This patch series adds Broadcom North Star 2 (NS2) SoC support. NS2 is an ARMv8
based SoC and under the Broadcom iProc family.
Sorry for tying this with the Broadcom iProc PCIe driver fixes for ARM64. I
have to tie them together because iProc PCIe support is enabled by default
when ARCH_BCM_IPROC is enabled. Without the fixes in the iProc PCIe driver,
enabling CONFIG_ARCH_BCM_IPROC would break the build for arm64 defconfig. Let
me know if there's a better way to handle this.
This patch series is generated based on v4.2-rc2 and tested on Broadcom NS2 SVK
Code available on GITHUB: https://github.com/Broadcom/arm64-linux.git
branch is ns2-core-v4
Changes from V3:
- Add more paragraph to the iProc PCIe arm64 support patch commit message to
explain the change and anticipated future change in more details
Changes from V2:
- Drop hardcoded earlycon kernel command line paramter in NS2 SVK dts file
because 1) earlycon is a debugging feature that can be enabled in the
bootloader and should not be enabled by default in the board dts file and 2)
of_earlycon should be used and support should be added to 8250 DW driver
Changes from V1:
- Took Arnd's advice to tweak the location of struct pci_sys_data within
struct iproc_pcie. This helps to get rid of most of the CONFIG_ARM wrap in
iProc PCIe core driver
- Use stdout-path and alias for serial console in NS2 SVK dts
- Add all 4 CPU descriptions in NS2 dtsi
- Remove "clock-frequency" property in the armv8 timer node so timer frequency
can be determined based on readings from CNTFRQ_EL0
- Remove config flag ARCH_BCM_NS2. Leave only ARCH_BCM_IPROC for all Broadcom
arm64 SoCs as advised
Ray Jui (4):
PCI: iproc: enable arm64 support for iProc PCIe
PCI: iproc: Fix ARM64 dependency in Kconfig
arm64: Add Broadcom iProc family support
arm64: dts: Add Broadcom North Star 2 support
Documentation/devicetree/bindings/arm/bcm/ns2.txt | 9 ++
arch/arm64/Kconfig | 5 +
arch/arm64/boot/dts/Makefile | 1 +
arch/arm64/boot/dts/broadcom/Makefile | 5 +
arch/arm64/boot/dts/broadcom/ns2-svk.dts | 59 +++++++++++
arch/arm64/boot/dts/broadcom/ns2.dtsi | 118 +++++++++++++++++++++
arch/arm64/configs/defconfig | 2 +
drivers/pci/host/Kconfig | 2 +-
drivers/pci/host/pcie-iproc.c | 15 +--
drivers/pci/host/pcie-iproc.h | 8 +-
10 files changed, 210 insertions(+), 14 deletions(-)
create mode 100644 Documentation/devicetree/bindings/arm/bcm/ns2.txt
create mode 100644 arch/arm64/boot/dts/broadcom/Makefile
create mode 100644 arch/arm64/boot/dts/broadcom/ns2-svk.dts
create mode 100644 arch/arm64/boot/dts/broadcom/ns2.dtsi
--
1.7.9.5
This patch enables arm64 support to the iProc PCIe driver
Note struct pci_sys_data is arm32 specific and will eventually be
removed. This change is done in such a way that when struct pci_sys_data
is removed from arm32, one only needs to also remove it from
pcie-iproc.h, no other change in the iProc PCIe core driver is needed
In addition, arm64 based PCI driver does not require call to
pci_fixup_irqs, as it implements OF based irq parsing and mapping in
pcibios_add_device
Signed-off-by: Ray Jui <[email protected]>
Reviewed-by: Scott Branden <[email protected]>
---
drivers/pci/host/pcie-iproc.c | 15 ++++-----------
drivers/pci/host/pcie-iproc.h | 8 ++++++--
2 files changed, 10 insertions(+), 13 deletions(-)
diff --git a/drivers/pci/host/pcie-iproc.c b/drivers/pci/host/pcie-iproc.c
index d77481e..8a556d5 100644
--- a/drivers/pci/host/pcie-iproc.c
+++ b/drivers/pci/host/pcie-iproc.c
@@ -58,11 +58,6 @@
#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;
-}
-
/**
* Note access to the configuration registers are protected at the higher layer
* by 'pci_lock' in drivers/pci/access.c
@@ -71,8 +66,7 @@ static void __iomem *iproc_pcie_map_cfg_bus(struct pci_bus *bus,
unsigned int devfn,
int where)
{
- struct pci_sys_data *sys = bus->sysdata;
- struct iproc_pcie *pcie = sys_to_pcie(sys);
+ struct iproc_pcie *pcie = bus->sysdata;
unsigned slot = PCI_SLOT(devfn);
unsigned fn = PCI_FUNC(devfn);
unsigned busno = bus->number;
@@ -208,10 +202,7 @@ int iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *res)
iproc_pcie_reset(pcie);
- pcie->sysdata.private_data = pcie;
-
- bus = pci_create_root_bus(pcie->dev, 0, &iproc_pcie_ops,
- &pcie->sysdata, res);
+ bus = pci_create_root_bus(pcie->dev, 0, &iproc_pcie_ops, pcie, res);
if (!bus) {
dev_err(pcie->dev, "unable to create PCI root bus\n");
ret = -ENOMEM;
@@ -229,7 +220,9 @@ int iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *res)
pci_scan_child_bus(bus);
pci_assign_unassigned_bus_resources(bus);
+#ifdef CONFIG_ARM
pci_fixup_irqs(pci_common_swizzle, pcie->map_irq);
+#endif
pci_bus_add_devices(bus);
return 0;
diff --git a/drivers/pci/host/pcie-iproc.h b/drivers/pci/host/pcie-iproc.h
index ba0a108..0ee9673 100644
--- a/drivers/pci/host/pcie-iproc.h
+++ b/drivers/pci/host/pcie-iproc.h
@@ -18,18 +18,22 @@
/**
* iProc PCIe device
+ * @sysdata: Per PCI controller data. This needs to be kept at the beginning of
+ * struct iproc_pcie, to enable support of both ARM32 and ARM64 platforms with
+ * minimal changes in the iProc PCIe core driver
* @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 {
+#ifdef CONFIG_ARM
+ struct pci_sys_data sysdata;
+#endif
struct device *dev;
void __iomem *base;
- struct pci_sys_data sysdata;
struct pci_bus *root_bus;
struct phy *phy;
int irqs[IPROC_PCIE_MAX_NUM_IRQS];
--
1.7.9.5
Allow Broadcom iProc PCIe core driver to be compiled for ARM64
Signed-off-by: Ray Jui <[email protected]>
Reviewed-by: Vikram Prakash <[email protected]>
Reviewed-by: Scott Branden <[email protected]>
---
drivers/pci/host/Kconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/pci/host/Kconfig b/drivers/pci/host/Kconfig
index c132bdd..d2c6144 100644
--- a/drivers/pci/host/Kconfig
+++ b/drivers/pci/host/Kconfig
@@ -117,7 +117,7 @@ config PCI_VERSATILE
config PCIE_IPROC
tristate "Broadcom iProc PCIe controller"
- depends on OF && ARM
+ depends on OF && (ARM || ARM64)
default n
help
This enables the iProc PCIe core controller support for Broadcom's
--
1.7.9.5
This patch adds support to Broadcom's iProc family of arm64 based SoCs
in the arm64 Kconfig and defconfig files
Signed-off-by: Ray Jui <[email protected]>
Reviewed-by: Scott Branden <[email protected]>
---
arch/arm64/Kconfig | 5 +++++
arch/arm64/configs/defconfig | 2 ++
2 files changed, 7 insertions(+)
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 318175f..969ef4a 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -162,6 +162,11 @@ source "kernel/Kconfig.freezer"
menu "Platform selection"
+config ARCH_BCM_IPROC
+ bool "Broadcom iProc SoC Family"
+ help
+ This enables support for Broadcom iProc based SoCs
+
config ARCH_EXYNOS
bool
help
diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig
index 4e17e7e..c83d51f 100644
--- a/arch/arm64/configs/defconfig
+++ b/arch/arm64/configs/defconfig
@@ -31,6 +31,7 @@ CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
# CONFIG_IOSCHED_DEADLINE is not set
+CONFIG_ARCH_BCM_IPROC=y
CONFIG_ARCH_EXYNOS7=y
CONFIG_ARCH_FSL_LS2085A=y
CONFIG_ARCH_HISI=y
@@ -102,6 +103,7 @@ CONFIG_SERIO_AMBAKMI=y
CONFIG_LEGACY_PTY_COUNT=16
CONFIG_SERIAL_8250=y
CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_SERIAL_8250_DW=y
CONFIG_SERIAL_8250_MT6577=y
CONFIG_SERIAL_AMBA_PL011=y
CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
--
1.7.9.5
Add Broadcom NS2 device tree binding document. Also add initial device
tree dtsi for Broadcom North Star 2 (NS2) SoC and board support for NS2
SVK board
Signed-off-by: Jon Mason <[email protected]>
Signed-off-by: Ray Jui <[email protected]>
Reviewed-by: Scott Branden <[email protected]>
---
Documentation/devicetree/bindings/arm/bcm/ns2.txt | 9 ++
arch/arm64/boot/dts/Makefile | 1 +
arch/arm64/boot/dts/broadcom/Makefile | 5 +
arch/arm64/boot/dts/broadcom/ns2-svk.dts | 59 +++++++++++
arch/arm64/boot/dts/broadcom/ns2.dtsi | 118 +++++++++++++++++++++
5 files changed, 192 insertions(+)
create mode 100644 Documentation/devicetree/bindings/arm/bcm/ns2.txt
create mode 100644 arch/arm64/boot/dts/broadcom/Makefile
create mode 100644 arch/arm64/boot/dts/broadcom/ns2-svk.dts
create mode 100644 arch/arm64/boot/dts/broadcom/ns2.dtsi
diff --git a/Documentation/devicetree/bindings/arm/bcm/ns2.txt b/Documentation/devicetree/bindings/arm/bcm/ns2.txt
new file mode 100644
index 0000000..35f056f
--- /dev/null
+++ b/Documentation/devicetree/bindings/arm/bcm/ns2.txt
@@ -0,0 +1,9 @@
+Broadcom North Star 2 (NS2) device tree bindings
+------------------------------------------------
+
+Boards with NS2 shall have the following properties:
+
+Required root node property:
+
+NS2 SVK board
+compatible = "brcm,ns2-svk", "brcm,ns2";
diff --git a/arch/arm64/boot/dts/Makefile b/arch/arm64/boot/dts/Makefile
index 38913be..9f95941 100644
--- a/arch/arm64/boot/dts/Makefile
+++ b/arch/arm64/boot/dts/Makefile
@@ -1,6 +1,7 @@
dts-dirs += amd
dts-dirs += apm
dts-dirs += arm
+dts-dirs += broadcom
dts-dirs += cavium
dts-dirs += exynos
dts-dirs += freescale
diff --git a/arch/arm64/boot/dts/broadcom/Makefile b/arch/arm64/boot/dts/broadcom/Makefile
new file mode 100644
index 0000000..e21fe66
--- /dev/null
+++ b/arch/arm64/boot/dts/broadcom/Makefile
@@ -0,0 +1,5 @@
+dtb-$(CONFIG_ARCH_BCM_IPROC) += ns2-svk.dtb
+
+always := $(dtb-y)
+subdir-y := $(dts-dirs)
+clean-files := *.dtb
diff --git a/arch/arm64/boot/dts/broadcom/ns2-svk.dts b/arch/arm64/boot/dts/broadcom/ns2-svk.dts
new file mode 100644
index 0000000..244baf8
--- /dev/null
+++ b/arch/arm64/boot/dts/broadcom/ns2-svk.dts
@@ -0,0 +1,59 @@
+/*
+ * BSD LICENSE
+ *
+ * Copyright(c) 2015 Broadcom Corporation. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * * Neither the name of Broadcom Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+/dts-v1/;
+
+#include "ns2.dtsi"
+
+/ {
+ model = "Broadcom NS2 SVK";
+ compatible = "brcm,ns2-svk", "brcm,ns2";
+
+ aliases {
+ serial0 = &uart3;
+ };
+
+ chosen {
+ stdout-path = "serial0:115200n8";
+ };
+
+ memory {
+ device_type = "memory";
+ reg = <0x000000000 0x80000000 0x00000000 0x40000000>;
+ };
+
+ soc: soc {
+ uart3: serial@66130000 {
+ status = "ok";
+ };
+ };
+};
diff --git a/arch/arm64/boot/dts/broadcom/ns2.dtsi b/arch/arm64/boot/dts/broadcom/ns2.dtsi
new file mode 100644
index 0000000..3c92d92
--- /dev/null
+++ b/arch/arm64/boot/dts/broadcom/ns2.dtsi
@@ -0,0 +1,118 @@
+/*
+ * BSD LICENSE
+ *
+ * Copyright(c) 2015 Broadcom Corporation. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * * Neither the name of Broadcom Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <dt-bindings/interrupt-controller/arm-gic.h>
+
+/memreserve/ 0x84b00000 0x00000008;
+
+/ {
+ compatible = "brcm,ns2";
+ interrupt-parent = <&gic>;
+ #address-cells = <2>;
+ #size-cells = <2>;
+
+ cpus {
+ #address-cells = <2>;
+ #size-cells = <0>;
+
+ cpu@0 {
+ device_type = "cpu";
+ compatible = "arm,cortex-a57", "arm,armv8";
+ reg = <0 0>;
+ enable-method = "spin-table";
+ cpu-release-addr = <0 0x84b00000>;
+ };
+
+ cpu@1 {
+ device_type = "cpu";
+ compatible = "arm,cortex-a57", "arm,armv8";
+ reg = <0 1>;
+ enable-method = "spin-table";
+ cpu-release-addr = <0 0x84b00000>;
+ };
+
+ cpu@2 {
+ device_type = "cpu";
+ compatible = "arm,cortex-a57", "arm,armv8";
+ reg = <0 2>;
+ enable-method = "spin-table";
+ cpu-release-addr = <0 0x84b00000>;
+ };
+
+ cpu@3 {
+ device_type = "cpu";
+ compatible = "arm,cortex-a57", "arm,armv8";
+ reg = <0 3>;
+ enable-method = "spin-table";
+ cpu-release-addr = <0 0x84b00000>;
+ };
+ };
+
+ timer {
+ compatible = "arm,armv8-timer";
+ interrupts = <GIC_PPI 13 (GIC_CPU_MASK_RAW(0xff) |
+ IRQ_TYPE_EDGE_RISING)>,
+ <GIC_PPI 14 (GIC_CPU_MASK_RAW(0xff) |
+ IRQ_TYPE_EDGE_RISING)>,
+ <GIC_PPI 11 (GIC_CPU_MASK_RAW(0xff) |
+ IRQ_TYPE_EDGE_RISING)>,
+ <GIC_PPI 10 (GIC_CPU_MASK_RAW(0xff) |
+ IRQ_TYPE_EDGE_RISING)>;
+ };
+
+ soc: soc {
+ compatible = "simple-bus";
+ #address-cells = <1>;
+ #size-cells = <1>;
+ ranges = <0 0 0 0xffffffff>;
+
+ gic: interrupt-controller@65210000 {
+ compatible = "arm,gic-400";
+ #interrupt-cells = <3>;
+ interrupt-controller;
+ reg = <0x65210000 0x1000>,
+ <0x65220000 0x1000>,
+ <0x65240000 0x2000>,
+ <0x65260000 0x1000>;
+ };
+
+ uart3: serial@66130000 {
+ compatible = "snps,dw-apb-uart";
+ reg = <0x66130000 0x100>;
+ interrupts = <GIC_SPI 393 IRQ_TYPE_LEVEL_HIGH>;
+ reg-shift = <2>;
+ reg-io-width = <4>;
+ clock-frequency = <23961600>;
+ status = "disabled";
+ };
+ };
+};
--
1.7.9.5
On Tue, Jul 21, 2015 at 06:29:40PM -0700, Ray Jui wrote:
> This patch enables arm64 support to the iProc PCIe driver
>
> Note struct pci_sys_data is arm32 specific and will eventually be
> removed. This change is done in such a way that when struct pci_sys_data
> is removed from arm32, one only needs to also remove it from
> pcie-iproc.h, no other change in the iProc PCIe core driver is needed
>
> In addition, arm64 based PCI driver does not require call to
> pci_fixup_irqs, as it implements OF based irq parsing and mapping in
> pcibios_add_device
>
> Signed-off-by: Ray Jui <[email protected]>
> Reviewed-by: Scott Branden <[email protected]>
> ---
> drivers/pci/host/pcie-iproc.c | 15 ++++-----------
> drivers/pci/host/pcie-iproc.h | 8 ++++++--
> 2 files changed, 10 insertions(+), 13 deletions(-)
>
> diff --git a/drivers/pci/host/pcie-iproc.c b/drivers/pci/host/pcie-iproc.c
> index d77481e..8a556d5 100644
> --- a/drivers/pci/host/pcie-iproc.c
> +++ b/drivers/pci/host/pcie-iproc.c
> @@ -58,11 +58,6 @@
> #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;
> -}
> -
> /**
> * Note access to the configuration registers are protected at the higher layer
> * by 'pci_lock' in drivers/pci/access.c
> @@ -71,8 +66,7 @@ static void __iomem *iproc_pcie_map_cfg_bus(struct pci_bus *bus,
> unsigned int devfn,
> int where)
> {
> - struct pci_sys_data *sys = bus->sysdata;
> - struct iproc_pcie *pcie = sys_to_pcie(sys);
> + struct iproc_pcie *pcie = bus->sysdata;
I'm thinking something like the following so we don't depend on
pci_sys_data being at the beginning of iproc_pcie. What do you think?
It has more ifdefs but feels a bit safer. And I don't mind if ugly code
comes with ugly ifdefs -- it's a little incentive to make the code cleaner.
commit 8d9bfe3702aaea457b3d59b09b86e9f03c322605
Author: Ray Jui <[email protected]>
Date: Tue Jul 21 18:29:40 2015 -0700
PCI: iproc: Add arm64 support
Add arm64 support to the iProc PCIe driver.
Note that on arm32, bus->sysdata points to the arm32-specific pci_sys_data
struct, and pci_sys_data.private_data contains the iproc_pcie pointer.
For arm64, there's nothing corresponding to pci_sys_data, so we keep the
iproc_pcie pointer directly in bus->sysdata.
In addition, arm64 does IRQ mapping in pcibios_add_device(), so it doesn't
need pci_fixup_irqs() as arm32 does.
Signed-off-by: Ray Jui <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
Reviewed-by: Scott Branden <[email protected]>
diff --git a/drivers/pci/host/pcie-iproc.c b/drivers/pci/host/pcie-iproc.c
index 9a00dca..fe2efb1 100644
--- a/drivers/pci/host/pcie-iproc.c
+++ b/drivers/pci/host/pcie-iproc.c
@@ -58,9 +58,17 @@
#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)
+static inline struct iproc_pcie *iproc_data(struct pci_bus *bus)
{
- return sys->private_data;
+ struct iproc_pcie *pcie;
+#ifdef CONFIG_ARM
+ struct pci_sys_data *sys = bus->sysdata;
+
+ pcie = sys->private_data;
+#else
+ pcie = bus->sysdata;
+#endif
+ return pcie;
}
/**
@@ -71,8 +79,7 @@ static void __iomem *iproc_pcie_map_cfg_bus(struct pci_bus *bus,
unsigned int devfn,
int where)
{
- struct pci_sys_data *sys = bus->sysdata;
- struct iproc_pcie *pcie = sys_to_pcie(sys);
+ struct iproc_pcie *pcie = iproc_data(bus);
unsigned slot = PCI_SLOT(devfn);
unsigned fn = PCI_FUNC(devfn);
unsigned busno = bus->number;
@@ -186,6 +193,7 @@ static void iproc_pcie_enable(struct iproc_pcie *pcie)
int iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *res)
{
int ret;
+ void *sysdata;
struct pci_bus *bus;
if (!pcie || !pcie->dev || !pcie->base)
@@ -205,10 +213,14 @@ int iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *res)
iproc_pcie_reset(pcie);
+#ifdef CONFIG_ARM
pcie->sysdata.private_data = pcie;
+ sysdata = &pcie->sysdata;
+#else
+ sysdata = pcie;
+#endif
- bus = pci_create_root_bus(pcie->dev, 0, &iproc_pcie_ops,
- &pcie->sysdata, res);
+ bus = pci_create_root_bus(pcie->dev, 0, &iproc_pcie_ops, sysdata, res);
if (!bus) {
dev_err(pcie->dev, "unable to create PCI root bus\n");
ret = -ENOMEM;
@@ -226,7 +238,9 @@ int iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *res)
pci_scan_child_bus(bus);
pci_assign_unassigned_bus_resources(bus);
+#ifdef CONFIG_ARM
pci_fixup_irqs(pci_common_swizzle, pcie->map_irq);
+#endif
pci_bus_add_devices(bus);
return 0;
diff --git a/drivers/pci/host/pcie-iproc.h b/drivers/pci/host/pcie-iproc.h
index ba0a108..c9e4c10 100644
--- a/drivers/pci/host/pcie-iproc.h
+++ b/drivers/pci/host/pcie-iproc.h
@@ -21,7 +21,7 @@
* @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
+ * @sysdata: Per PCI controller data (ARM-specific)
* @root_bus: pointer to root bus
* @phy: optional PHY device that controls the Serdes
* @irqs: interrupt IDs
@@ -29,7 +29,9 @@
struct iproc_pcie {
struct device *dev;
void __iomem *base;
+#ifdef CONFIG_ARM
struct pci_sys_data sysdata;
+#endif
struct pci_bus *root_bus;
struct phy *phy;
int irqs[IPROC_PCIE_MAX_NUM_IRQS];
On 7/22/2015 1:53 PM, Bjorn Helgaas wrote:
> On Tue, Jul 21, 2015 at 06:29:40PM -0700, Ray Jui wrote:
>> This patch enables arm64 support to the iProc PCIe driver
>>
>> Note struct pci_sys_data is arm32 specific and will eventually be
>> removed. This change is done in such a way that when struct pci_sys_data
>> is removed from arm32, one only needs to also remove it from
>> pcie-iproc.h, no other change in the iProc PCIe core driver is needed
>>
>> In addition, arm64 based PCI driver does not require call to
>> pci_fixup_irqs, as it implements OF based irq parsing and mapping in
>> pcibios_add_device
>>
>> Signed-off-by: Ray Jui <[email protected]>
>> Reviewed-by: Scott Branden <[email protected]>
>> ---
>> drivers/pci/host/pcie-iproc.c | 15 ++++-----------
>> drivers/pci/host/pcie-iproc.h | 8 ++++++--
>> 2 files changed, 10 insertions(+), 13 deletions(-)
>>
>> diff --git a/drivers/pci/host/pcie-iproc.c b/drivers/pci/host/pcie-iproc.c
>> index d77481e..8a556d5 100644
>> --- a/drivers/pci/host/pcie-iproc.c
>> +++ b/drivers/pci/host/pcie-iproc.c
>> @@ -58,11 +58,6 @@
>> #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;
>> -}
>> -
>> /**
>> * Note access to the configuration registers are protected at the higher layer
>> * by 'pci_lock' in drivers/pci/access.c
>> @@ -71,8 +66,7 @@ static void __iomem *iproc_pcie_map_cfg_bus(struct pci_bus *bus,
>> unsigned int devfn,
>> int where)
>> {
>> - struct pci_sys_data *sys = bus->sysdata;
>> - struct iproc_pcie *pcie = sys_to_pcie(sys);
>> + struct iproc_pcie *pcie = bus->sysdata;
>
> I'm thinking something like the following so we don't depend on
> pci_sys_data being at the beginning of iproc_pcie. What do you think?
> It has more ifdefs but feels a bit safer. And I don't mind if ugly code
> comes with ugly ifdefs -- it's a little incentive to make the code cleaner.
>
>
> commit 8d9bfe3702aaea457b3d59b09b86e9f03c322605
> Author: Ray Jui <[email protected]>
> Date: Tue Jul 21 18:29:40 2015 -0700
>
> PCI: iproc: Add arm64 support
>
> Add arm64 support to the iProc PCIe driver.
>
> Note that on arm32, bus->sysdata points to the arm32-specific pci_sys_data
> struct, and pci_sys_data.private_data contains the iproc_pcie pointer.
> For arm64, there's nothing corresponding to pci_sys_data, so we keep the
> iproc_pcie pointer directly in bus->sysdata.
>
> In addition, arm64 does IRQ mapping in pcibios_add_device(), so it doesn't
> need pci_fixup_irqs() as arm32 does.
>
> Signed-off-by: Ray Jui <[email protected]>
> Signed-off-by: Bjorn Helgaas <[email protected]>
> Reviewed-by: Scott Branden <[email protected]>
>
> diff --git a/drivers/pci/host/pcie-iproc.c b/drivers/pci/host/pcie-iproc.c
> index 9a00dca..fe2efb1 100644
> --- a/drivers/pci/host/pcie-iproc.c
> +++ b/drivers/pci/host/pcie-iproc.c
> @@ -58,9 +58,17 @@
> #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)
> +static inline struct iproc_pcie *iproc_data(struct pci_bus *bus)
> {
> - return sys->private_data;
> + struct iproc_pcie *pcie;
> +#ifdef CONFIG_ARM
> + struct pci_sys_data *sys = bus->sysdata;
> +
> + pcie = sys->private_data;
> +#else
> + pcie = bus->sysdata;
> +#endif
> + return pcie;
> }
>
> /**
> @@ -71,8 +79,7 @@ static void __iomem *iproc_pcie_map_cfg_bus(struct pci_bus *bus,
> unsigned int devfn,
> int where)
> {
> - struct pci_sys_data *sys = bus->sysdata;
> - struct iproc_pcie *pcie = sys_to_pcie(sys);
> + struct iproc_pcie *pcie = iproc_data(bus);
> unsigned slot = PCI_SLOT(devfn);
> unsigned fn = PCI_FUNC(devfn);
> unsigned busno = bus->number;
> @@ -186,6 +193,7 @@ static void iproc_pcie_enable(struct iproc_pcie *pcie)
> int iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *res)
> {
> int ret;
> + void *sysdata;
> struct pci_bus *bus;
>
> if (!pcie || !pcie->dev || !pcie->base)
> @@ -205,10 +213,14 @@ int iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *res)
>
> iproc_pcie_reset(pcie);
>
> +#ifdef CONFIG_ARM
> pcie->sysdata.private_data = pcie;
> + sysdata = &pcie->sysdata;
> +#else
> + sysdata = pcie;
> +#endif
>
> - bus = pci_create_root_bus(pcie->dev, 0, &iproc_pcie_ops,
> - &pcie->sysdata, res);
> + bus = pci_create_root_bus(pcie->dev, 0, &iproc_pcie_ops, sysdata, res);
> if (!bus) {
> dev_err(pcie->dev, "unable to create PCI root bus\n");
> ret = -ENOMEM;
> @@ -226,7 +238,9 @@ int iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *res)
>
> pci_scan_child_bus(bus);
> pci_assign_unassigned_bus_resources(bus);
> +#ifdef CONFIG_ARM
> pci_fixup_irqs(pci_common_swizzle, pcie->map_irq);
> +#endif
> pci_bus_add_devices(bus);
>
> return 0;
> diff --git a/drivers/pci/host/pcie-iproc.h b/drivers/pci/host/pcie-iproc.h
> index ba0a108..c9e4c10 100644
> --- a/drivers/pci/host/pcie-iproc.h
> +++ b/drivers/pci/host/pcie-iproc.h
> @@ -21,7 +21,7 @@
> * @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
> + * @sysdata: Per PCI controller data (ARM-specific)
> * @root_bus: pointer to root bus
> * @phy: optional PHY device that controls the Serdes
> * @irqs: interrupt IDs
> @@ -29,7 +29,9 @@
> struct iproc_pcie {
> struct device *dev;
> void __iomem *base;
> +#ifdef CONFIG_ARM
> struct pci_sys_data sysdata;
> +#endif
> struct pci_bus *root_bus;
> struct phy *phy;
> int irqs[IPROC_PCIE_MAX_NUM_IRQS];
>
I'm fine with the entire change. Both of us understand it's temporary
and will be removed as soon as struct pci_sys_data is removed from ARM32.
Note I don't know how the code merge should work here. The iProc PCIe
driver is enabled when CONFIG_ARCH_BCM_IPROC is enabled. Note patch #3
of this series enables CONFIG_ARCH_BCM_IPROC for arm64. Without this
patch, arm64 build will be broken because of struct pci_sys_data.
I know you are taking PCI changes and I assume Catalin will be merging
arm64 related changes. But patches in this patch series need to go
together to keep things intact.
Thanks,
Ray
On Wed, Jul 22, 2015 at 02:09:24PM -0700, Ray Jui wrote:
> I'm fine with the entire change. Both of us understand it's temporary
> and will be removed as soon as struct pci_sys_data is removed from ARM32.
>
> Note I don't know how the code merge should work here. The iProc PCIe
> driver is enabled when CONFIG_ARCH_BCM_IPROC is enabled. Note patch #3
> of this series enables CONFIG_ARCH_BCM_IPROC for arm64. Without this
> patch, arm64 build will be broken because of struct pci_sys_data.
>
> I know you are taking PCI changes and I assume Catalin will be merging
> arm64 related changes. But patches in this patch series need to go
> together to keep things intact.
The easiest thing to do would be to merge them all through one tree.
Catalin could ack the arm64 config changes and I could merge everything via
PCI, or I could ack the PCI changes and he could merge them.
I propose the former since the PCI changes are more substantive, but if the
latter works better, feel free to add my:
Acked-by: Bjorn Helgaas <[email protected]>
to the revised first patch and the second one.
Bjorn
(adding Olof)
On Wed, Jul 22, 2015 at 02:29:42AM +0100, Ray Jui wrote:
> This patch adds support to Broadcom's iProc family of arm64 based SoCs
> in the arm64 Kconfig and defconfig files
>
> Signed-off-by: Ray Jui <[email protected]>
> Reviewed-by: Scott Branden <[email protected]>
> ---
> arch/arm64/Kconfig | 5 +++++
> arch/arm64/configs/defconfig | 2 ++
> 2 files changed, 7 insertions(+)
>
> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> index 318175f..969ef4a 100644
> --- a/arch/arm64/Kconfig
> +++ b/arch/arm64/Kconfig
> @@ -162,6 +162,11 @@ source "kernel/Kconfig.freezer"
>
> menu "Platform selection"
>
> +config ARCH_BCM_IPROC
> + bool "Broadcom iProc SoC Family"
> + help
> + This enables support for Broadcom iProc based SoCs
> +
The arm-soc guys have a patch moving the platform-related config options
out into Kconfig.platforms, so I think you need to co-ordinate with them.
> config ARCH_EXYNOS
> bool
> help
> diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig
> index 4e17e7e..c83d51f 100644
> --- a/arch/arm64/configs/defconfig
> +++ b/arch/arm64/configs/defconfig
> @@ -31,6 +31,7 @@ CONFIG_MODULES=y
> CONFIG_MODULE_UNLOAD=y
> # CONFIG_BLK_DEV_BSG is not set
> # CONFIG_IOSCHED_DEADLINE is not set
> +CONFIG_ARCH_BCM_IPROC=y
> CONFIG_ARCH_EXYNOS7=y
> CONFIG_ARCH_FSL_LS2085A=y
> CONFIG_ARCH_HISI=y
> @@ -102,6 +103,7 @@ CONFIG_SERIO_AMBAKMI=y
> CONFIG_LEGACY_PTY_COUNT=16
> CONFIG_SERIAL_8250=y
> CONFIG_SERIAL_8250_CONSOLE=y
> +CONFIG_SERIAL_8250_DW=y
> CONFIG_SERIAL_8250_MT6577=y
> CONFIG_SERIAL_AMBA_PL011=y
> CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
I'm fine with this.
Will
Hi Will,
On 7/24/2015 1:51 AM, Will Deacon wrote:
> (adding Olof)
>
> On Wed, Jul 22, 2015 at 02:29:42AM +0100, Ray Jui wrote:
>> This patch adds support to Broadcom's iProc family of arm64 based SoCs
>> in the arm64 Kconfig and defconfig files
>>
>> Signed-off-by: Ray Jui <[email protected]>
>> Reviewed-by: Scott Branden <[email protected]>
>> ---
>> arch/arm64/Kconfig | 5 +++++
>> arch/arm64/configs/defconfig | 2 ++
>> 2 files changed, 7 insertions(+)
>>
>> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
>> index 318175f..969ef4a 100644
>> --- a/arch/arm64/Kconfig
>> +++ b/arch/arm64/Kconfig
>> @@ -162,6 +162,11 @@ source "kernel/Kconfig.freezer"
>>
>> menu "Platform selection"
>>
>> +config ARCH_BCM_IPROC
>> + bool "Broadcom iProc SoC Family"
>> + help
>> + This enables support for Broadcom iProc based SoCs
>> +
>
> The arm-soc guys have a patch moving the platform-related config options
> out into Kconfig.platforms, so I think you need to co-ordinate with them.
>
Thanks for letting me know. I see that those patches are already merged
into for-next under arm-soc. I guess I should generate a new patch
series based on for-next and move this config option to Kconfig.platform?
Olof, any comment?
>> config ARCH_EXYNOS
>> bool
>> help
>> diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig
>> index 4e17e7e..c83d51f 100644
>> --- a/arch/arm64/configs/defconfig
>> +++ b/arch/arm64/configs/defconfig
>> @@ -31,6 +31,7 @@ CONFIG_MODULES=y
>> CONFIG_MODULE_UNLOAD=y
>> # CONFIG_BLK_DEV_BSG is not set
>> # CONFIG_IOSCHED_DEADLINE is not set
>> +CONFIG_ARCH_BCM_IPROC=y
>> CONFIG_ARCH_EXYNOS7=y
>> CONFIG_ARCH_FSL_LS2085A=y
>> CONFIG_ARCH_HISI=y
>> @@ -102,6 +103,7 @@ CONFIG_SERIO_AMBAKMI=y
>> CONFIG_LEGACY_PTY_COUNT=16
>> CONFIG_SERIAL_8250=y
>> CONFIG_SERIAL_8250_CONSOLE=y
>> +CONFIG_SERIAL_8250_DW=y
>> CONFIG_SERIAL_8250_MT6577=y
>> CONFIG_SERIAL_AMBA_PL011=y
>> CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
>
> I'm fine with this.
>
> Will
>
On Fri, Jul 24, 2015 at 5:42 PM, Ray Jui <[email protected]> wrote:
> Hi Will,
>
> On 7/24/2015 1:51 AM, Will Deacon wrote:
>> (adding Olof)
>>
>> On Wed, Jul 22, 2015 at 02:29:42AM +0100, Ray Jui wrote:
>>> This patch adds support to Broadcom's iProc family of arm64 based SoCs
>>> in the arm64 Kconfig and defconfig files
>>>
>>> Signed-off-by: Ray Jui <[email protected]>
>>> Reviewed-by: Scott Branden <[email protected]>
>>> ---
>>> arch/arm64/Kconfig | 5 +++++
>>> arch/arm64/configs/defconfig | 2 ++
>>> 2 files changed, 7 insertions(+)
>>>
>>> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
>>> index 318175f..969ef4a 100644
>>> --- a/arch/arm64/Kconfig
>>> +++ b/arch/arm64/Kconfig
>>> @@ -162,6 +162,11 @@ source "kernel/Kconfig.freezer"
>>>
>>> menu "Platform selection"
>>>
>>> +config ARCH_BCM_IPROC
>>> + bool "Broadcom iProc SoC Family"
>>> + help
>>> + This enables support for Broadcom iProc based SoCs
>>> +
>>
>> The arm-soc guys have a patch moving the platform-related config options
>> out into Kconfig.platforms, so I think you need to co-ordinate with them.
>>
>
> Thanks for letting me know. I see that those patches are already merged
> into for-next under arm-soc. I guess I should generate a new patch
> series based on for-next and move this config option to Kconfig.platform?
>
> Olof, any comment?
Yeah, please do that and send that series with [email protected] on cc,
we'll apply it from there!
Thanks,
-Olof
Hi Olof,
On 7/27/2015 11:35 AM, Olof Johansson wrote:
> On Fri, Jul 24, 2015 at 5:42 PM, Ray Jui <[email protected]> wrote:
>> Hi Will,
>>
>> On 7/24/2015 1:51 AM, Will Deacon wrote:
>>> (adding Olof)
>>>
>>> On Wed, Jul 22, 2015 at 02:29:42AM +0100, Ray Jui wrote:
>>>> This patch adds support to Broadcom's iProc family of arm64 based SoCs
>>>> in the arm64 Kconfig and defconfig files
>>>>
>>>> Signed-off-by: Ray Jui <[email protected]>
>>>> Reviewed-by: Scott Branden <[email protected]>
>>>> ---
>>>> arch/arm64/Kconfig | 5 +++++
>>>> arch/arm64/configs/defconfig | 2 ++
>>>> 2 files changed, 7 insertions(+)
>>>>
>>>> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
>>>> index 318175f..969ef4a 100644
>>>> --- a/arch/arm64/Kconfig
>>>> +++ b/arch/arm64/Kconfig
>>>> @@ -162,6 +162,11 @@ source "kernel/Kconfig.freezer"
>>>>
>>>> menu "Platform selection"
>>>>
>>>> +config ARCH_BCM_IPROC
>>>> + bool "Broadcom iProc SoC Family"
>>>> + help
>>>> + This enables support for Broadcom iProc based SoCs
>>>> +
>>>
>>> The arm-soc guys have a patch moving the platform-related config options
>>> out into Kconfig.platforms, so I think you need to co-ordinate with them.
>>>
>>
>> Thanks for letting me know. I see that those patches are already merged
>> into for-next under arm-soc. I guess I should generate a new patch
>> series based on for-next and move this config option to Kconfig.platform?
>>
>> Olof, any comment?
>
> Yeah, please do that and send that series with [email protected] on cc,
> we'll apply it from there!
>
>
> Thanks,
>
> -Olof
>
Note this patch series contains some iProc PCIe patches as per-requsite
that need to be applied before the ARCH_BCM_IPROC arm64 patch. Without
the PCIe patch, arm64 build out of defconfig would be broken.
Bjorn has acked the PCIe patches. I'll re-send the entire patch series
to you and [email protected] and you can merge them there.
Thanks,
Ray