Skip to contentSkip to conversation

pci: iproc: Add Broadcom iProc PCIe support

PATCH Superseded Ray Jui 4 participants 12 messages Mar 10, 2015 – Mar 11, 2015

Subsystems (4)

v5 of 9 Superseded Latest is v9 →

lore
Share & tools
lore

Ray Jui

author maintainer
#0 lore raw

v5 cover pci: iproc: Add Broadcom iProc PCIe support

This patch series adds the support for Broadcom iProc PCIe controller

pcie-iproc.c servers as the common core driver, and front-end bus interface needs to be added to support different bus interfaces

pcie-iproc-pltfm.c contains the support for the platform bus interface

Changes from v4:
- iProc PCIe driver module support was not included in patch series v4. This patch series (v5) fixes it

Changes from v3:
- Export symbols of several PCI functions so they can be used by drivers compiled as kernel module
- Add additional support to the Broadcom iProc PCIe driver so it can be installed/uninstalled as kernel loadable module

Changes from v2:
- Major rework of the PCIe driver to factor out common generic code from front-end bus interface. Support for generic platform bus interface is also added
- Adapt to several new PCI APIs that have been introduced lately

Changes from v1:
- Add standard PCI interrupt DT properties "#interrupt-cells", "interrupt-map-mask" and "interrupt-map" so legacy INTx interrupts can be supported by using standard PCI OF IRQ parsing function
- Get rid of custom IRQ mapping function in the driver. Use of_irq_parse_and_map_pci instead

Ray Jui (4):
PCI: Export symbols of PCI functions
pci: iProc: define iProc PCIe platform bus binding
pci: iproc: Add Broadcom iProc PCIe support
ARM: dts: enable PCIe support for Cygnus

 .../devicetree/bindings/pci/brcm,iproc-pcie.txt    |   63 ++++
 arch/arm/boot/dts/bcm-cygnus.dtsi                  |   42 +++
 arch/arm/boot/dts/bcm958300k.dts                   |    8 +
 drivers/pci/host/Kconfig                           |   17 +
 drivers/pci/host/Makefile                          |    2 +
 drivers/pci/host/pcie-iproc-pltfm.c                |  108 ++++++
 drivers/pci/host/pcie-iproc.c                      |  351 ++++++++++++++++++++
 drivers/pci/host/pcie-iproc.h                      |   42 +++
 drivers/pci/pci.c                                  |    1 +
 drivers/pci/probe.c                                |    1 +
 drivers/pci/remove.c                               |    2 +
 drivers/pci/setup-bus.c                            |    1 +
 drivers/pci/setup-irq.c                            |    1 +
 13 files changed, 639 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/pci/brcm,iproc-pcie.txt
 create mode 100644 drivers/pci/host/pcie-iproc-pltfm.c
 create mode 100644 drivers/pci/host/pcie-iproc.c
 create mode 100644 drivers/pci/host/pcie-iproc.h
Signature · 1 line
1.7.9.5
author maintainer
#1 lore raw

PATCH v5 1/4 PCI: Export symbols of PCI functions

Export symbols of the following PCI functions so they can be referenced by a PCI driver compiled as a kernel loadable module:

pci_common_swizzle
pci_create_root_bus
pci_stop_root_bus
pci_remove_root_bus
pci_assign_unassigned_bus_resources
pci_fixup_irqs

---
 drivers/pci/pci.c       |    1 +
 drivers/pci/probe.c     |    1 +
 drivers/pci/remove.c    |    2 ++
 drivers/pci/setup-bus.c |    1 +
 drivers/pci/setup-irq.c |    1 +
 5 files changed, 6 insertions(+)
drivers/pci/pci.c +1 -0
index 81f06e8..fb4a3da 100644
@@ -2492,6 +2492,7 @@ u8 pci_common_swizzle(struct pci_dev *dev, u8 *pinp)
 	*pinp = pin;
 	return PCI_SLOT(dev->devfn);
 }
+EXPORT_SYMBOL(pci_common_swizzle);
 
 /**
  *	pci_release_region - Release a PCI bar
drivers/pci/probe.c +1 -0
index 8d2f400..eb24ef9 100644
@@ -1993,6 +1993,7 @@ err_out:
 	kfree(b);
 	return NULL;
 }
+EXPORT_SYMBOL(pci_create_root_bus);
 
 int pci_bus_insert_busn_res(struct pci_bus *b, int bus, int bus_max)
 {
drivers/pci/remove.c +2 -0
index 8bd76c9..c523159 100644
@@ -139,6 +139,7 @@ void pci_stop_root_bus(struct pci_bus *bus)
 	/* stop the host bridge */
 	device_release_driver(&host_bridge->dev);
 }
+EXPORT_SYMBOL(pci_stop_root_bus);
 
 void pci_remove_root_bus(struct pci_bus *bus)
 {
@@ -158,3 +159,4 @@ void pci_remove_root_bus(struct pci_bus *bus)
 	/* remove the host bridge */
 	device_unregister(&host_bridge->dev);
 }
+EXPORT_SYMBOL(pci_remove_root_bus);
drivers/pci/setup-bus.c +1 -0
index e3e17f3..932c9e5 100644
@@ -1750,3 +1750,4 @@ void pci_assign_unassigned_bus_resources(struct pci_bus *bus)
 	__pci_bus_assign_resources(bus, &add_list, NULL);
 	BUG_ON(!list_empty(&add_list));
 }
+EXPORT_SYMBOL(pci_assign_unassigned_bus_resources);
drivers/pci/setup-irq.c +1 -0
index 4e2d595..9d2bbb6 100644
@@ -65,3 +65,4 @@ void pci_fixup_irqs(u8 (*swizzle)(struct pci_dev *, u8 *),
 	for_each_pci_dev(dev)
 		pdev_fixup_irq(dev, swizzle, map_irq);
 }
+EXPORT_SYMBOL(pci_fixup_irqs);
Signature · 1 line
1.7.9.5
maintainer
#2 lore raw

Re: [PATCH v5 1/4] PCI: Export symbols of PCI functions

Quoted 9 lines · Ray Jui

Export symbols of the following PCI functions so they can be referenced by a PCI driver compiled as a kernel loadable module:

pci_common_swizzle
pci_create_root_bus
pci_stop_root_bus
pci_remove_root_bus
pci_assign_unassigned_bus_resources
pci_fixup_irqs

Have you tested your driver as a loadable module?

If it works to load/unload/reload/etc., I'm OK with exporting these symbols. I would prefer EXPORT_SYMBOL_GPL, though, because these don't constitute a very coherent interface right now, and they will likely be reworked. For example, see Yijing Wang's work to sort out the pci_bus_add_devices() / resource assignment mess.

Bjorn

Quoted 71 lines
---
 drivers/pci/pci.c       |    1 +
 drivers/pci/probe.c     |    1 +
 drivers/pci/remove.c    |    2 ++
 drivers/pci/setup-bus.c |    1 +
 drivers/pci/setup-irq.c |    1 +
 5 files changed, 6 insertions(+)
drivers/pci/pci.c +1 -0
index 81f06e8..fb4a3da 100644
@@ -2492,6 +2492,7 @@ u8 pci_common_swizzle(struct pci_dev *dev, u8 *pinp)
 	*pinp = pin;
 	return PCI_SLOT(dev->devfn);
 }
+EXPORT_SYMBOL(pci_common_swizzle);
 
 /**
  *	pci_release_region - Release a PCI bar
drivers/pci/probe.c +1 -0
index 8d2f400..eb24ef9 100644
@@ -1993,6 +1993,7 @@ err_out:
 	kfree(b);
 	return NULL;
 }
+EXPORT_SYMBOL(pci_create_root_bus);
 
 int pci_bus_insert_busn_res(struct pci_bus *b, int bus, int bus_max)
 {
drivers/pci/remove.c +2 -0
index 8bd76c9..c523159 100644
@@ -139,6 +139,7 @@ void pci_stop_root_bus(struct pci_bus *bus)
 	/* stop the host bridge */
 	device_release_driver(&host_bridge->dev);
 }
+EXPORT_SYMBOL(pci_stop_root_bus);
 
 void pci_remove_root_bus(struct pci_bus *bus)
 {
@@ -158,3 +159,4 @@ void pci_remove_root_bus(struct pci_bus *bus)
 	/* remove the host bridge */
 	device_unregister(&host_bridge->dev);
 }
+EXPORT_SYMBOL(pci_remove_root_bus);
drivers/pci/setup-bus.c +1 -0
index e3e17f3..932c9e5 100644
@@ -1750,3 +1750,4 @@ void pci_assign_unassigned_bus_resources(struct pci_bus *bus)
 	__pci_bus_assign_resources(bus, &add_list, NULL);
 	BUG_ON(!list_empty(&add_list));
 }
+EXPORT_SYMBOL(pci_assign_unassigned_bus_resources);
drivers/pci/setup-irq.c +1 -0
index 4e2d595..9d2bbb6 100644
@@ -65,3 +65,4 @@ void pci_fixup_irqs(u8 (*swizzle)(struct pci_dev *, u8 *),
 	for_each_pci_dev(dev)
 		pdev_fixup_irq(dev, swizzle, map_irq);
 }
+EXPORT_SYMBOL(pci_fixup_irqs);
Signature · 1 line
1.7.9.5
author maintainer
#3 lore raw

Re: [PATCH v5 1/4] PCI: Export symbols of PCI functions

Hi Bjorn,

Quoted 13 lines · Bjorn Helgaas
Quoted 9 lines · Ray Jui

Export symbols of the following PCI functions so they can be referenced by a PCI driver compiled as a kernel loadable module:

pci_common_swizzle
pci_create_root_bus
pci_stop_root_bus
pci_remove_root_bus
pci_assign_unassigned_bus_resources
pci_fixup_irqs

Have you tested your driver as a loadable module?

Quoted 7 lines

If it works to load/unload/reload/etc., I'm OK with exporting these symbols. I would prefer EXPORT_SYMBOL_GPL, though, because these don't constitute a very coherent interface right now, and they will likely be reworked. For example, see Yijing Wang's work to sort out the pci_bus_add_devices() / resource assignment mess.

Bjorn

Yes, I tested the driver with load/unload/reload on Cygnus. Console logs are included in my reply to Arnd's email from another email thread.

I'll wait for more feedback from you on the iProc PCIe driver itself. But if you have reviewed them and think they are okay then I can go ahead and submit v6 with changes to use EXPORT_SYMBOL_GPL.

Let me know.

Thanks,

Ray

author maintainer
#4 lore raw

PATCH v5 2/4 pci: iProc: define iProc PCIe platform bus binding

In reply to Ray Jui

Document the Broadcom iProc PCIe platform interface device tree binding

---
 .../devicetree/bindings/pci/brcm,iproc-pcie.txt    |   63 ++++++++++++++++++++
 1 file changed, 63 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/pci/brcm,iproc-pcie.txt
Documentation/devicetree/bindings/pci/brcm,iproc-pcie.txt new file +63 -0
new file mode 100644
index 0000000..f7ce50e
@@ -0,0 +1,63 @@
+* Broadcom iProc PCIe controller with the platform bus interface
+
+Required properties:
+- compatible: Must be "brcm,iproc-pcie"
+- reg: base address and length of the PCIe controller I/O register space
+- #interrupt-cells: set to <1>
+- interrupt-map-mask and interrupt-map, standard PCI properties to define the
+  mapping of the PCIe interface to interrupt numbers
+- linux,pci-domain: PCI domain ID. Should be unique for each host controller
+- bus-range: PCI bus numbers covered
+- #address-cells: set to <3>
+- #size-cells: set to <2>
+- device_type: set to "pci"
+- ranges: ranges for the PCI memory and I/O regions
+
+Optional properties:
+- phys: phandle of the PCIe PHY device
+- phy-names: must be "pcie-phy"
+
+Example:
+	pcie0: pcie@18012000 {
+		compatible = "brcm,iproc-pcie";
+		reg = <0x18012000 0x1000>;
+
+		#interrupt-cells = <1>;
+		interrupt-map-mask = <0 0 0 0>;
+		interrupt-map = <0 0 0 0 &gic GIC_SPI 100 IRQ_TYPE_NONE>;
+
+		linux,pci-domain = <0>;
+
+		bus-range = <0x00 0xff>;
+
+		#address-cells = <3>;
+		#size-cells = <2>;
+		device_type = "pci";
+		ranges = <0x81000000 0 0	  0x28000000 0 0x00010000
+			  0x82000000 0 0x20000000 0x20000000 0 0x04000000>;
+
+		phys = <&phy 0 5>;
+		phy-names = "pcie-phy";
+	};
+
+	pcie1: pcie@18013000 {
+		compatible = "brcm,iproc-pcie";
+		reg = <0x18013000 0x1000>;
+
+		#interrupt-cells = <1>;
+		interrupt-map-mask = <0 0 0 0>;
+		interrupt-map = <0 0 0 0 &gic GIC_SPI 106 IRQ_TYPE_NONE>;
+
+		linux,pci-domain = <1>;
+
+		bus-range = <0x00 0xff>;
+
+		#address-cells = <3>;
+		#size-cells = <2>;
+		device_type = "pci";
+		ranges = <0x81000000 0 0	  0x48000000 0 0x00010000
+			  0x82000000 0 0x40000000 0x40000000 0 0x04000000>;
+
+		phys = <&phy 1 6>;
+		phy-names = "pcie-phy";
+	};
Signature · 1 line
1.7.9.5
author maintainer
#5 lore raw

PATCH v5 3/4 pci: iproc: Add Broadcom iProc PCIe support

In reply to Ray Jui

This adds the support for Broadcom iProc PCIe controller

pcie-iproc.c servers as the common core driver, and front-end bus interface needs to be added to support different bus interfaces

pcie-iproc-pltfm.c contains the support for the platform bus interface

---
 drivers/pci/host/Kconfig            |   17 ++
 drivers/pci/host/Makefile           |    2 +
 drivers/pci/host/pcie-iproc-pltfm.c |  108 +++++++++++
 drivers/pci/host/pcie-iproc.c       |  351 +++++++++++++++++++++++++++++++++++
 drivers/pci/host/pcie-iproc.h       |   42 +++++
 5 files changed, 520 insertions(+)
 create mode 100644 drivers/pci/host/pcie-iproc-pltfm.c
 create mode 100644 drivers/pci/host/pcie-iproc.c
 create mode 100644 drivers/pci/host/pcie-iproc.h
drivers/pci/host/Kconfig +17 -0
index 7b892a9..f4d9c90 100644
@@ -106,4 +106,21 @@ config PCI_VERSATILE
 	bool "ARM Versatile PB PCI controller"
 	depends on ARCH_VERSATILE
 
+config PCIE_IPROC
+	tristate "Broadcom iProc PCIe controller"
+	help
+	  This enables the iProc PCIe core controller support for Broadcom's
+	  iProc family of SoCs. An appropriate bus interface driver also needs
+	  to be enabled
+
+config PCIE_IPROC_PLTFM
+	tristate "Broadcom iProc PCIe platform bus driver"
+	depends on ARCH_BCM_IPROC || COMPILE_TEST
+	depends on OF
+	select PCIE_IPROC
+	default ARCH_BCM_IPROC
+	help
+	  Say Y here if you want to use the Broadcom iProc PCIe controller
+	  through the generic platform bus interface
+
 endmenu
drivers/pci/host/Makefile +2 -0
index e61d91c..2e02d20 100644
@@ -13,3 +13,5 @@ obj-$(CONFIG_PCIE_XILINX) += pcie-xilinx.o
 obj-$(CONFIG_PCI_XGENE) += pci-xgene.o
 obj-$(CONFIG_PCI_LAYERSCAPE) += pci-layerscape.o
 obj-$(CONFIG_PCI_VERSATILE) += pci-versatile.o
+obj-$(CONFIG_PCIE_IPROC) += pcie-iproc.o
+obj-$(CONFIG_PCIE_IPROC_PLTFM) += pcie-iproc-pltfm.o
drivers/pci/host/pcie-iproc-pltfm.c new file +108 -0

108 lines not on this page · raw message

drivers/pci/host/pcie-iproc.c new file +351 -0

351 lines not on this page · raw message

drivers/pci/host/pcie-iproc.h new file +42 -0

42 lines not on this page · raw message

Signature · 1 line
1.7.9.5
maintainer
#6 lore raw

[+cc Rob, Yijing]

Quoted 34 lines · Ray Jui

This adds the support for Broadcom iProc PCIe controller

pcie-iproc.c servers as the common core driver, and front-end bus interface needs to be added to support different bus interfaces

pcie-iproc-pltfm.c contains the support for the platform bus interface

---
 drivers/pci/host/Kconfig            |   17 ++
 drivers/pci/host/Makefile           |    2 +
 drivers/pci/host/pcie-iproc-pltfm.c |  108 +++++++++++
 drivers/pci/host/pcie-iproc.c       |  351 +++++++++++++++++++++++++++++++++++
 drivers/pci/host/pcie-iproc.h       |   42 +++++
 5 files changed, 520 insertions(+)
 create mode 100644 drivers/pci/host/pcie-iproc-pltfm.c
 create mode 100644 drivers/pci/host/pcie-iproc.c
 create mode 100644 drivers/pci/host/pcie-iproc.h
drivers/pci/host/Kconfig +6 -0
index 7b892a9..f4d9c90 100644
@@ -106,4 +106,21 @@ config PCI_VERSATILE
 	bool "ARM Versatile PB PCI controller"
 	depends on ARCH_VERSATILE
 
+config PCIE_IPROC
+	tristate "Broadcom iProc PCIe controller"
+	help
+	  This enables the iProc PCIe core controller support for Broadcom's
+	  iProc family of SoCs. An appropriate bus interface driver also needs
+	  to be enabled

drivers/pci/host/Kconfig config PCI_VERSATILE · line 124

Quoted 10 lines
drivers/pci/host/Kconfig +10 -0
+
+config PCIE_IPROC_PLTFM
+	tristate "Broadcom iProc PCIe platform bus driver"
+	depends on ARCH_BCM_IPROC || COMPILE_TEST
+	depends on OF
+	select PCIE_IPROC
+	default ARCH_BCM_IPROC
+	help
+	  Say Y here if you want to use the Broadcom iProc PCIe controller
+	  through the generic platform bus interface

Do you anticipate additional front-end bus interfaces? If not, and maybe even if you do, you might squash everything into pcie-iproc.c. Then you only need one file (no .h file needed) and the package is a little simpler. I think it's pretty common to have multiple driver registration methods in the same file (OF, PCI, ACPI, etc.) And I think it's common to have those methods guarded by the generic config symbol (CONFIG_PCI, CONFIG_OF, etc.) rather than defining new ones specific to the driver.

Quoted 195 lines
drivers/pci/host/Kconfig +1 -0
+
 endmenu
drivers/pci/host/Makefile +2 -0
index e61d91c..2e02d20 100644
@@ -13,3 +13,5 @@ obj-$(CONFIG_PCIE_XILINX) += pcie-xilinx.o
 obj-$(CONFIG_PCI_XGENE) += pci-xgene.o
 obj-$(CONFIG_PCI_LAYERSCAPE) += pci-layerscape.o
 obj-$(CONFIG_PCI_VERSATILE) += pci-versatile.o
+obj-$(CONFIG_PCIE_IPROC) += pcie-iproc.o
+obj-$(CONFIG_PCIE_IPROC_PLTFM) += pcie-iproc-pltfm.o
drivers/pci/host/pcie-iproc-pltfm.c new file +108 -0
new file mode 100644
index 0000000..af935ea
@@ -0,0 +1,108 @@
+/*
+ * Copyright (C) 2015 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/clk.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/of_address.h>
+#include <linux/of_pci.h>
+#include <linux/of_irq.h>
+#include <linux/of_platform.h>
+#include <linux/phy/phy.h>
+
+#include "pcie-iproc.h"
+
+static int __init iproc_pcie_pltfm_probe(struct platform_device *pdev)
+{
+	struct iproc_pcie *pcie;
+	struct device_node *np = pdev->dev.of_node;
+	struct resource reg;
+	resource_size_t iobase = 0;
+	LIST_HEAD(res);
+	int ret;
+
+	pcie = devm_kzalloc(&pdev->dev, sizeof(struct iproc_pcie), GFP_KERNEL);
+	if (!pcie)
+		return -ENOMEM;
+
+	pcie->dev = &pdev->dev;
+	platform_set_drvdata(pdev, pcie);
+
+	ret = of_address_to_resource(np, 0, &reg);
+	if (ret < 0) {
+		dev_err(pcie->dev, "unable to obtain controller resources\n");
+		return ret;
+	}
+
+	pcie->base = devm_ioremap(pcie->dev, reg.start, resource_size(&reg));
+	if (!pcie->base) {
+		dev_err(pcie->dev, "unable to map controller registers\n");
+		return -ENOMEM;
+	}
+
+	/* PHY use is optional */
+	pcie->phy = devm_phy_get(&pdev->dev, "pcie-phy");
+	if (IS_ERR(pcie->phy)) {
+		if (PTR_ERR(pcie->phy) == -EPROBE_DEFER)
+			return -EPROBE_DEFER;
+		pcie->phy = NULL;
+	}
+
+	ret = of_pci_get_host_bridge_resources(np, 0, 0xff, &res, &iobase);
+	if (ret) {
+		dev_err(pcie->dev,
+			"unable to get PCI host bridge resources\n");
+		return ret;
+	}
+
+	pcie->resources = &res;
+
+	ret = iproc_pcie_setup(pcie);
+	if (ret) {
+		dev_err(pcie->dev, "PCIe controller setup failed\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int iproc_pcie_pltfm_remove(struct platform_device *pdev)
+{
+	struct iproc_pcie *pcie = platform_get_drvdata(pdev);
+
+	return iproc_pcie_remove(pcie);
+}
+
+static const struct of_device_id iproc_pcie_of_match_table[] = {
+	{ .compatible = "brcm,iproc-pcie", },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, iproc_pcie_of_match_table);
+
+static struct platform_driver iproc_pcie_pltfm_driver = {
+	.driver = {
+		.name = "iproc-pcie",
+		.of_match_table = of_match_ptr(iproc_pcie_of_match_table),
+	},
+	.probe = iproc_pcie_pltfm_probe,
+	.remove = iproc_pcie_pltfm_remove,
+};
+module_platform_driver(iproc_pcie_pltfm_driver);
+
+MODULE_AUTHOR("Ray Jui <[email protected]>");
+MODULE_DESCRIPTION("Broadcom iPROC PCIe platform driver");
+MODULE_LICENSE("GPL v2");
drivers/pci/host/pcie-iproc.c new file +63 -0
new file mode 100644
index 0000000..ad6c89e
@@ -0,0 +1,351 @@
+/*
+ * Copyright (C) 2014 Hauke Mehrtens <[email protected]>
+ * Copyright (C) 2015 Broadcom Corporatcommon ion
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/msi.h>
+#include <linux/clk.h>
+#include <linux/module.h>
+#include <linux/mbus.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/of_address.h>
+#include <linux/of_pci.h>
+#include <linux/of_irq.h>
+#include <linux/of_platform.h>
+#include <linux/phy/phy.h>
+
+#include "pcie-iproc.h"
+
+#define CLK_CONTROL_OFFSET           0x000
+#define EP_MODE_SURVIVE_PERST_SHIFT  1
+#define EP_MODE_SURVIVE_PERST        BIT(EP_MODE_SURVIVE_PERST_SHIFT)
+#define RC_PCIE_RST_OUTPUT_SHIFT     0
+#define RC_PCIE_RST_OUTPUT           BIT(RC_PCIE_RST_OUTPUT_SHIFT)
+
+#define CFG_IND_ADDR_OFFSET          0x120
+#define CFG_IND_ADDR_MASK            0x00001ffc
+
+#define CFG_IND_DATA_OFFSET          0x124
+
+#define CFG_ADDR_OFFSET              0x1f8
+#define CFG_ADDR_BUS_NUM_SHIFT       20
+#define CFG_ADDR_BUS_NUM_MASK        0x0ff00000
+#define CFG_ADDR_DEV_NUM_SHIFT       15
+#define CFG_ADDR_DEV_NUM_MASK        0x000f8000
+#define CFG_ADDR_FUNC_NUM_SHIFT      12
+#define CFG_ADDR_FUNC_NUM_MASK       0x00007000
+#define CFG_ADDR_REG_NUM_SHIFT       2
+#define CFG_ADDR_REG_NUM_MASK        0x00000ffc
+#define CFG_ADDR_CFG_TYPE_SHIFT      0
+#define CFG_ADDR_CFG_TYPE_MASK       0x00000003
+
+#define CFG_DATA_OFFSET              0x1fc
+
+#define SYS_RC_INTX_EN               0x330
+#define SYS_RC_INTX_MASK             0xf
+
+static inline struct iproc_pcie *sys_to_pcie(struct pci_sys_data *sys)
+{
+	return sys->private_data;

drivers/pci/host/pcie-iproc.c line 73

Quoted 10 lines
drivers/pci/host/pcie-iproc.c new file +10 -0
+}
+
+#define INVALID_ACCESS_OFFSET 0xffffffff
+static u32 iproc_pcie_cfg_base(struct iproc_pcie *pcie, int busno,
+			       unsigned int devfn, int where)
+{
+	int slot = PCI_SLOT(devfn);
+	int fn = PCI_FUNC(devfn);
+	u32 val;
+

Would you mind adding a comment to the effect that CFG_IND_ADDR_OFFSET/CFG_IND_DATA_OFFSET and CFG_ADDR_OFFSET/CFG_DATA_OFFSET are protected by pci_lock?

They obviously need a mutex, and while I don't have any plans to change it, I'm not completely 100% sure that pci_lock is the best place for it.

Quoted 47 lines
drivers/pci/host/pcie-iproc.c new file +47 -0
+	/* root complex access */
+	if (busno == 0) {
+		if (slot >= 1)
+			return INVALID_ACCESS_OFFSET;
+		writel(where & CFG_IND_ADDR_MASK,
+		       pcie->base + CFG_IND_ADDR_OFFSET);
+		return CFG_IND_DATA_OFFSET;
+	}
+
+	if (fn > 1)
+		return INVALID_ACCESS_OFFSET;
+
+	/* EP device access */
+	val = (busno << CFG_ADDR_BUS_NUM_SHIFT) |
+		(slot << CFG_ADDR_DEV_NUM_SHIFT) |
+		(fn << CFG_ADDR_FUNC_NUM_SHIFT) |
+		(where & CFG_ADDR_REG_NUM_MASK) |
+		(1 & CFG_ADDR_CFG_TYPE_MASK);
+	writel(val, pcie->base + CFG_ADDR_OFFSET);
+
+	return CFG_DATA_OFFSET;
+}
+
+#define INVALID_CFG_RD 0xffffffff
+static int iproc_pcie_read_conf(struct iproc_pcie *pcie, int busno,
+				unsigned int devfn, int where, int size,
+				u32 *val)
+{
+	u32 offset, mask, shift;
+
+	*val = INVALID_CFG_RD;
+
+	if (size != 1 && size != 2 && size != 4)
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+	else if ((size == 2) && (where & 1))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+	else if ((size == 4) && (where & 3))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	offset = iproc_pcie_cfg_base(pcie, busno, devfn, where);
+	if (offset == INVALID_ACCESS_OFFSET)
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	*val = readl(pcie->base + offset);
+
+	if (size != 4) {
+		mask = (1 << (size * BITS_PER_BYTE)) - 1;

drivers/pci/host/pcie-iproc.c line 130

Quoted 10 lines
drivers/pci/host/pcie-iproc.c new file +10 -0
+		shift = (where % 4) * BITS_PER_BYTE;
+		*val = (*val >> shift) & mask;
+	}
+
+	dev_dbg(pcie->dev,
+		"conf rd: busn=%d devfn=%u where=%d size=%d val=0x%08x\n",
+		busno, devfn, where, size, *val);
+
+	return PCIBIOS_SUCCESSFUL;
+}

This looks like maybe it could use Rob Herring's (CC'd) new generic config accessors? For example, see these:

1f94a94f67e1 ("PCI: Add generic config accessors") 21186728865e ("PCI: generic: Convert to use generic config accessors")

Quoted 176 lines
drivers/pci/host/pcie-iproc.c new file +176 -0
+
+static int iproc_pcie_write_conf(struct iproc_pcie *pcie, int busno,
+				 unsigned int devfn, int where, int size,
+				 u32 val)
+{
+	u32 offset, mask, shift, data;
+
+	if (size != 1 && size != 2 && size != 4)
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+	else if ((size == 2) && (where & 1))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+	else if ((size == 4) && (where & 3))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	offset = iproc_pcie_cfg_base(pcie, busno, devfn, where);
+	if (offset == INVALID_ACCESS_OFFSET)
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	data = readl(pcie->base + offset);
+
+	if (size != 4) {
+		mask = (1 << (size * BITS_PER_BYTE)) - 1;
+		shift = (where % 4) * BITS_PER_BYTE;
+		data &= ~(mask << shift);
+		data |= (val & mask) << shift;
+	} else {
+		data = val;
+	}
+
+	writel(data, pcie->base + offset);
+
+	dev_dbg(pcie->dev,
+		"conf wr: busn=%d devfn=%u where=%d size=%d data=0x%08x\n",
+		busno, devfn, where, size, data);
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int iproc_pcie_read(struct pci_bus *bus, unsigned int devfn,
+			   int where, int size, u32 *val)
+{
+	struct iproc_pcie *pcie = sys_to_pcie(bus->sysdata);
+
+	return iproc_pcie_read_conf(pcie, bus->number, devfn, where, size,
+				    val);
+}
+
+static int iproc_pcie_write(struct pci_bus *bus, unsigned int devfn,
+			    int where, int size, u32 val)
+{
+	struct iproc_pcie *pcie = sys_to_pcie(bus->sysdata);
+
+	return iproc_pcie_write_conf(pcie, bus->number, devfn, where, size,
+				     val);
+}
+
+static struct pci_ops iproc_pcie_ops = {
+	.read = iproc_pcie_read,
+	.write = iproc_pcie_write,
+};
+
+static void iproc_pcie_reset(struct iproc_pcie *pcie)
+{
+	u32 val;
+
+	/*
+	 * Configure the PCIe controller as root complex and send a downstream
+	 * reset
+	 */
+	val = EP_MODE_SURVIVE_PERST | RC_PCIE_RST_OUTPUT;
+	writel(val, pcie->base + CLK_CONTROL_OFFSET);
+	udelay(250);
+	val &= ~EP_MODE_SURVIVE_PERST;
+	writel(val, pcie->base + CLK_CONTROL_OFFSET);
+	msleep(250);
+}
+
+static int iproc_pcie_check_link(struct iproc_pcie *pcie, struct pci_bus *bus)
+{
+	u8 hdr_type;
+	u32 link_ctrl;
+	u16 pos, link_status;
+	int link_is_active = 0;
+
+	/* make sure we are not in EP mode */
+	pci_bus_read_config_byte(bus, 0, PCI_HEADER_TYPE, &hdr_type);
+	if ((hdr_type & 0x7f) != PCI_HEADER_TYPE_BRIDGE) {
+		dev_err(pcie->dev, "in EP mode, hdr=0x08%x\n", hdr_type);
+		return -EFAULT;
+	}
+
+	/* force class to PCI_CLASS_BRIDGE_PCI (0x0604) */
+	pci_bus_write_config_word(bus, 0, PCI_CLASS_DEVICE,
+				  PCI_CLASS_BRIDGE_PCI);
+
+	/* check link status to see if link is active */
+	pos = pci_bus_find_capability(bus, 0, PCI_CAP_ID_EXP);
+	pci_bus_read_config_word(bus, 0, pos + PCI_EXP_LNKSTA, &link_status);
+	if (link_status & PCI_EXP_LNKSTA_NLW)
+		link_is_active = 1;
+
+	if (!link_is_active) {
+		/* try GEN 1 link speed */
+#define PCI_LINK_STATUS_CTRL_2_OFFSET 0x0dc
+#define PCI_TARGET_LINK_SPEED_MASK    0xf
+#define PCI_TARGET_LINK_SPEED_GEN2    0x2
+#define PCI_TARGET_LINK_SPEED_GEN1    0x1
+		pci_bus_read_config_dword(bus, 0,
+					  PCI_LINK_STATUS_CTRL_2_OFFSET,
+					  &link_ctrl);
+		if ((link_ctrl & PCI_TARGET_LINK_SPEED_MASK) ==
+		    PCI_TARGET_LINK_SPEED_GEN2) {
+			link_ctrl &= ~PCI_TARGET_LINK_SPEED_MASK;
+			link_ctrl |= PCI_TARGET_LINK_SPEED_GEN1;
+			pci_bus_write_config_dword(bus, 0,
+					   PCI_LINK_STATUS_CTRL_2_OFFSET,
+					   link_ctrl);
+			msleep(100);
+
+			pos = pci_bus_find_capability(bus, 0, PCI_CAP_ID_EXP);
+			pci_bus_read_config_word(bus, 0, pos + PCI_EXP_LNKSTA,
+						 &link_status);
+			if (link_status & PCI_EXP_LNKSTA_NLW)
+				link_is_active = 1;
+		}
+	}
+
+	dev_info(pcie->dev, "link: %s\n", link_is_active ? "UP" : "DOWN");
+
+	return link_is_active ? 0 : -ENODEV;
+}
+
+static void iproc_pcie_enable(struct iproc_pcie *pcie)
+{
+	writel(SYS_RC_INTX_MASK, pcie->base + SYS_RC_INTX_EN);
+}
+
+int iproc_pcie_setup(struct iproc_pcie *pcie)
+{
+	int ret;
+	struct pci_bus *bus;
+
+	if (!pcie || !pcie->dev || !pcie->base)
+		return -EINVAL;
+
+	if (pcie->phy) {
+		ret = phy_init(pcie->phy);
+		if (ret) {
+			dev_err(pcie->dev, "unable to initialize PCIe PHY\n");
+			return ret;
+		}
+
+		ret = phy_power_on(pcie->phy);
+		if (ret) {
+			dev_err(pcie->dev, "unable to power on PCIe PHY\n");
+			goto err_exit_phy;
+		}
+
+	}
+
+	iproc_pcie_reset(pcie);
+
+	pcie->sysdata.private_data = pcie;
+
+	bus = pci_create_root_bus(pcie->dev, 0, &iproc_pcie_ops,
+				  &pcie->sysdata, pcie->resources);
+	if (!bus) {
+		dev_err(pcie->dev, "unable to create PCI root bus\n");
+		ret = -ENOMEM;
+		goto err_power_off_phy;
+	}
+	pcie->root_bus = bus;
+
+	ret = iproc_pcie_check_link(pcie, bus);
+	if (ret) {
+		dev_err(pcie->dev, "no PCIe EP device detected\n");

drivers/pci/host/pcie-iproc.c line 316

Quoted 10 lines
drivers/pci/host/pcie-iproc.c new file +10 -0
+		goto err_rm_root_bus;
+	}
+
+	iproc_pcie_enable(pcie);
+
+	pci_scan_child_bus(bus);
+	pci_assign_unassigned_bus_resources(bus);
+	pci_bus_add_devices(bus);
+
+	pci_fixup_irqs(pci_common_swizzle, of_irq_parse_and_map_pci);

I think the IRQ fixup should be before pci_bus_add_devices(). CC'd Yijing, who's been fixing similar issues.

pci_bus_add_devices() is the point where drivers can claim these devices, and we don't want to change things after a driver claims the device.

Quoted 21 lines
drivers/pci/host/pcie-iproc.c new file +21 -0
+
+	return 0;
+
+err_rm_root_bus:
+	pci_stop_root_bus(bus);
+	pci_remove_root_bus(bus);
+
+err_power_off_phy:
+	if (pcie->phy)
+		phy_power_off(pcie->phy);
+err_exit_phy:
+	if (pcie->phy)
+		phy_exit(pcie->phy);
+
+	return ret;
+}
+EXPORT_SYMBOL(iproc_pcie_setup);
+
+int iproc_pcie_remove(struct iproc_pcie *pcie)
+{
+	pci_stop_root_bus(pcie->root_bus);

drivers/pci/host/pcie-iproc.c line 347

Quoted 10 lines
drivers/pci/host/pcie-iproc.c new file +10 -0
+	pci_remove_root_bus(pcie->root_bus);
+
+	if (pcie->phy) {
+		phy_power_off(pcie->phy);
+		phy_exit(pcie->phy);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(iproc_pcie_remove);

These exports wouldn't be needed if this were all squashed into one file.

Quoted 40 lines
drivers/pci/host/pcie-iproc.c new file +4 -0
+
+MODULE_AUTHOR("Ray Jui <[email protected]>");
+MODULE_DESCRIPTION("Broadcom iPROC PCIe common driver");
+MODULE_LICENSE("GPL v2");
drivers/pci/host/pcie-iproc.h new file +30 -0
new file mode 100644
index 0000000..569bc04
@@ -0,0 +1,42 @@
+/*
+ * Copyright (C) 2014-2015 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef _PCIE_IPROC_H
+#define _PCIE_IPROC_H
+
+#define IPROC_PCIE_MAX_NUM_IRQS 6
+
+/**
+ * iProc PCIe device
+ * @dev: pointer to device data structure
+ * @base: PCIe host controller I/O register base
+ * @resources: linked list of all PCI resources
+ * @sysdata: Per PCI controller data
+ * @root_bus: pointer to root bus
+ * @phy: optional PHY device that controls the Serdes
+ * @irqs: interrupt IDs
+ */
+struct iproc_pcie {
+	struct device *dev;

drivers/pci/host/pcie-iproc.h line 40

Quoted 10 lines
drivers/pci/host/pcie-iproc.h new file +10 -0
+	void __iomem *base;
+	struct list_head *resources;
+	struct pci_sys_data sysdata;
+	struct pci_bus *root_bus;
+	struct phy *phy;
+	int irqs[IPROC_PCIE_MAX_NUM_IRQS];
+};
+
+extern int iproc_pcie_setup(struct iproc_pcie *pcie);
+extern int iproc_pcie_remove(struct iproc_pcie *pcie);

Hopefully you can squash this all into one file, but if you can't, my preference is to omit "extern" on function declarations.

Quoted 5 lines
drivers/pci/host/pcie-iproc.h new file +2 -0
+
+#endif /* _PCIE_IPROC_H */
Signature · 1 line
1.7.9.5
author maintainer
#7 lore raw

Hi Bjorn,

Quoted 48 lines · Bjorn Helgaas

[+cc Rob, Yijing]

Quoted 34 lines · Ray Jui

This adds the support for Broadcom iProc PCIe controller

pcie-iproc.c servers as the common core driver, and front-end bus interface needs to be added to support different bus interfaces

pcie-iproc-pltfm.c contains the support for the platform bus interface

---
 drivers/pci/host/Kconfig            |   17 ++
 drivers/pci/host/Makefile           |    2 +
 drivers/pci/host/pcie-iproc-pltfm.c |  108 +++++++++++
 drivers/pci/host/pcie-iproc.c       |  351 +++++++++++++++++++++++++++++++++++
 drivers/pci/host/pcie-iproc.h       |   42 +++++
 5 files changed, 520 insertions(+)
 create mode 100644 drivers/pci/host/pcie-iproc-pltfm.c
 create mode 100644 drivers/pci/host/pcie-iproc.c
 create mode 100644 drivers/pci/host/pcie-iproc.h
drivers/pci/host/Kconfig +6 -0
index 7b892a9..f4d9c90 100644
@@ -106,4 +106,21 @@ config PCI_VERSATILE
 	bool "ARM Versatile PB PCI controller"
 	depends on ARCH_VERSATILE
 
+config PCIE_IPROC
+	tristate "Broadcom iProc PCIe controller"
+	help
+	  This enables the iProc PCIe core controller support for Broadcom's
+	  iProc family of SoCs. An appropriate bus interface driver also needs
+	  to be enabled

drivers/pci/host/Kconfig config PCI_VERSATILE · line 124

Quoted 10 lines
drivers/pci/host/Kconfig +10 -0
+
+config PCIE_IPROC_PLTFM
+	tristate "Broadcom iProc PCIe platform bus driver"
+	depends on ARCH_BCM_IPROC || COMPILE_TEST
+	depends on OF
+	select PCIE_IPROC
+	default ARCH_BCM_IPROC
+	help
+	  Say Y here if you want to use the Broadcom iProc PCIe controller
+	  through the generic platform bus interface
Quoted 8 lines

Do you anticipate additional front-end bus interfaces? If not, and maybe even if you do, you might squash everything into pcie-iproc.c. Then you only need one file (no .h file needed) and the package is a little simpler. I think it's pretty common to have multiple driver registration methods in the same file (OF, PCI, ACPI, etc.) And I think it's common to have those methods guarded by the generic config symbol (CONFIG_PCI, CONFIG_OF, etc.) rather than defining new ones specific to the driver.

Yes I do expect Hauke (CCed) to add BCMA bus front end support later.

I still think having the front end bus driver separated is cleaner and may be less troublesome for Hauke to add BCMA support in the future. But if you strongly favor having everything stuffed in one single file, I can make that change. Please let me know.

Quoted 210 lines
Quoted 195 lines
drivers/pci/host/Kconfig +1 -0
+
 endmenu
drivers/pci/host/Makefile +2 -0
index e61d91c..2e02d20 100644
@@ -13,3 +13,5 @@ obj-$(CONFIG_PCIE_XILINX) += pcie-xilinx.o
 obj-$(CONFIG_PCI_XGENE) += pci-xgene.o
 obj-$(CONFIG_PCI_LAYERSCAPE) += pci-layerscape.o
 obj-$(CONFIG_PCI_VERSATILE) += pci-versatile.o
+obj-$(CONFIG_PCIE_IPROC) += pcie-iproc.o
+obj-$(CONFIG_PCIE_IPROC_PLTFM) += pcie-iproc-pltfm.o
drivers/pci/host/pcie-iproc-pltfm.c new file +108 -0
new file mode 100644
index 0000000..af935ea
@@ -0,0 +1,108 @@
+/*
+ * Copyright (C) 2015 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/clk.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/of_address.h>
+#include <linux/of_pci.h>
+#include <linux/of_irq.h>
+#include <linux/of_platform.h>
+#include <linux/phy/phy.h>
+
+#include "pcie-iproc.h"
+
+static int __init iproc_pcie_pltfm_probe(struct platform_device *pdev)
+{
+	struct iproc_pcie *pcie;
+	struct device_node *np = pdev->dev.of_node;
+	struct resource reg;
+	resource_size_t iobase = 0;
+	LIST_HEAD(res);
+	int ret;
+
+	pcie = devm_kzalloc(&pdev->dev, sizeof(struct iproc_pcie), GFP_KERNEL);
+	if (!pcie)
+		return -ENOMEM;
+
+	pcie->dev = &pdev->dev;
+	platform_set_drvdata(pdev, pcie);
+
+	ret = of_address_to_resource(np, 0, &reg);
+	if (ret < 0) {
+		dev_err(pcie->dev, "unable to obtain controller resources\n");
+		return ret;
+	}
+
+	pcie->base = devm_ioremap(pcie->dev, reg.start, resource_size(&reg));
+	if (!pcie->base) {
+		dev_err(pcie->dev, "unable to map controller registers\n");
+		return -ENOMEM;
+	}
+
+	/* PHY use is optional */
+	pcie->phy = devm_phy_get(&pdev->dev, "pcie-phy");
+	if (IS_ERR(pcie->phy)) {
+		if (PTR_ERR(pcie->phy) == -EPROBE_DEFER)
+			return -EPROBE_DEFER;
+		pcie->phy = NULL;
+	}
+
+	ret = of_pci_get_host_bridge_resources(np, 0, 0xff, &res, &iobase);
+	if (ret) {
+		dev_err(pcie->dev,
+			"unable to get PCI host bridge resources\n");
+		return ret;
+	}
+
+	pcie->resources = &res;
+
+	ret = iproc_pcie_setup(pcie);
+	if (ret) {
+		dev_err(pcie->dev, "PCIe controller setup failed\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int iproc_pcie_pltfm_remove(struct platform_device *pdev)
+{
+	struct iproc_pcie *pcie = platform_get_drvdata(pdev);
+
+	return iproc_pcie_remove(pcie);
+}
+
+static const struct of_device_id iproc_pcie_of_match_table[] = {
+	{ .compatible = "brcm,iproc-pcie", },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, iproc_pcie_of_match_table);
+
+static struct platform_driver iproc_pcie_pltfm_driver = {
+	.driver = {
+		.name = "iproc-pcie",
+		.of_match_table = of_match_ptr(iproc_pcie_of_match_table),
+	},
+	.probe = iproc_pcie_pltfm_probe,
+	.remove = iproc_pcie_pltfm_remove,
+};
+module_platform_driver(iproc_pcie_pltfm_driver);
+
+MODULE_AUTHOR("Ray Jui <[email protected]>");
+MODULE_DESCRIPTION("Broadcom iPROC PCIe platform driver");
+MODULE_LICENSE("GPL v2");
drivers/pci/host/pcie-iproc.c new file +63 -0
new file mode 100644
index 0000000..ad6c89e
@@ -0,0 +1,351 @@
+/*
+ * Copyright (C) 2014 Hauke Mehrtens <[email protected]>
+ * Copyright (C) 2015 Broadcom Corporatcommon ion
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/msi.h>
+#include <linux/clk.h>
+#include <linux/module.h>
+#include <linux/mbus.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/of_address.h>
+#include <linux/of_pci.h>
+#include <linux/of_irq.h>
+#include <linux/of_platform.h>
+#include <linux/phy/phy.h>
+
+#include "pcie-iproc.h"
+
+#define CLK_CONTROL_OFFSET           0x000
+#define EP_MODE_SURVIVE_PERST_SHIFT  1
+#define EP_MODE_SURVIVE_PERST        BIT(EP_MODE_SURVIVE_PERST_SHIFT)
+#define RC_PCIE_RST_OUTPUT_SHIFT     0
+#define RC_PCIE_RST_OUTPUT           BIT(RC_PCIE_RST_OUTPUT_SHIFT)
+
+#define CFG_IND_ADDR_OFFSET          0x120
+#define CFG_IND_ADDR_MASK            0x00001ffc
+
+#define CFG_IND_DATA_OFFSET          0x124
+
+#define CFG_ADDR_OFFSET              0x1f8
+#define CFG_ADDR_BUS_NUM_SHIFT       20
+#define CFG_ADDR_BUS_NUM_MASK        0x0ff00000
+#define CFG_ADDR_DEV_NUM_SHIFT       15
+#define CFG_ADDR_DEV_NUM_MASK        0x000f8000
+#define CFG_ADDR_FUNC_NUM_SHIFT      12
+#define CFG_ADDR_FUNC_NUM_MASK       0x00007000
+#define CFG_ADDR_REG_NUM_SHIFT       2
+#define CFG_ADDR_REG_NUM_MASK        0x00000ffc
+#define CFG_ADDR_CFG_TYPE_SHIFT      0
+#define CFG_ADDR_CFG_TYPE_MASK       0x00000003
+
+#define CFG_DATA_OFFSET              0x1fc
+
+#define SYS_RC_INTX_EN               0x330
+#define SYS_RC_INTX_MASK             0xf
+
+static inline struct iproc_pcie *sys_to_pcie(struct pci_sys_data *sys)
+{
+	return sys->private_data;

drivers/pci/host/pcie-iproc.c line 73

Quoted 10 lines
drivers/pci/host/pcie-iproc.c new file +10 -0
+}
+
+#define INVALID_ACCESS_OFFSET 0xffffffff
+static u32 iproc_pcie_cfg_base(struct iproc_pcie *pcie, int busno,
+			       unsigned int devfn, int where)
+{
+	int slot = PCI_SLOT(devfn);
+	int fn = PCI_FUNC(devfn);
+	u32 val;
+

Would you mind adding a comment to the effect that CFG_IND_ADDR_OFFSET/CFG_IND_DATA_OFFSET and CFG_ADDR_OFFSET/CFG_DATA_OFFSET are protected by pci_lock?

Quoted 4 lines

They obviously need a mutex, and while I don't have any plans to change it, I'm not completely 100% sure that pci_lock is the best place for it.

Sorry I don't get what you want me to do here. Do you want me to add some comment to explain that the struct pci_ops read/write callbacks are already protected at the upper layer by the pci_lock spinlock and therefore no lock is required in this driver?

Quoted 61 lines
Quoted 47 lines
drivers/pci/host/pcie-iproc.c new file +47 -0
+	/* root complex access */
+	if (busno == 0) {
+		if (slot >= 1)
+			return INVALID_ACCESS_OFFSET;
+		writel(where & CFG_IND_ADDR_MASK,
+		       pcie->base + CFG_IND_ADDR_OFFSET);
+		return CFG_IND_DATA_OFFSET;
+	}
+
+	if (fn > 1)
+		return INVALID_ACCESS_OFFSET;
+
+	/* EP device access */
+	val = (busno << CFG_ADDR_BUS_NUM_SHIFT) |
+		(slot << CFG_ADDR_DEV_NUM_SHIFT) |
+		(fn << CFG_ADDR_FUNC_NUM_SHIFT) |
+		(where & CFG_ADDR_REG_NUM_MASK) |
+		(1 & CFG_ADDR_CFG_TYPE_MASK);
+	writel(val, pcie->base + CFG_ADDR_OFFSET);
+
+	return CFG_DATA_OFFSET;
+}
+
+#define INVALID_CFG_RD 0xffffffff
+static int iproc_pcie_read_conf(struct iproc_pcie *pcie, int busno,
+				unsigned int devfn, int where, int size,
+				u32 *val)
+{
+	u32 offset, mask, shift;
+
+	*val = INVALID_CFG_RD;
+
+	if (size != 1 && size != 2 && size != 4)
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+	else if ((size == 2) && (where & 1))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+	else if ((size == 4) && (where & 3))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	offset = iproc_pcie_cfg_base(pcie, busno, devfn, where);
+	if (offset == INVALID_ACCESS_OFFSET)
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	*val = readl(pcie->base + offset);
+
+	if (size != 4) {
+		mask = (1 << (size * BITS_PER_BYTE)) - 1;

drivers/pci/host/pcie-iproc.c line 130

Quoted 10 lines
drivers/pci/host/pcie-iproc.c new file +10 -0
+		shift = (where % 4) * BITS_PER_BYTE;
+		*val = (*val >> shift) & mask;
+	}
+
+	dev_dbg(pcie->dev,
+		"conf rd: busn=%d devfn=%u where=%d size=%d val=0x%08x\n",
+		busno, devfn, where, size, *val);
+
+	return PCIBIOS_SUCCESSFUL;
+}

This looks like maybe it could use Rob Herring's (CC'd) new generic config accessors? For example, see these:

Quoted 3 lines

1f94a94f67e1 ("PCI: Add generic config accessors") 21186728865e ("PCI: generic: Convert to use generic config accessors")

Okay I will switch to use the generic code for config space access. I'll add map_bus callback for address calculation.

Quoted 190 lines
Quoted 176 lines
drivers/pci/host/pcie-iproc.c new file +176 -0
+
+static int iproc_pcie_write_conf(struct iproc_pcie *pcie, int busno,
+				 unsigned int devfn, int where, int size,
+				 u32 val)
+{
+	u32 offset, mask, shift, data;
+
+	if (size != 1 && size != 2 && size != 4)
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+	else if ((size == 2) && (where & 1))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+	else if ((size == 4) && (where & 3))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	offset = iproc_pcie_cfg_base(pcie, busno, devfn, where);
+	if (offset == INVALID_ACCESS_OFFSET)
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	data = readl(pcie->base + offset);
+
+	if (size != 4) {
+		mask = (1 << (size * BITS_PER_BYTE)) - 1;
+		shift = (where % 4) * BITS_PER_BYTE;
+		data &= ~(mask << shift);
+		data |= (val & mask) << shift;
+	} else {
+		data = val;
+	}
+
+	writel(data, pcie->base + offset);
+
+	dev_dbg(pcie->dev,
+		"conf wr: busn=%d devfn=%u where=%d size=%d data=0x%08x\n",
+		busno, devfn, where, size, data);
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int iproc_pcie_read(struct pci_bus *bus, unsigned int devfn,
+			   int where, int size, u32 *val)
+{
+	struct iproc_pcie *pcie = sys_to_pcie(bus->sysdata);
+
+	return iproc_pcie_read_conf(pcie, bus->number, devfn, where, size,
+				    val);
+}
+
+static int iproc_pcie_write(struct pci_bus *bus, unsigned int devfn,
+			    int where, int size, u32 val)
+{
+	struct iproc_pcie *pcie = sys_to_pcie(bus->sysdata);
+
+	return iproc_pcie_write_conf(pcie, bus->number, devfn, where, size,
+				     val);
+}
+
+static struct pci_ops iproc_pcie_ops = {
+	.read = iproc_pcie_read,
+	.write = iproc_pcie_write,
+};
+
+static void iproc_pcie_reset(struct iproc_pcie *pcie)
+{
+	u32 val;
+
+	/*
+	 * Configure the PCIe controller as root complex and send a downstream
+	 * reset
+	 */
+	val = EP_MODE_SURVIVE_PERST | RC_PCIE_RST_OUTPUT;
+	writel(val, pcie->base + CLK_CONTROL_OFFSET);
+	udelay(250);
+	val &= ~EP_MODE_SURVIVE_PERST;
+	writel(val, pcie->base + CLK_CONTROL_OFFSET);
+	msleep(250);
+}
+
+static int iproc_pcie_check_link(struct iproc_pcie *pcie, struct pci_bus *bus)
+{
+	u8 hdr_type;
+	u32 link_ctrl;
+	u16 pos, link_status;
+	int link_is_active = 0;
+
+	/* make sure we are not in EP mode */
+	pci_bus_read_config_byte(bus, 0, PCI_HEADER_TYPE, &hdr_type);
+	if ((hdr_type & 0x7f) != PCI_HEADER_TYPE_BRIDGE) {
+		dev_err(pcie->dev, "in EP mode, hdr=0x08%x\n", hdr_type);
+		return -EFAULT;
+	}
+
+	/* force class to PCI_CLASS_BRIDGE_PCI (0x0604) */
+	pci_bus_write_config_word(bus, 0, PCI_CLASS_DEVICE,
+				  PCI_CLASS_BRIDGE_PCI);
+
+	/* check link status to see if link is active */
+	pos = pci_bus_find_capability(bus, 0, PCI_CAP_ID_EXP);
+	pci_bus_read_config_word(bus, 0, pos + PCI_EXP_LNKSTA, &link_status);
+	if (link_status & PCI_EXP_LNKSTA_NLW)
+		link_is_active = 1;
+
+	if (!link_is_active) {
+		/* try GEN 1 link speed */
+#define PCI_LINK_STATUS_CTRL_2_OFFSET 0x0dc
+#define PCI_TARGET_LINK_SPEED_MASK    0xf
+#define PCI_TARGET_LINK_SPEED_GEN2    0x2
+#define PCI_TARGET_LINK_SPEED_GEN1    0x1
+		pci_bus_read_config_dword(bus, 0,
+					  PCI_LINK_STATUS_CTRL_2_OFFSET,
+					  &link_ctrl);
+		if ((link_ctrl & PCI_TARGET_LINK_SPEED_MASK) ==
+		    PCI_TARGET_LINK_SPEED_GEN2) {
+			link_ctrl &= ~PCI_TARGET_LINK_SPEED_MASK;
+			link_ctrl |= PCI_TARGET_LINK_SPEED_GEN1;
+			pci_bus_write_config_dword(bus, 0,
+					   PCI_LINK_STATUS_CTRL_2_OFFSET,
+					   link_ctrl);
+			msleep(100);
+
+			pos = pci_bus_find_capability(bus, 0, PCI_CAP_ID_EXP);
+			pci_bus_read_config_word(bus, 0, pos + PCI_EXP_LNKSTA,
+						 &link_status);
+			if (link_status & PCI_EXP_LNKSTA_NLW)
+				link_is_active = 1;
+		}
+	}
+
+	dev_info(pcie->dev, "link: %s\n", link_is_active ? "UP" : "DOWN");
+
+	return link_is_active ? 0 : -ENODEV;
+}
+
+static void iproc_pcie_enable(struct iproc_pcie *pcie)
+{
+	writel(SYS_RC_INTX_MASK, pcie->base + SYS_RC_INTX_EN);
+}
+
+int iproc_pcie_setup(struct iproc_pcie *pcie)
+{
+	int ret;
+	struct pci_bus *bus;
+
+	if (!pcie || !pcie->dev || !pcie->base)
+		return -EINVAL;
+
+	if (pcie->phy) {
+		ret = phy_init(pcie->phy);
+		if (ret) {
+			dev_err(pcie->dev, "unable to initialize PCIe PHY\n");
+			return ret;
+		}
+
+		ret = phy_power_on(pcie->phy);
+		if (ret) {
+			dev_err(pcie->dev, "unable to power on PCIe PHY\n");
+			goto err_exit_phy;
+		}
+
+	}
+
+	iproc_pcie_reset(pcie);
+
+	pcie->sysdata.private_data = pcie;
+
+	bus = pci_create_root_bus(pcie->dev, 0, &iproc_pcie_ops,
+				  &pcie->sysdata, pcie->resources);
+	if (!bus) {
+		dev_err(pcie->dev, "unable to create PCI root bus\n");
+		ret = -ENOMEM;
+		goto err_power_off_phy;
+	}
+	pcie->root_bus = bus;
+
+	ret = iproc_pcie_check_link(pcie, bus);
+	if (ret) {
+		dev_err(pcie->dev, "no PCIe EP device detected\n");

drivers/pci/host/pcie-iproc.c line 316

Quoted 10 lines
drivers/pci/host/pcie-iproc.c new file +10 -0
+		goto err_rm_root_bus;
+	}
+
+	iproc_pcie_enable(pcie);
+
+	pci_scan_child_bus(bus);
+	pci_assign_unassigned_bus_resources(bus);
+	pci_bus_add_devices(bus);
+
+	pci_fixup_irqs(pci_common_swizzle, of_irq_parse_and_map_pci);

I think the IRQ fixup should be before pci_bus_add_devices(). CC'd Yijing, who's been fixing similar issues.

Quoted 3 lines

pci_bus_add_devices() is the point where drivers can claim these devices, and we don't want to change things after a driver claims the device.

Right. I will fix this.

Quoted 25 lines
Quoted 25 lines
drivers/pci/host/pcie-iproc.c new file +25 -0
+
+	return 0;
+
+err_rm_root_bus:
+	pci_stop_root_bus(bus);
+	pci_remove_root_bus(bus);
+
+err_power_off_phy:
+	if (pcie->phy)
+		phy_power_off(pcie->phy);
+err_exit_phy:
+	if (pcie->phy)
+		phy_exit(pcie->phy);
+
+	return ret;
+}
+EXPORT_SYMBOL(iproc_pcie_setup);
+
+int iproc_pcie_remove(struct iproc_pcie *pcie)
+{
+	pci_stop_root_bus(pcie->root_bus);
+	pci_remove_root_bus(pcie->root_bus);
+
+	if (pcie->phy) {
+		phy_power_off(pcie->phy);
Quoted 9 lines

drivers/pci/host/pcie-iproc.c line 347

Quoted 6 lines
drivers/pci/host/pcie-iproc.c new file +6 -0
+		phy_exit(pcie->phy);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(iproc_pcie_remove);

These exports wouldn't be needed if this were all squashed into one file.

Understood. Let's decide whether or not to keep everything in one file.

Quoted 54 lines
Quoted 40 lines
drivers/pci/host/pcie-iproc.c new file +4 -0
+
+MODULE_AUTHOR("Ray Jui <[email protected]>");
+MODULE_DESCRIPTION("Broadcom iPROC PCIe common driver");
+MODULE_LICENSE("GPL v2");
drivers/pci/host/pcie-iproc.h new file +30 -0
new file mode 100644
index 0000000..569bc04
@@ -0,0 +1,42 @@
+/*
+ * Copyright (C) 2014-2015 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef _PCIE_IPROC_H
+#define _PCIE_IPROC_H
+
+#define IPROC_PCIE_MAX_NUM_IRQS 6
+
+/**
+ * iProc PCIe device
+ * @dev: pointer to device data structure
+ * @base: PCIe host controller I/O register base
+ * @resources: linked list of all PCI resources
+ * @sysdata: Per PCI controller data
+ * @root_bus: pointer to root bus
+ * @phy: optional PHY device that controls the Serdes
+ * @irqs: interrupt IDs
+ */
+struct iproc_pcie {
+	struct device *dev;

drivers/pci/host/pcie-iproc.h line 40

Quoted 10 lines
drivers/pci/host/pcie-iproc.h new file +10 -0
+	void __iomem *base;
+	struct list_head *resources;
+	struct pci_sys_data sysdata;
+	struct pci_bus *root_bus;
+	struct phy *phy;
+	int irqs[IPROC_PCIE_MAX_NUM_IRQS];
+};
+
+extern int iproc_pcie_setup(struct iproc_pcie *pcie);
+extern int iproc_pcie_remove(struct iproc_pcie *pcie);

Hopefully you can squash this all into one file, but if you can't, my preference is to omit "extern" on function declarations.

Quoted 5 lines
Quoted 5 lines
drivers/pci/host/pcie-iproc.h new file +2 -0
+
+#endif /* _PCIE_IPROC_H */
Signature · 1 line
1.7.9.5

Thanks for the review!

Ray

#8 lore raw
Quoted 9 lines · On Tuesday 10 March 2015 15:22:36 Ray Jui wrote:
Quoted 8 lines

Do you anticipate additional front-end bus interfaces? If not, and maybe even if you do, you might squash everything into pcie-iproc.c. Then you only need one file (no .h file needed) and the package is a little simpler. I think it's pretty common to have multiple driver registration methods in the same file (OF, PCI, ACPI, etc.) And I think it's common to have those methods guarded by the generic config symbol (CONFIG_PCI, CONFIG_OF, etc.) rather than defining new ones specific to the driver.

Quoted 7 lines

Yes I do expect Hauke (CCed) to add BCMA bus front end support later.

I still think having the front end bus driver separated is cleaner and may be less troublesome for Hauke to add BCMA support in the future. But if you strongly favor having everything stuffed in one single file, I can make that change. Please let me know.

I was the one that initially suggested splitting the driver into files like this, and I still think it's the right strategy with the BCMA driver coming up.

Arnd

maintainer
#9 lore raw
In reply to Ray Jui
Quoted 59 lines · Ray Jui

Hi Bjorn,

Quoted 48 lines · Bjorn Helgaas

[+cc Rob, Yijing]

Quoted 34 lines · Ray Jui

This adds the support for Broadcom iProc PCIe controller

pcie-iproc.c servers as the common core driver, and front-end bus interface needs to be added to support different bus interfaces

pcie-iproc-pltfm.c contains the support for the platform bus interface

---
 drivers/pci/host/Kconfig            |   17 ++
 drivers/pci/host/Makefile           |    2 +
 drivers/pci/host/pcie-iproc-pltfm.c |  108 +++++++++++
 drivers/pci/host/pcie-iproc.c       |  351 +++++++++++++++++++++++++++++++++++
 drivers/pci/host/pcie-iproc.h       |   42 +++++
 5 files changed, 520 insertions(+)
 create mode 100644 drivers/pci/host/pcie-iproc-pltfm.c
 create mode 100644 drivers/pci/host/pcie-iproc.c
 create mode 100644 drivers/pci/host/pcie-iproc.h
drivers/pci/host/Kconfig +6 -0
index 7b892a9..f4d9c90 100644
@@ -106,4 +106,21 @@ config PCI_VERSATILE
 	bool "ARM Versatile PB PCI controller"
 	depends on ARCH_VERSATILE
 
+config PCIE_IPROC
+	tristate "Broadcom iProc PCIe controller"
+	help
+	  This enables the iProc PCIe core controller support for Broadcom's
+	  iProc family of SoCs. An appropriate bus interface driver also needs
+	  to be enabled

drivers/pci/host/Kconfig config PCI_VERSATILE · line 124

Quoted 10 lines
drivers/pci/host/Kconfig +10 -0
+
+config PCIE_IPROC_PLTFM
+	tristate "Broadcom iProc PCIe platform bus driver"
+	depends on ARCH_BCM_IPROC || COMPILE_TEST
+	depends on OF
+	select PCIE_IPROC
+	default ARCH_BCM_IPROC
+	help
+	  Say Y here if you want to use the Broadcom iProc PCIe controller
+	  through the generic platform bus interface
Quoted 7 lines

Do you anticipate additional front-end bus interfaces? If not, and maybe even if you do, you might squash everything into pcie-iproc.c. Then you only need one file (no .h file needed) and the package is a little simpler. I think it's pretty common to have multiple driver registration methods in the same file (OF, PCI, ACPI, etc.) And I think it's common to have those methods guarded by the generic config symbol (CONFIG_PCI, CONFIG_OF, etc.) rather than defining new ones specific to the driver.

Quoted 6 lines

Yes I do expect Hauke (CCed) to add BCMA bus front end support later.

I still think having the front end bus driver separated is cleaner and may be less troublesome for Hauke to add BCMA support in the future. But if you strongly favor having everything stuffed in one single file, I can make that change. Please let me know.

OK, just leave it as-is.

Quoted 13 lines
Quoted 13 lines

drivers/pci/host/Kconfig config PCI_VERSATILE

Quoted 8 lines
drivers/pci/host/Kconfig +8 -0
+#define INVALID_ACCESS_OFFSET 0xffffffff
+static u32 iproc_pcie_cfg_base(struct iproc_pcie *pcie, int busno,
+			       unsigned int devfn, int where)
+{
+	int slot = PCI_SLOT(devfn);
+	int fn = PCI_FUNC(devfn);
+	u32 val;
+

Would you mind adding a comment to the effect that CFG_IND_ADDR_OFFSET/CFG_IND_DATA_OFFSET and CFG_ADDR_OFFSET/CFG_DATA_OFFSET are protected by pci_lock?

Quoted 8 lines
Quoted 3 lines

They obviously need a mutex, and while I don't have any plans to change it, I'm not completely 100% sure that pci_lock is the best place for it.

Sorry I don't get what you want me to do here. Do you want me to add some comment to explain that the struct pci_ops read/write callbacks are already protected at the upper layer by the pci_lock spinlock and therefore no lock is required in this driver?

Nothing fancy; something like this that "git grep pci_lock" will find:

/* addr/data must used atomically and are protected by pci_lock */

Bjorn

#10 lore raw
In reply to Bjorn Helgaas

...

Quoted 26 lines
Quoted 12 lines
drivers/pci/host/pcie-iproc.c new file +12 -0
+	bus = pci_create_root_bus(pcie->dev, 0, &iproc_pcie_ops,
+				  &pcie->sysdata, pcie->resources);
+	if (!bus) {
+		dev_err(pcie->dev, "unable to create PCI root bus\n");
+		ret = -ENOMEM;
+		goto err_power_off_phy;
+	}
+	pcie->root_bus = bus;
+
+	ret = iproc_pcie_check_link(pcie, bus);
+	if (ret) {
+		dev_err(pcie->dev, "no PCIe EP device detected\n");

drivers/pci/host/pcie-iproc.c line 316

Quoted 10 lines
drivers/pci/host/pcie-iproc.c new file +10 -0
+		goto err_rm_root_bus;
+	}
+
+	iproc_pcie_enable(pcie);
+
+	pci_scan_child_bus(bus);
+	pci_assign_unassigned_bus_resources(bus);
+	pci_bus_add_devices(bus);
+
+	pci_fixup_irqs(pci_common_swizzle, of_irq_parse_and_map_pci);

I think the IRQ fixup should be before pci_bus_add_devices(). CC'd Yijing, who's been fixing similar issues.

Quoted 2 lines

pci_bus_add_devices() is the point where drivers can claim these devices, and we don't want to change things after a driver claims the device.

Yes, we should call pci_bus_add_devices() at last, after all resources(mem/io/irq) are configured properly. Otherwise, this code could not be run normally in non system boot up path.

Thanks!
Yijing.

Quoted 97 lines
Quoted 22 lines
drivers/pci/host/pcie-iproc.c new file +21 -0
+
+	return 0;
+
+err_rm_root_bus:
+	pci_stop_root_bus(bus);
+	pci_remove_root_bus(bus);
+
+err_power_off_phy:
+	if (pcie->phy)
+		phy_power_off(pcie->phy);
+err_exit_phy:
+	if (pcie->phy)
+		phy_exit(pcie->phy);
+
+	return ret;
+}
+EXPORT_SYMBOL(iproc_pcie_setup);
+
+int iproc_pcie_remove(struct iproc_pcie *pcie)
+{
+	pci_stop_root_bus(pcie->root_bus);

drivers/pci/host/pcie-iproc.c line 347

Quoted 10 lines
drivers/pci/host/pcie-iproc.c new file +10 -0
+	pci_remove_root_bus(pcie->root_bus);
+
+	if (pcie->phy) {
+		phy_power_off(pcie->phy);
+		phy_exit(pcie->phy);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(iproc_pcie_remove);

These exports wouldn't be needed if this were all squashed into one file.

Quoted 40 lines
drivers/pci/host/pcie-iproc.c new file +4 -0
+
+MODULE_AUTHOR("Ray Jui <[email protected]>");
+MODULE_DESCRIPTION("Broadcom iPROC PCIe common driver");
+MODULE_LICENSE("GPL v2");
drivers/pci/host/pcie-iproc.h new file +30 -0
new file mode 100644
index 0000000..569bc04
@@ -0,0 +1,42 @@
+/*
+ * Copyright (C) 2014-2015 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef _PCIE_IPROC_H
+#define _PCIE_IPROC_H
+
+#define IPROC_PCIE_MAX_NUM_IRQS 6
+
+/**
+ * iProc PCIe device
+ * @dev: pointer to device data structure
+ * @base: PCIe host controller I/O register base
+ * @resources: linked list of all PCI resources
+ * @sysdata: Per PCI controller data
+ * @root_bus: pointer to root bus
+ * @phy: optional PHY device that controls the Serdes
+ * @irqs: interrupt IDs
+ */
+struct iproc_pcie {
+	struct device *dev;

drivers/pci/host/pcie-iproc.h line 40

Quoted 10 lines
drivers/pci/host/pcie-iproc.h new file +10 -0
+	void __iomem *base;
+	struct list_head *resources;
+	struct pci_sys_data sysdata;
+	struct pci_bus *root_bus;
+	struct phy *phy;
+	int irqs[IPROC_PCIE_MAX_NUM_IRQS];
+};
+
+extern int iproc_pcie_setup(struct iproc_pcie *pcie);
+extern int iproc_pcie_remove(struct iproc_pcie *pcie);

Hopefully you can squash this all into one file, but if you can't, my preference is to omit "extern" on function declarations.

Quoted 5 lines
drivers/pci/host/pcie-iproc.h new file +2 -0
+
+#endif /* _PCIE_IPROC_H */
Signature · 1 line
1.7.9.5

.

Signature · 2 lines
Thanks!
Yijing
author maintainer
#11 lore raw

PATCH v5 4/4 ARM: dts: enable PCIe support for Cygnus

In reply to Ray Jui

Add PCIe device nodes in bcm-cygnus.dtsi but keep them disabled there. Only enable them for bcm958300k where PCIe interfaces are populated

---
 arch/arm/boot/dts/bcm-cygnus.dtsi |   42 +++++++++++++++++++++++++++++++++++++
 arch/arm/boot/dts/bcm958300k.dts  |    8 +++++++
 2 files changed, 50 insertions(+)
arch/arm/boot/dts/bcm-cygnus.dtsi +42 -0
index ff5fb6a..ee0c6e4 100644
@@ -90,6 +90,48 @@
 		status = "disabled";
 	};
 
+	pcie0: pcie@18012000 {
+		compatible = "brcm,iproc-pcie";
+		reg = <0x18012000 0x1000>;
+
+		#interrupt-cells = <1>;
+		interrupt-map-mask = <0 0 0 0>;
+		interrupt-map = <0 0 0 0 &gic GIC_SPI 100 IRQ_TYPE_NONE>;
+
+		linux,pci-domain = <0>;
+
+		bus-range = <0x00 0xff>;
+
+		#address-cells = <3>;
+		#size-cells = <2>;
+		device_type = "pci";
+		ranges = <0x81000000 0 0	  0x28000000 0 0x00010000
+			  0x82000000 0 0x20000000 0x20000000 0 0x04000000>;
+
+		status = "disabled";
+	};
+
+	pcie1: pcie@18013000 {
+		compatible = "brcm,iproc-pcie";
+		reg = <0x18013000 0x1000>;
+
+		#interrupt-cells = <1>;
+		interrupt-map-mask = <0 0 0 0>;
+		interrupt-map = <0 0 0 0 &gic GIC_SPI 106 IRQ_TYPE_NONE>;
+
+		linux,pci-domain = <1>;
+
+		bus-range = <0x00 0xff>;
+
+		#address-cells = <3>;
+		#size-cells = <2>;
+		device_type = "pci";
+		ranges = <0x81000000 0 0	  0x48000000 0 0x00010000
+			  0x82000000 0 0x40000000 0x40000000 0 0x04000000>;
+
+		status = "disabled";
+	};
+
 	uart0: serial@18020000 {
 		compatible = "snps,dw-apb-uart";
 		reg = <0x18020000 0x100>;
arch/arm/boot/dts/bcm958300k.dts +8 -0
index f1bb36f..c9eb856 100644
@@ -47,6 +47,14 @@
 		bootargs = "console=ttyS0,115200";
 	};
 
+	pcie0: pcie@18012000 {
+		status = "okay";
+	};
+
+	pcie1: pcie@18013000 {
+		status = "okay";
+	};
+
 	uart3: serial@18023000 {
 		status = "okay";
 	};
Signature · 1 line
1.7.9.5