[PATCH u-boot v2019.04-aspeed-openbmc 1/4] arm/dts: Add Genesis3 board
Joel Stanley
joel at jms.id.au
Thu Apr 21 15:52:18 AEST 2022
On Wed, 20 Apr 2022 at 10:25, Patrick Rudolph
<patrick.rudolph at 9elements.com> wrote:
>
> Add IBM Genesis3 board and devicetree source file.
>
> Signed-off-by: Patrick Rudolph <patrick.rudolph at 9elements.com>
> ---
> arch/arm/dts/Makefile | 1 +
> arch/arm/dts/ibm-genesis3.dts | 83 ++++++++++++++++++++++++++++
> arch/arm/mach-aspeed/ast2500/Kconfig | 11 ++++
> board/ibm/genesis3/Kconfig | 13 +++++
> board/ibm/genesis3/Makefile | 1 +
> board/ibm/genesis3/ibm_genesis3.c | 5 ++
> include/configs/ibm_genesis3.h | 19 +++++++
> 7 files changed, 133 insertions(+)
> create mode 100644 arch/arm/dts/ibm-genesis3.dts
> create mode 100644 board/ibm/genesis3/Kconfig
> create mode 100644 board/ibm/genesis3/Makefile
> create mode 100644 board/ibm/genesis3/ibm_genesis3.c
> create mode 100644 include/configs/ibm_genesis3.h
>
> diff --git a/arch/arm/dts/Makefile b/arch/arm/dts/Makefile
> index a79f885f54..fae23e6bc1 100755
> --- a/arch/arm/dts/Makefile
> +++ b/arch/arm/dts/Makefile
> @@ -677,6 +677,7 @@ dtb-$(CONFIG_ARCH_BCM6858) += \
> dtb-$(CONFIG_ARCH_ASPEED) += \
> ast2400-evb.dtb \
> ast2500-evb.dtb \
> + ibm-genesis3.dtb \
> ast2600a0-evb.dtb \
> ast2600a1-evb.dtb \
> ast2600-bletchley.dtb \
> diff --git a/arch/arm/dts/ibm-genesis3.dts b/arch/arm/dts/ibm-genesis3.dts
> new file mode 100644
> index 0000000000..ba610c9032
> --- /dev/null
> +++ b/arch/arm/dts/ibm-genesis3.dts
> @@ -0,0 +1,83 @@
> +/dts-v1/;
> +
> +#include "ast2500-u-boot.dtsi"
> +
> +/ {
> + model = "IBM Genesis3";
> + compatible = "ibm,genesis3", "aspeed,ast2500";
> +
> + memory {
> + device_type = "memory";
> + reg = <0x80000000 0x20000000>;
> + };
> +
> + chosen {
> + stdout-path = &uart5;
> + };
> +
> + aliases {
> + spi0 = &fmc;
> + ethernet0 = &mac0;
> + ethernet1 = &mac1;
> + };
> +};
> +
> +&uart5 {
> + u-boot,dm-pre-reloc;
> + status = "okay";
> +};
> +
> +&sdrammc {
> + clock-frequency = <400000000>;
> +};
> +
> +&wdt1 {
> + u-boot,dm-pre-reloc;
> + status = "okay";
> +};
> +
> +&wdt2 {
> + u-boot,dm-pre-reloc;
> + status = "okay";
> +};
> +
> +&wdt3 {
> + u-boot,dm-pre-reloc;
> + status = "okay";
> +};
> +
> +&mac0 {
> + status = "okay";
> + phy-mode = "rgmii";
> +
> + pinctrl-names = "default";
> + pinctrl-0 = <&pinctrl_mac1link_default &pinctrl_mdio1_default>;
> +};
> +
> +&mac1 {
> + status = "okay";
> + phy-mode = "rgmii";
> + pinctrl-names = "default";
> + pinctrl-0 = <&pinctrl_mac2link_default &pinctrl_mdio2_default>;
> +};
> +
> +&fmc {
> + status = "okay";
> + timing-calibration-disabled;
> + flash at 0 {
> + compatible = "spi-flash", "spansion,s25fl256l";
> + status = "okay";
> + spi-max-frequency = <50000000>;
> + spi-tx-bus-width = <2>;
> + spi-rx-bus-width = <2>;
> + };
> +
> + flash at 1 {
> + compatible = "spi-flash", "spansion,s25fl256l";
> + status = "okay";
> + spi-max-frequency = <50000000>;
> + spi-tx-bus-width = <2>;
> + spi-rx-bus-width = <2>;
> + };
> +};
> +
> diff --git a/arch/arm/mach-aspeed/ast2500/Kconfig b/arch/arm/mach-aspeed/ast2500/Kconfig
> index 4f992f442d..81c4fd0c2f 100644
> --- a/arch/arm/mach-aspeed/ast2500/Kconfig
> +++ b/arch/arm/mach-aspeed/ast2500/Kconfig
> @@ -15,8 +15,19 @@ config TARGET_EVB_AST2500
> It has 512M of RAM, 32M of SPI flash, two Ethernet ports,
> 4 Serial ports, 4 USB ports, VGA port, PCIe, SD card slot,
> 20 pin JTAG, pinouts for 14 I2Cs, 3 SPIs and eSPI, 8 PWMs.
> +
> +config TARGET_IBM_GENESIS3
> + bool "IBM-Genesis3"
> + help
> + IBM Genesis3 is an IBM evaluation board for Intel Xeon
> + and AST2500 BMC. It has 512M of RAM, 32M of SPI flash,
> + two Ethernet ports, 1 debug uart, 1 USB ports and 1 PCIe
> + port connected to the PCH.
> + It has support for JTAG, pinouts for 9 I2Cs and eSPI.
Given this is another aspeed system, would it make sense to group it
the same way we've grouped the existing boards under an "aspeed"
prefix? (The rainier system is an IBM system, FWIW).
The board file looks to be similar enough to the existing evb-ast2500.
Can you make a case for maintaining another one, or should we re-use
what's there like the other platforms have done?
> +
> endchoice
>
> source "board/aspeed/evb_ast2500/Kconfig"
> +source "board/ibm/genesis3/Kconfig"
>
> endif
> diff --git a/board/ibm/genesis3/Kconfig b/board/ibm/genesis3/Kconfig
> new file mode 100644
> index 0000000000..09fe60262b
> --- /dev/null
> +++ b/board/ibm/genesis3/Kconfig
> @@ -0,0 +1,13 @@
> +if TARGET_IBM_GENESIS3
> +
> +config SYS_BOARD
> + default "genesis3"
> +
> +config SYS_VENDOR
> + default "ibm"
> +
> +config SYS_CONFIG_NAME
> + string "board configuration name"
> + default "ibm_genesis3"
> +
> +endif
> diff --git a/board/ibm/genesis3/Makefile b/board/ibm/genesis3/Makefile
> new file mode 100644
> index 0000000000..2bb8917fab
> --- /dev/null
> +++ b/board/ibm/genesis3/Makefile
> @@ -0,0 +1 @@
> +obj-y += ibm_genesis3.o
> diff --git a/board/ibm/genesis3/ibm_genesis3.c b/board/ibm/genesis3/ibm_genesis3.c
> new file mode 100644
> index 0000000000..9de8531806
> --- /dev/null
> +++ b/board/ibm/genesis3/ibm_genesis3.c
> @@ -0,0 +1,5 @@
> +// SPDX-License-Identifier: GPL-2.0+
> +/*
> + * Copyright (c) 2022 9elements GmbH
> + */
> +#include <common.h>
> diff --git a/include/configs/ibm_genesis3.h b/include/configs/ibm_genesis3.h
> new file mode 100644
> index 0000000000..f40aeed8a9
> --- /dev/null
> +++ b/include/configs/ibm_genesis3.h
> @@ -0,0 +1,19 @@
> +/* SPDX-License-Identifier: GPL-2.0+ */
> +/*
> + * Copyright 2022 9elements GmbH.
> + */
> +
> +#ifndef __CONFIG_H
> +#define __CONFIG_H
> +
> +#include <configs/aspeed-common.h>
> +
> +#define CONFIG_SYS_MEMTEST_START (CONFIG_SYS_SDRAM_BASE + 0x300000)
> +#define CONFIG_SYS_MEMTEST_END (CONFIG_SYS_MEMTEST_START + 0x5000000)
> +
> +#define CONFIG_SYS_UBOOT_BASE CONFIG_SYS_TEXT_BASE
> +
> +/* Memory Info */
> +#define CONFIG_SYS_LOAD_ADDR 0x83000000
> +
> +#endif /* __CONFIG_H */
> --
> 2.35.1
>
More information about the openbmc
mailing list