pci/controller/dwc/pcie-amd-mdb.c
> +#define AMD_MDB_TLP_PCIE_INTX_MASK GENMASK(23, 16)
> +
> +#define AMD_MDB_PCIE_INTX_BIT(x) FIELD_PREP(AMD_MDB_TLP_PCIE_INTX_MASK, BIT(x))
> +
> +#define AMD_MDB_PCIE_INTR_INTX_ASSERT(x) BIT((x) * 2)
> +#define AMD_MDB_PCIE_INTR_INTX_DEASSERT(x) BIT(((x) * 2) + 1)
> +static void amd_mdb_intx_irq_mask(struct irq_data *data)
> +{
> + struct amd_mdb_pcie *pcie = irq_data_get_irq_chip_data(data);
> + struct dw_pcie *pci = &pcie->pci;
> + struct dw_pcie_rp *port = &pci->pp;
> + unsigned long flags;
> + u32 val;
> +
> + raw_spin_lock_irqsave(&port->lock, flags);
> + val = pcie_read(pcie, AMD_MDB_TLP_IR_MASK_MISC);
> + val &= ~AMD_MDB_PCIE_INTX_BIT(data->hwirq);
This doesn't look right to me. hwirq should be 0, 1, 2, or 3 (INTA,
INTB, INTC, or INTD):
AMD_MDB_PCIE_INTX_BIT(0) == 0001 0000 (INTA assert)
AMD_MDB_PCIE_INTX_BIT(1) == 0002 0000 (INTA deassert)
AMD_MDB_PCIE_INTX_BIT(2) == 0004 0000 (INTB assert)
AMD_MDB_PCIE_INTX_BIT(3) == 0008 0000 (INTB deassert)
Maybe the AMD_MDB_TLP_IR_ENABLE_MISC register is laid out differently
than AMD_MDB_TLP_IR_STATUS_MISC? If so, and you're updating a
four-bit field, it needs a different GENMASK.
> + pcie_write(pcie, val, AMD_MDB_TLP_IR_ENABLE_MISC);
This *looks* like it's supposed to be a read/modify/write of
AMD_MDB_TLP_IR_MASK_MISC, but you read AMD_MDB_TLP_IR_MASK_MISC and
then write AMD_MDB_TLP_IR_ENABLE_MISC. Same below in
amd_mdb_intx_irq_unmask().
> + raw_spin_unlock_irqrestore(&port->lock, flags);
> +}
> +
> +static void amd_mdb_intx_irq_unmask(struct irq_data *data)
> +{
> + struct amd_mdb_pcie *pcie = irq_data_get_irq_chip_data(data);
> + struct dw_pcie *pci = &pcie->pci;
> + struct dw_pcie_rp *port = &pci->pp;
> + unsigned long flags;
> + u32 val;
> +
> + raw_spin_lock_irqsave(&port->lock, flags);
> + val = pcie_read(pcie, AMD_MDB_TLP_IR_MASK_MISC);
> + val |= AMD_MDB_PCIE_INTX_BIT(data->hwirq);
> + pcie_write(pcie, val, AMD_MDB_TLP_IR_ENABLE_MISC);
> + raw_spin_unlock_irqrestore(&port->lock, flags);
> +}
> +static irqreturn_t dw_pcie_rp_intx_flow(int irq, void *args)
It'd be nice if this were close in the source file to the INTx
mask/unmask above.
> +{
> + struct amd_mdb_pcie *pcie = args;
> + unsigned long val;
> + int i;
> +
> + val = FIELD_GET(AMD_MDB_TLP_PCIE_INTX_MASK,
> + pcie_read(pcie, AMD_MDB_TLP_IR_STATUS_MISC));
> +
> + for (i = 0; i < PCI_NUM_INTX; i++) {
> + if (val & AMD_MDB_PCIE_INTR_INTX_ASSERT(i))
> + generic_handle_domain_irq(pcie->intx_domain, i);
> + if (val & AMD_MDB_PCIE_INTR_INTX_DEASSERT(i)
> + generic_handle_domain_irq(pcie->intx_domain, (i * 2) + 1);
Why call generic_handle_domain_irq() for deassert? No other drivers
do that AFAIK. If you do need it, "(i * 2) + 1" looks completely
wrong; it should be the hwirq value.
> + }
> +
> + return IRQ_HANDLED;
> +}
> +static irqreturn_t amd_mdb_pcie_intr_handler(int irq, void *args)
> +{
> + struct amd_mdb_pcie *pcie = args;
> + struct device *dev;
> + struct irq_data *d;
> +
> + dev = pcie->pci.dev;
> +
> + /**
/* (not /**)
> + * In future, error reporting will be hooked to the AER subsystem.
> + * Currently, the driver prints a warning message to the user.
> + */
> + d = irq_domain_get_irq_data(pcie->mdb_domain, irq);
> + if (intr_cause[d->hwirq].str)
> + dev_warn(dev, "%s\n", intr_cause[d->hwirq].str);
> + else
> + dev_warn_once(dev, "Unknown IRQ %ld\n", d->hwirq);
> +
> + return IRQ_HANDLED;
> +}
> +static int amd_mdb_setup_irq(struct amd_mdb_pcie *pcie,
> + struct platform_device *pdev)
> +{
> ...
> +
> + /* Plug the main event handler */
> + err = devm_request_irq(dev, pp->irq, amd_mdb_pcie_event_flow,
> + IRQF_SHARED | IRQF_NO_THREAD, "amd_mdb pcie_irq", pcie);
Wrap to fit in 80 columns like the rest of the file.
Bjorn
^ permalink raw reply [flat|nested] 6+ messages in thread
* RE: [PATCH v10 3/3] PCI: amd-mdb: Add AMD MDB Root Port driver
2025-02-11 18:33 ` Bjorn Helgaas
@ 2025-02-17 9:59 ` Havalige, Thippeswamy
0 siblings, 0 replies; 6+ messages in thread
From: Havalige, Thippeswamy @ 2025-02-17 9:59 UTC (permalink / raw)
To: Bjorn Helgaas
Cc: bhelgaas@google.com, lpieralisi@kernel.org, kw@linux.com,
manivannan.sadhasivam@linaro.org, robh@kernel.org,
krzk+dt@kernel.org, conor+dt@kernel.org,
linux-pci@vger.kernel.org, devicetree@vger.kernel.org,
linux-kernel@vger.kernel.org, jingoohan1@gmail.com, Simek, Michal,
Gogada, Bharat Kumar
Hi Bjorn,
> -----Original Message-----
> From: Bjorn Helgaas <helgaas@kernel.org>
> Sent: Wednesday, February 12, 2025 12:04 AM
> To: Havalige, Thippeswamy <thippeswamy.havalige@amd.com>
> Cc: bhelgaas@google.com; lpieralisi@kernel.org; kw@linux.com;
> manivannan.sadhasivam@linaro.org; robh@kernel.org; krzk+dt@kernel.org;
> conor+dt@kernel.org; linux-pci@vger.kernel.org; devicetree@vger.kernel.org;
> linux-kernel@vger.kernel.org; jingoohan1@gmail.com; Simek, Michal
> <michal.simek@amd.com>; Gogada, Bharat Kumar
> <bharat.kumar.gogada@amd.com>
> Subject: Re: [PATCH v10 3/3] PCI: amd-mdb: Add AMD MDB Root Port driver
>
> On Tue, Feb 11, 2025 at 12:08:51PM +0530, Thippeswamy Havalige wrote:
> > Add support for AMD MDB (Multimedia DMA Bridge) IP core as Root Port.
>
> > +++ b/drivers/pci/controller/dwc/Kconfig
> > @@ -27,6 +27,17 @@ config PCIE_AL
> > required only for DT-based platforms. ACPI platforms with the
> > Annapurna Labs PCIe controller don't need to enable this.
> >
> > +config PCIE_AMD_MDB
> > + bool "AMD MDB Versal2 PCIe Host controller"
> > + depends on OF || COMPILE_TEST
> > + depends on PCI && PCI_MSI
> > + select PCIE_DW_HOST
> > + help
> > + Say Y here if you want to enable PCIe controller support on AMD
> > + Versal2 SoCs. The AMD MDB Versal2 PCIe controller is based on
> DesignWare
> > + IP and therefore the driver re-uses the Designware core functions to
> > + implement the driver.
>
> Wrap to fit in 75-78 columns like the rest of the file. This gets
> chopped off in an 80 column menuconfig window.
Thank you for reviewing, I will update it in next patch.
>
> > +++ b/drivers/pci/controller/dwc/pcie-amd-mdb.c
>
> > +#define AMD_MDB_TLP_PCIE_INTX_MASK GENMASK(23, 16)
> > +
> > +#define AMD_MDB_PCIE_INTX_BIT(x)
> FIELD_PREP(AMD_MDB_TLP_PCIE_INTX_MASK, BIT(x))
> > +
> > +#define AMD_MDB_PCIE_INTR_INTX_ASSERT(x) BIT((x) * 2)
> > +#define AMD_MDB_PCIE_INTR_INTX_DEASSERT(x) BIT(((x) * 2) + 1)
>
> > +static void amd_mdb_intx_irq_mask(struct irq_data *data)
> > +{
> > + struct amd_mdb_pcie *pcie = irq_data_get_irq_chip_data(data);
> > + struct dw_pcie *pci = &pcie->pci;
> > + struct dw_pcie_rp *port = &pci->pp;
> > + unsigned long flags;
> > + u32 val;
> > +
> > + raw_spin_lock_irqsave(&port->lock, flags);
> > + val = pcie_read(pcie, AMD_MDB_TLP_IR_MASK_MISC);
> > + val &= ~AMD_MDB_PCIE_INTX_BIT(data->hwirq);
>
> This doesn't look right to me. hwirq should be 0, 1, 2, or 3 (INTA,
> INTB, INTC, or INTD):
Thank you for reviewing,
Will update macro to following
#define AMD_MDB_PCIE_INTX_BIT(x) (1U << (2 * (x) + AMD_MDB_PCIE_INTR_INTX)).
>
> AMD_MDB_PCIE_INTX_BIT(0) == 0001 0000 (INTA assert)
> AMD_MDB_PCIE_INTX_BIT(1) == 0002 0000 (INTA deassert)
> AMD_MDB_PCIE_INTX_BIT(2) == 0004 0000 (INTB assert)
> AMD_MDB_PCIE_INTX_BIT(3) == 0008 0000 (INTB deassert)
>
> Maybe the AMD_MDB_TLP_IR_ENABLE_MISC register is laid out differently
> than AMD_MDB_TLP_IR_STATUS_MISC? If so, and you're updating a
> four-bit field, it needs a different GENMASK.
>
> > + pcie_write(pcie, val, AMD_MDB_TLP_IR_ENABLE_MISC);
>
> This *looks* like it's supposed to be a read/modify/write of
> AMD_MDB_TLP_IR_MASK_MISC, but you read
> AMD_MDB_TLP_IR_MASK_MISC and
> then write AMD_MDB_TLP_IR_ENABLE_MISC. Same below in
> amd_mdb_intx_irq_unmask().
- Thank you for reviewing, Here *_STATUS/ENABLE/MASK_MISC register are
Laid in the same way.
>
> > + raw_spin_unlock_irqrestore(&port->lock, flags);
> > +}
> > +
> > +static void amd_mdb_intx_irq_unmask(struct irq_data *data)
> > +{
> > + struct amd_mdb_pcie *pcie = irq_data_get_irq_chip_data(data);
> > + struct dw_pcie *pci = &pcie->pci;
> > + struct dw_pcie_rp *port = &pci->pp;
> > + unsigned long flags;
> > + u32 val;
> > +
> > + raw_spin_lock_irqsave(&port->lock, flags);
> > + val = pcie_read(pcie, AMD_MDB_TLP_IR_MASK_MISC);
> > + val |= AMD_MDB_PCIE_INTX_BIT(data->hwirq);
> > + pcie_write(pcie, val, AMD_MDB_TLP_IR_ENABLE_MISC);
> > + raw_spin_unlock_irqrestore(&port->lock, flags);
> > +}
>
> > +static irqreturn_t dw_pcie_rp_intx_flow(int irq, void *args)
>
> It'd be nice if this were close in the source file to the INTx
> mask/unmask above.
Thank you for reviewing, Will move this in next patch.
>
> > +{
> > + struct amd_mdb_pcie *pcie = args;
> > + unsigned long val;
> > + int i;
> > +
> > + val = FIELD_GET(AMD_MDB_TLP_PCIE_INTX_MASK,
> > + pcie_read(pcie, AMD_MDB_TLP_IR_STATUS_MISC));
> > +
> > + for (i = 0; i < PCI_NUM_INTX; i++) {
> > + if (val & AMD_MDB_PCIE_INTR_INTX_ASSERT(i))
> > + generic_handle_domain_irq(pcie->intx_domain, i);
> > + if (val & AMD_MDB_PCIE_INTR_INTX_DEASSERT(i)
> > + generic_handle_domain_irq(pcie->intx_domain, (i *
> 2) + 1);
>
> Why call generic_handle_domain_irq() for deassert? No other drivers
> do that AFAIK. If you do need it, "(i * 2) + 1" looks completely
> wrong; it should be the hwirq value.
Thank you for reviewing, will remove this deassert generic handler in next patch.
>
> > + }
> > +
> > + return IRQ_HANDLED;
> > +}
>
> > +static irqreturn_t amd_mdb_pcie_intr_handler(int irq, void *args)
> > +{
> > + struct amd_mdb_pcie *pcie = args;
> > + struct device *dev;
> > + struct irq_data *d;
> > +
> > + dev = pcie->pci.dev;
> > +
> > + /**
>
> /* (not /**)
- Thank you for reviewing, Will update this in next patch.
>
> > + * In future, error reporting will be hooked to the AER subsystem.
> > + * Currently, the driver prints a warning message to the user.
> > + */
> > + d = irq_domain_get_irq_data(pcie->mdb_domain, irq);
> > + if (intr_cause[d->hwirq].str)
> > + dev_warn(dev, "%s\n", intr_cause[d->hwirq].str);
> > + else
> > + dev_warn_once(dev, "Unknown IRQ %ld\n", d->hwirq);
> > +
> > + return IRQ_HANDLED;
> > +}
>
> > +static int amd_mdb_setup_irq(struct amd_mdb_pcie *pcie,
> > + struct platform_device *pdev)
> > +{
> > ...
>
> > +
> > + /* Plug the main event handler */
> > + err = devm_request_irq(dev, pp->irq, amd_mdb_pcie_event_flow,
> > + IRQF_SHARED | IRQF_NO_THREAD, "amd_mdb
> pcie_irq", pcie);
>
> Wrap to fit in 80 columns like the rest of the file.
-Thank you for reviewing, will update this in next patch.
>
> Bjorn
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2025-02-17 10:00 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-02-11 6:38 [PATCH v10 0/3] Add support for AMD MDB IP as Root Port Thippeswamy Havalige
2025-02-11 6:38 ` [PATCH v10 1/3] dt-bindings: PCI: dwc: Add AMD Versal2 mdb slcr support Thippeswamy Havalige
2025-02-11 6:38 ` [PATCH v10 2/3] dt-bindings: PCI: amd-mdb: Add AMD Versal2 MDB PCIe Root Port Bridge Thippeswamy Havalige
2025-02-11 6:38 ` [PATCH v10 3/3] PCI: amd-mdb: Add AMD MDB Root Port driver Thippeswamy Havalige
2025-02-11 18:33 ` Bjorn Helgaas
2025-02-17 9:59 ` Havalige, Thippeswamy
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox