2015-07-22 01:29:31

by Ray Jui

[permalink] [raw]
Subject: [PATCH v4 0/4] Add Broadcom North Star 2 support

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


2015-07-22 01:29:34

by Ray Jui

[permalink] [raw]
Subject: [PATCH v4 1/4] PCI: iproc: enable arm64 support for iProc PCIe

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

2015-07-22 01:30:47

by Ray Jui

[permalink] [raw]
Subject: [PATCH v4 2/4] PCI: iproc: Fix ARM64 dependency in Kconfig

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

2015-07-22 01:29:39

by Ray Jui

[permalink] [raw]
Subject: [PATCH v4 3/4] arm64: Add Broadcom iProc family support

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

2015-07-22 01:29:59

by Ray Jui

[permalink] [raw]
Subject: [PATCH v4 4/4] arm64: dts: Add Broadcom North Star 2 support

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

2015-07-22 20:53:25

by Bjorn Helgaas

[permalink] [raw]
Subject: Re: [PATCH v4 1/4] PCI: iproc: enable arm64 support for iProc PCIe

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];

2015-07-22 21:09:35

by Ray Jui

[permalink] [raw]
Subject: Re: [PATCH v4 1/4] PCI: iproc: enable arm64 support for iProc PCIe



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

2015-07-22 21:40:14

by Bjorn Helgaas

[permalink] [raw]
Subject: Re: [PATCH v4 1/4] PCI: iproc: enable arm64 support for iProc PCIe

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

2015-07-24 08:51:09

by Will Deacon

[permalink] [raw]
Subject: Re: [PATCH v4 3/4] arm64: Add Broadcom iProc family support

(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

2015-07-24 15:42:52

by Ray Jui

[permalink] [raw]
Subject: Re: [PATCH v4 3/4] arm64: Add Broadcom iProc family support

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
>

2015-07-27 18:35:10

by Olof Johansson

[permalink] [raw]
Subject: Re: [PATCH v4 3/4] arm64: Add Broadcom iProc family support

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

2015-07-27 18:45:04

by Ray Jui

[permalink] [raw]
Subject: Re: [PATCH v4 3/4] arm64: Add Broadcom iProc family support

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