Add EEPROM-based board detect support for TI
Subsystems (3)
cover Add EEPROM-based board detect support for TI
This series adds the pieces needed to fully configure the TI AM654x EVM and its associated (optional) daughtercards using a GPIO-based card presence detection scheme. It then reads out and parses any I2C EEPROM data present on any such boards for example board name, serial #, Ethernet MAC addresses, etc. and incorporates this data into the current U-Boot environment. One important aspect is the application of daughercard-specific DTB overlays right before the Kernel boot happens.
This series also adds basic I2C support for TI K3 family devices which is a prerequisite for the EEPROM access.
--
Andreas Dannenberg
Texas Instruments Inc
Andreas Dannenberg (16):
arm: dts: k3-am65: Move pinctrl nodes out of U-Boot specific dtsi
arm: dts: k3-am65: Add I2C nodes
arm: dts: k3-am654-base-board: Enable wkup_i2c0 across all boot stages
configs: am65x_evm_defconfig: Enable I2C support
configs: am65x_hs_evm_defconfig: Enable I2C support
arm64: dts: k3-am654-base-board: Add I2C GPIO expander @ 0x38
configs: am65x_evm_a53: Enable PCA953X-type GPIO driver
configs: am65x_hs_evm_a53: Enable PCA953X-type GPIO driver
configs: am65x_evm_a53: Increase malloc pool before relocation
configs: am65x_hs_evm_a53: Increase malloc pool before relocation
arm: K3: am654: Map common EEPROM data into SRAM scratch space
ti: common: am6: Add support for board description EEPROM
ti: common: am6: Add support for setting MAC addresses
board: ti: am654: Use EEPROM-based board detection
configs: am65x_evm: Add support for applying overlays
configs: am65x_hs_evm: Add support for applying overlays
Vignesh R (2):
i2c: omap24xx_i2c: Adapt driver to support K3 devices
arm: omap_i2c: Remove unwanted header file inclusion
arch/arm/dts/k3-am65-main.dtsi | 60 +++++ arch/arm/dts/k3-am65-mcu.dtsi | 11 + arch/arm/dts/k3-am65-wakeup.dtsi | 19 ++ arch/arm/dts/k3-am65.dtsi | 7 + arch/arm/dts/k3-am654-base-board-u-boot.dtsi | 35 +-- arch/arm/dts/k3-am654-base-board.dts | 22 ++ arch/arm/dts/k3-am654-r5-base-board.dts | 13 + arch/arm/include/asm/omap_i2c.h | 2 - arch/arm/mach-k3/am6_init.c | 4 + arch/arm/mach-k3/include/mach/am6_hardware.h | 3 + arch/arm/mach-k3/include/mach/sys_proto.h | 2 + board/ti/am65x/Kconfig | 7 + board/ti/am65x/evm.c | 241 ++++++++++++++++++ board/ti/common/board_detect.c | 246 +++++++++++++++++++ board/ti/common/board_detect.h | 206 ++++++++++++++++ configs/am65x_evm_a53_defconfig | 11 +- configs/am65x_evm_r5_defconfig | 4 + configs/am65x_hs_evm_a53_defconfig | 11 +- configs/am65x_hs_evm_r5_defconfig | 4 + drivers/i2c/Kconfig | 2 +- include/configs/am65x_evm.h | 18 +- 21 files changed, 892 insertions(+), 36 deletions(-)
Signature · 1 line
2.17.1
PATCH 1/18 i2c: omap24xx_i2c: Adapt driver to support K3 devices
From: Vignesh R <[email protected]>
K3 devices have I2C IP that is same as OMAP2+ family. Allow driver to be compiled for ARCH_K3.
--- drivers/i2c/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/i2c/Kconfig +1 -1
@@ -327,7 +327,7 @@ endif config SYS_I2C_OMAP24XX bool "TI OMAP2+ I2C driver" - depends on ARCH_OMAP2PLUS + depends on ARCH_OMAP2PLUS || ARCH_K3 help Add support for the OMAP2+ I2C driver.
Signature · 1 line
2.17.1
PATCH 1/18 i2c: omap24xx_i2c: Adapt driver to support K3 devices
Hello Andreas,
Am 08.05.2019 um 23:37 schrieb Andreas Dannenberg:
Quoted 10 lines
From: Vignesh R <[email protected]>
K3 devices have I2C IP that is same as OMAP2+ family. Allow driver to be compiled for
ARCH_K3.Signed-off-by Vignesh R <[email protected]>Signed-off-by Andreas Dannenberg <[email protected]>--- drivers/i2c/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
bye,
Heiko
Signature · 3 lines
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: +49-8142-66989-52 Fax: +49-8142-66989-80 Email: hs at denx.de
PATCH 2/18 arm: omap_i2c: Remove unwanted header file inclusion
In reply to Andreas DannenbergFrom: Vignesh R <[email protected]>
There is no need for to include this header here, so drop it.
--- arch/arm/include/asm/omap_i2c.h | 2 -- 1 file changed, 2 deletions(-)
arch/arm/include/asm/omap_i2c.h +0 -2
@@ -3,8 +3,6 @@#ifndef _OMAP_I2C_H#define _OMAP_I2C_H -#include <asm/arch/cpu.h> -#ifdef CONFIG_DM_I2C /* Information about a GPIO bank */
Signature · 1 line
2.17.1
PATCH 2/18 arm: omap_i2c: Remove unwanted header file inclusion
Hello Andreas,
Am 08.05.2019 um 23:37 schrieb Andreas Dannenberg:
Quoted 8 lines
From: Vignesh R <[email protected]>
There is no need for to include this header here, so drop it.
Signed-off-by Vignesh R <[email protected]>--- arch/arm/include/asm/omap_i2c.h | 2 -- 1 file changed, 2 deletions(-)
bye,
Heiko
Signature · 3 lines
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: +49-8142-66989-52 Fax: +49-8142-66989-80 Email: hs at denx.de
PATCH 3/18 arm: dts: k3-am65: Move pinctrl nodes out of U-Boot specific dtsi
In reply to Andreas DannenbergOnly U-Boot specifc DT properties or overrides, must be in -u-boot.dtsi. Pinctrl nodes does not belong here. Now that pinctrl nodes are in kernel DT, there is no reason to be keep these in -u-boot.dtsi. Move them to proper places so that it would ease copying DT entries from kernel DT.
--- arch/arm/dts/k3-am65-main.dtsi | 16 ++++++++++++++ arch/arm/dts/k3-am65-wakeup.dtsi | 8 +++++++ arch/arm/dts/k3-am654-base-board-u-boot.dtsi | 23 -------------------- 3 files changed, 24 insertions(+), 23 deletions(-)
arch/arm/dts/k3-am65-main.dtsi +16 -0
@@ -69,4 +69,20 @@ clock-frequency = <48000000>; current-speed = <115200>; }; + + main_pmx0: pinmux at 11c000 { + compatible = "pinctrl-single"; + reg = <0x0 0x11c000 0x0 0x2e4>; + #pinctrl-cells = <1>; + pinctrl-single,register-width = <32>; + pinctrl-single,function-mask = <0xffffffff>; + }; + + main_pmx1: pinmux at 11c2e8 { + compatible = "pinctrl-single"; + reg = <0x0 0x11c2e8 0x0 0x24>; + #pinctrl-cells = <1>; + pinctrl-single,register-width = <32>; + pinctrl-single,function-mask = <0xffffffff>; + }; };
arch/arm/dts/k3-am65-wakeup.dtsi +8 -0
@@ -34,6 +34,14 @@ }; }; + wkup_pmx0: pinmux at 4301c000 { + compatible = "pinctrl-single"; + reg = <0x4301c000 0x118>; + #pinctrl-cells = <1>; + pinctrl-single,register-width = <32>; + pinctrl-single,function-mask = <0xffffffff>; + }; + wkup_uart0: serial at 42300000 { compatible = "ti,am654-uart"; reg = <0x42300000 0x100>;
arch/arm/dts/k3-am654-base-board-u-boot.dtsi +0 -23
@@ -19,22 +19,6 @@ &cbass_main{ u-boot,dm-spl; - main_pmx0: pinmux at 11c000 { - compatible = "pinctrl-single"; - reg = <0x0 0x11c000 0x0 0x2e4>; - #pinctrl-cells = <1>; - pinctrl-single,register-width = <32>; - pinctrl-single,function-mask = <0xffffffff>; - }; - - main_pmx1: pinmux at 11c2e8 { - compatible = "pinctrl-single"; - reg = <0x0 0x11c2e8 0x0 0x24>; - #pinctrl-cells = <1>; - pinctrl-single,register-width = <32>; - pinctrl-single,function-mask = <0xffffffff>; - }; - sdhci0: sdhci at 04F80000 { compatible = "arasan,sdhci-5.1"; reg = <0x0 0x4F80000 0x0 0x1000>,
@@ -57,13 +41,6 @@ &cbass_mcu { u-boot,dm-spl; - wkup_pmx0: pinmux at 4301c000 { - compatible = "pinctrl-single"; - reg = <0x0 0x4301c000 0x0 0x118>; - #pinctrl-cells = <1>; - pinctrl-single,register-width = <32>; - pinctrl-single,function-mask = <0xffffffff>; - }; navss_mcu: navss-mcu { compatible = "simple-bus";
Signature · 1 line
2.17.1
PATCH 4/18 arm: dts: k3-am65: Add I2C nodes
In reply to Andreas DannenbergAdd I2C DT nodes
--- arch/arm/dts/k3-am65-main.dtsi | 44 ++++++++++++++++++++++++++++++++ arch/arm/dts/k3-am65-mcu.dtsi | 11 ++++++++ arch/arm/dts/k3-am65-wakeup.dtsi | 11 ++++++++ arch/arm/dts/k3-am65.dtsi | 6 +++++ 4 files changed, 72 insertions(+)
arch/arm/dts/k3-am65-main.dtsi +44 -0
@@ -85,4 +85,48 @@ pinctrl-single,register-width = <32>; pinctrl-single,function-mask = <0xffffffff>; }; + + main_i2c0: i2c at 2000000 { + compatible = "ti,am654-i2c", "ti,omap4-i2c"; + reg = <0x0 0x2000000 0x0 0x100>; + interrupts = <GIC_SPI 200 IRQ_TYPE_LEVEL_HIGH>; + #address-cells = <1>; + #size-cells = <0>; + clock-names = "fck"; + clocks = <&k3_clks 110 1>; + power-domains = <&k3_pds 110>; + }; + + main_i2c1: i2c at 2010000 { + compatible = "ti,am654-i2c", "ti,omap4-i2c"; + reg = <0x0 0x2010000 0x0 0x100>; + interrupts = <GIC_SPI 201 IRQ_TYPE_LEVEL_HIGH>; + #address-cells = <1>; + #size-cells = <0>; + clock-names = "fck"; + clocks = <&k3_clks 111 1>; + power-domains = <&k3_pds 111>; + }; + + main_i2c2: i2c at 2020000 { + compatible = "ti,am654-i2c", "ti,omap4-i2c"; + reg = <0x0 0x2020000 0x0 0x100>; + interrupts = <GIC_SPI 202 IRQ_TYPE_LEVEL_HIGH>; + #address-cells = <1>; + #size-cells = <0>; + clock-names = "fck"; + clocks = <&k3_clks 112 1>; + power-domains = <&k3_pds 112>; + }; + + main_i2c3: i2c at 2030000 { + compatible = "ti,am654-i2c", "ti,omap4-i2c"; + reg = <0x0 0x2030000 0x0 0x100>; + interrupts = <GIC_SPI 203 IRQ_TYPE_LEVEL_HIGH>; + #address-cells = <1>; + #size-cells = <0>; + clock-names = "fck"; + clocks = <&k3_clks 113 1>; + power-domains = <&k3_pds 113>; + }; };
arch/arm/dts/k3-am65-mcu.dtsi +11 -0
@@ -15,4 +15,15 @@ clock-frequency = <96000000>; current-speed = <115200>; }; + + mcu_i2c0: i2c at 40b00000 { + compatible = "ti,am654-i2c", "ti,omap4-i2c"; + reg = <0x0 0x40b00000 0x0 0x100>; + interrupts = <GIC_SPI 564 IRQ_TYPE_LEVEL_HIGH>; + #address-cells = <1>; + #size-cells = <0>; + clock-names = "fck"; + clocks = <&k3_clks 114 1>; + power-domains = <&k3_pds 114>; + }; };
arch/arm/dts/k3-am65-wakeup.dtsi +11 -0
@@ -51,4 +51,15 @@ clock-frequency = <48000000>; current-speed = <115200>; }; + + wkup_i2c0: i2c at 42120000 { + compatible = "ti,am654-i2c", "ti,omap4-i2c"; + reg = <0x42120000 0x100>; + interrupts = <GIC_SPI 696 IRQ_TYPE_LEVEL_HIGH>; + #address-cells = <1>; + #size-cells = <0>; + clock-names = "fck"; + clocks = <&k3_clks 115 1>; + power-domains = <&k3_pds 115>; + }; };
arch/arm/dts/k3-am65.dtsi +6 -0
@@ -22,6 +22,12 @@ serial2 = &main_uart0; serial3 = &main_uart1; serial4 = &main_uart2; + i2c0 = &wkup_i2c0; + i2c1 = &mcu_i2c0; + i2c2 = &main_i2c0; + i2c3 = &main_i2c1; + i2c4 = &main_i2c2; + i2c5 = &main_i2c3; }; chosen { };
Signature · 1 line
2.17.1
PATCH 5/18 arm: dts: k3-am654-base-board: Enable wkup_i2c0 across all boot stages
In reply to Andreas DannenbergTo enable the use of an EEPROM-based board detection scheme we need to be able to access the I2C bus associated with the EEPROMs across all 3 stages of U-Boot: R5 SPL, A53 SPL, and A53 U-Boot (proper). So go ahead and add/update the wkup_i2c0 peripheral module DTS definitions and its associated pinmux node accordingly.
--- arch/arm/dts/k3-am65.dtsi | 1 + arch/arm/dts/k3-am654-base-board-u-boot.dtsi | 12 ++++++++++++ arch/arm/dts/k3-am654-base-board.dts | 15 +++++++++++++++ arch/arm/dts/k3-am654-r5-base-board.dts | 13 +++++++++++++ 4 files changed, 41 insertions(+)
arch/arm/dts/k3-am65.dtsi +1 -0
@@ -8,6 +8,7 @@ #include <dt-bindings/gpio/gpio.h> #include <dt-bindings/interrupt-controller/irq.h> #include <dt-bindings/interrupt-controller/arm-gic.h> +#include <dt-bindings/pinctrl/k3.h> / { model = "Texas Instruments K3 AM654 SoC";
arch/arm/dts/k3-am654-base-board-u-boot.dtsi +12 -0
@@ -117,6 +117,14 @@ u-boot,dm-spl; }; +&wkup_pmx0 { + u-boot,dm-spl; + + wkup_i2c0_pins_default { + u-boot,dm-spl; + }; +}; + &main_pmx0 { u-boot,dm-spl; main_uart0_pins_default: main_uart0_pins_default {
@@ -189,3 +197,7 @@ pinctrl-0 = <&main_mmc1_pins_default>; sdhci-caps-mask = <0x7 0x0>; }; + +&wkup_i2c0 { + u-boot,dm-spl; +};
arch/arm/dts/k3-am654-base-board.dts +15 -0
@@ -34,3 +34,18 @@ }; }; }; + +&wkup_pmx0 { + wkup_i2c0_pins_default: wkup-i2c0-pins-default { + pinctrl-single,pins = < + AM65X_WKUP_IOPAD(0x00e0, PIN_INPUT, 0) /* (AC7) WKUP_I2C0_SCL */ + AM65X_WKUP_IOPAD(0x00e4, PIN_INPUT, 0) /* (AD6) WKUP_I2C0_SDA */ + >; + }; +}; + +&wkup_i2c0 { + pinctrl-names = "default"; + pinctrl-0 = <&wkup_i2c0_pins_default>; + clock-frequency = <400000>; +};
arch/arm/dts/k3-am654-r5-base-board.dts +13 -0
@@ -130,6 +130,13 @@ >; u-boot,dm-spl; }; + + wkup_i2c0_pins_default: wkup-i2c0-pins-default { + pinctrl-single,pins = < + AM65X_WKUP_IOPAD(0x00e0, PIN_INPUT, 0) /* (AC7) WKUP_I2C0_SCL */ + AM65X_WKUP_IOPAD(0x00e4, PIN_INPUT, 0) /* (AD6) WKUP_I2C0_SDA */ + >; + }; }; &memorycontroller {
@@ -137,3 +144,9 @@ pinctrl-names = "default"; pinctrl-0 = <&wkup_vtt_pins_default>; }; + +&wkup_i2c0 { + pinctrl-names = "default"; + pinctrl-0 = <&wkup_i2c0_pins_default>; + clock-frequency = <400000>; +};
Signature · 1 line
2.17.1
PATCH 6/18 configs: am65x_evm_defconfig: Enable I2C support
In reply to Andreas DannenbergEnable I2C via driver model as well as the associated set of U-Boot commands to allow us interacting with various I2C slave devices.
--- configs/am65x_evm_a53_defconfig | 4 ++++ configs/am65x_evm_r5_defconfig | 4 ++++ 2 files changed, 8 insertions(+)
configs/am65x_evm_a53_defconfig +4 -0
@@ -30,6 +30,7 @@ CONFIG_SPL_REMOTEPROC=y CONFIG_SPL_YMODEM_SUPPORT=y CONFIG_CMD_ASKENV=y # CONFIG_CMD_FLASH is not set +CONFIG_CMD_I2C=y CONFIG_CMD_MMC=y CONFIG_CMD_REMOTEPROC=y # CONFIG_CMD_SETEXPR is not set
@@ -53,6 +54,9 @@ CONFIG_CLK_TI_SCI=y CONFIG_DMA_CHANNELS=y CONFIG_TI_K3_NAVSS_UDMA=y CONFIG_TI_SCI_PROTOCOL=y +CONFIG_DM_I2C=y +CONFIG_I2C_SET_DEFAULT_BUS_NUM=y +CONFIG_SYS_I2C_OMAP24XX=y CONFIG_DM_MAILBOX=y CONFIG_K3_SEC_PROXY=y CONFIG_DM_MMC=y
configs/am65x_evm_r5_defconfig +4 -0
@@ -35,6 +35,7 @@ CONFIG_CMD_BOOTZ=y CONFIG_CMD_ASKENV=y # CONFIG_CMD_FLASH is not set CONFIG_CMD_GPT=y +CONFIG_CMD_I2C=y CONFIG_CMD_MMC=y CONFIG_CMD_REMOTEPROC=y # CONFIG_CMD_SETEXPR is not set
@@ -58,6 +59,9 @@ CONFIG_CLK_TI_SCI=y CONFIG_TI_SCI_PROTOCOL=y CONFIG_DM_GPIO=y CONFIG_DA8XX_GPIO=y +CONFIG_DM_I2C=y +CONFIG_I2C_SET_DEFAULT_BUS_NUM=y +CONFIG_SYS_I2C_OMAP24XX=y CONFIG_DM_MAILBOX=y CONFIG_K3_SEC_PROXY=y CONFIG_MISC=y
Signature · 1 line
2.17.1
PATCH 7/18 configs: am65x_hs_evm_defconfig: Enable I2C support
In reply to Andreas DannenbergEnable I2C via driver model as well as the associated set of U-Boot commands to allow us interacting with various I2C slave devices.
--- configs/am65x_hs_evm_a53_defconfig | 4 ++++ configs/am65x_hs_evm_r5_defconfig | 4 ++++ 2 files changed, 8 insertions(+)
configs/am65x_hs_evm_a53_defconfig +4 -0
@@ -33,6 +33,7 @@ CONFIG_SPL_REMOTEPROC=y CONFIG_SPL_YMODEM_SUPPORT=y CONFIG_CMD_ASKENV=y # CONFIG_CMD_FLASH is not set +CONFIG_CMD_I2C=y CONFIG_CMD_MMC=y CONFIG_CMD_REMOTEPROC=y # CONFIG_CMD_SETEXPR is not set
@@ -56,6 +57,9 @@ CONFIG_CLK_TI_SCI=y CONFIG_DMA_CHANNELS=y CONFIG_TI_K3_NAVSS_UDMA=y CONFIG_TI_SCI_PROTOCOL=y +CONFIG_DM_I2C=y +CONFIG_I2C_SET_DEFAULT_BUS_NUM=y +CONFIG_SYS_I2C_OMAP24XX=y CONFIG_DM_MAILBOX=y CONFIG_K3_SEC_PROXY=y CONFIG_DM_MMC=y
configs/am65x_hs_evm_r5_defconfig +4 -0
@@ -37,6 +37,7 @@ CONFIG_CMD_BOOTZ=y CONFIG_CMD_ASKENV=y # CONFIG_CMD_FLASH is not set CONFIG_CMD_GPT=y +CONFIG_CMD_I2C=y CONFIG_CMD_MMC=y CONFIG_CMD_REMOTEPROC=y # CONFIG_CMD_SETEXPR is not set
@@ -60,6 +61,9 @@ CONFIG_CLK_TI_SCI=y CONFIG_TI_SCI_PROTOCOL=y CONFIG_DM_GPIO=y CONFIG_DA8XX_GPIO=y +CONFIG_DM_I2C=y +CONFIG_I2C_SET_DEFAULT_BUS_NUM=y +CONFIG_SYS_I2C_OMAP24XX=y CONFIG_DM_MAILBOX=y CONFIG_K3_SEC_PROXY=y CONFIG_MISC=y
Signature · 1 line
2.17.1
PATCH 7/18 configs: am65x_hs_evm_defconfig: Enable I2C support
Quoted 7 lines · Andreas Dannenberg
Enable I2C via driver model as well as the associated set of U-Boot commands to allow us interacting with various I2C slave devices.
Signed-off-by Andreas Dannenberg <[email protected]>--- configs/am65x_hs_evm_a53_defconfig | 4 ++++ configs/am65x_hs_evm_r5_defconfig | 4 ++++
Thanks for not forgetting HS :)
For all the _hs defconfig changes in this series:
Quoted 47 lines
2 files changed, 8 insertions(+)configs/am65x_hs_evm_a53_defconfig +4
-0@@ -33,6 +33,7 @@ CONFIG_SPL_REMOTEPROC=y CONFIG_SPL_YMODEM_SUPPORT=y CONFIG_CMD_ASKENV=y # CONFIG_CMD_FLASH is not set +CONFIG_CMD_I2C=y CONFIG_CMD_MMC=y CONFIG_CMD_REMOTEPROC=y # CONFIG_CMD_SETEXPR is not set@@ -56,6 +57,9 @@ CONFIG_CLK_TI_SCI=y CONFIG_DMA_CHANNELS=y CONFIG_TI_K3_NAVSS_UDMA=y CONFIG_TI_SCI_PROTOCOL=y +CONFIG_DM_I2C=y +CONFIG_I2C_SET_DEFAULT_BUS_NUM=y +CONFIG_SYS_I2C_OMAP24XX=y CONFIG_DM_MAILBOX=y CONFIG_K3_SEC_PROXY=y CONFIG_DM_MMC=yconfigs/am65x_hs_evm_r5_defconfig +4
-0@@ -37,6 +37,7 @@ CONFIG_CMD_BOOTZ=y CONFIG_CMD_ASKENV=y # CONFIG_CMD_FLASH is not set CONFIG_CMD_GPT=y +CONFIG_CMD_I2C=y CONFIG_CMD_MMC=y CONFIG_CMD_REMOTEPROC=y # CONFIG_CMD_SETEXPR is not set@@ -60,6 +61,9 @@ CONFIG_CLK_TI_SCI=y CONFIG_TI_SCI_PROTOCOL=y CONFIG_DM_GPIO=y CONFIG_DA8XX_GPIO=y +CONFIG_DM_I2C=y +CONFIG_I2C_SET_DEFAULT_BUS_NUM=y +CONFIG_SYS_I2C_OMAP24XX=y CONFIG_DM_MAILBOX=y CONFIG_K3_SEC_PROXY=y CONFIG_MISC=y
PATCH 8/18 arm64: dts: k3-am654-base-board: Add I2C GPIO expander @ 0x38
In reply to Andreas DannenbergThe AM654 base board has a TCA9554/PCA9554-type GPIO expander on the wkup_i2c0 bus at address 0x38 that is used to detect the presence of daughter cards. Add a respective DTS description of this expander to enable its use.
--- arch/arm/dts/k3-am654-base-board.dts | 7 +++++++ 1 file changed, 7 insertions(+)
arch/arm/dts/k3-am654-base-board.dts +7 -0
@@ -48,4 +48,11 @@ pinctrl-names = "default"; pinctrl-0 = <&wkup_i2c0_pins_default>; clock-frequency = <400000>; + + tca9554: gpio at 38 { + compatible = "nxp,pca9554"; + reg = <0x38>; + gpio-controller; + #gpio-cells = <2>; + }; };
Signature · 1 line
2.17.1
PATCH 9/18 configs: am65x_evm_a53: Enable PCA953X-type GPIO driver
In reply to Andreas DannenbergEnable GPIO driver for PCA953X-compatible I2C-based I/O expander which includes support for the TCA9554/PCA9554-type expander found on the AM654 EVM base board.
--- configs/am65x_evm_a53_defconfig | 2 ++ 1 file changed, 2 insertions(+)
configs/am65x_evm_a53_defconfig +2 -0
@@ -54,6 +54,8 @@ CONFIG_CLK_TI_SCI=y CONFIG_DMA_CHANNELS=y CONFIG_TI_K3_NAVSS_UDMA=y CONFIG_TI_SCI_PROTOCOL=y +CONFIG_DM_GPIO=y +CONFIG_DM_PCA953X=y CONFIG_DM_I2C=y CONFIG_I2C_SET_DEFAULT_BUS_NUM=y CONFIG_SYS_I2C_OMAP24XX=y
Signature · 1 line
2.17.1
PATCH 10/18 configs: am65x_hs_evm_a53: Enable PCA953X-type GPIO driver
In reply to Andreas DannenbergEnable GPIO driver for PCA953X-compatible I2C-based I/O expander which includes support for the TCA9554/PCA9554-type expander found on the AM654 EVM base board.
--- configs/am65x_hs_evm_a53_defconfig | 2 ++ 1 file changed, 2 insertions(+)
configs/am65x_hs_evm_a53_defconfig +2 -0
@@ -57,6 +57,8 @@ CONFIG_CLK_TI_SCI=y CONFIG_DMA_CHANNELS=y CONFIG_TI_K3_NAVSS_UDMA=y CONFIG_TI_SCI_PROTOCOL=y +CONFIG_DM_GPIO=y +CONFIG_DM_PCA953X=y CONFIG_DM_I2C=y CONFIG_I2C_SET_DEFAULT_BUS_NUM=y CONFIG_SYS_I2C_OMAP24XX=y
Signature · 1 line
2.17.1
PATCH 11/18 configs: am65x_evm_a53: Increase malloc pool before relocation
In reply to Andreas DannenbergThe malloc pool used before relocation is getting tight leading to out of memory errors doing certain DM-related calls. Since we are running the A53 SPL out of DDR let's just go ahead and bump its size as used in U-Boot proper as well as SPL (via Kconfig default value) from 8KB to 32KB.
--- configs/am65x_evm_a53_defconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
configs/am65x_evm_a53_defconfig +1 -1
@@ -2,7 +2,7 @@ CONFIG_ARM=y CONFIG_ARCH_K3=y CONFIG_SPL_LIBCOMMON_SUPPORT=y CONFIG_SPL_LIBGENERIC_SUPPORT=y -CONFIG_SYS_MALLOC_F_LEN=0x2000 +CONFIG_SYS_MALLOC_F_LEN=0x8000 CONFIG_SOC_K3_AM6=y CONFIG_TARGET_AM654_A53_EVM=y CONFIG_SPL_MMC_SUPPORT=y
Signature · 1 line
2.17.1
PATCH 12/18 configs: am65x_hs_evm_a53: Increase malloc pool before relocation
In reply to Andreas DannenbergThe malloc pool used before relocation is getting tight leading to out of memory errors doing certain DM-related calls. Since we are running the A53 SPL out of DDR let's just go ahead and bump its size as used in U-Boot proper as well as SPL (via Kconfig default value) from 8KB to 32KB.
--- configs/am65x_hs_evm_a53_defconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
configs/am65x_hs_evm_a53_defconfig +1 -1
@@ -3,7 +3,7 @@ CONFIG_ARCH_K3=y CONFIG_TI_SECURE_DEVICE=y CONFIG_SPL_LIBCOMMON_SUPPORT=y CONFIG_SPL_LIBGENERIC_SUPPORT=y -CONFIG_SYS_MALLOC_F_LEN=0x2000 +CONFIG_SYS_MALLOC_F_LEN=0x8000 CONFIG_SOC_K3_AM6=y CONFIG_TARGET_AM654_A53_EVM=y CONFIG_SPL_MMC_SUPPORT=y
Signature · 1 line
2.17.1
PATCH 13/18 arm: K3: am654: Map common EEPROM data into SRAM scratch space
In reply to Andreas DannenbergThe board detection scheme employed on various TI EVMs makes use of SRAM scratch space to share data read from an on-board EEPROM between the different bootloading stages. Map the associated definition that's used to locate this data into the SRAM scratch space we use on AM654x.
--- arch/arm/mach-k3/include/mach/am6_hardware.h | 3 +++ 1 file changed, 3 insertions(+)
arch/arm/mach-k3/include/mach/am6_hardware.h +3 -0
@@ -44,4 +44,7 @@#define CTRLMMR_LOCK_KICK1 0x0100c#define CTRLMMR_LOCK_KICK1_UNLOCK_VAL 0xd172bc5a +/* MCU SCRATCHPAD usage */ +#define TI_SRAM_SCRATCH_BOARD_EEPROM_START CONFIG_SYS_K3_MCU_SCRATCHPAD_BASE +#endif /* __ASM_ARCH_AM6_HARDWARE_H */
Signature · 1 line
2.17.1
PATCH 13/18 arm: K3: am654: Map common EEPROM data into SRAM scratch space
Quoted 15 lines · Andreas Dannenberg
The board detection scheme employed on various TI EVMs makes use of SRAM scratch space to share data read from an on-board EEPROM between the different bootloading stages. Map the associated definition that's used to locate this data into the SRAM scratch space we use on AM654x.
Signed-off-by Andreas Dannenberg <[email protected]>--- arch/arm/mach-k3/include/mach/am6_hardware.h | 3 +++ 1 file changed, 3 insertions(+)arch/arm/mach-k3/include/mach/am6_hardware.h +0
-0
arch/arm/mach-k3/include/mach/am6_hardware.h line 48
Quoted 5 lines
arch/arm/mach-k3/include/mach/am6_hardware.h +2
-0@@ -44,4 +44,7 @@#define CTRLMMR_LOCK_KICK1 0x0100c#define CTRLMMR_LOCK_KICK1_UNLOCK_VAL 0xd172bc5a +/* MCU SCRATCHPAD usage */ +#define TI_SRAM_SCRATCH_BOARD_EEPROM_START CONFIG_SYS_K3_MCU_SCRATCHPAD_BASE
Won't HS devices fail while accessing this region? We should drop it altogether.
Thanks and regards,
Lokesh
Quoted 3 lines
arch/arm/mach-k3/include/mach/am6_hardware.h +1
-0+#endif /* __ASM_ARCH_AM6_HARDWARE_H */
PATCH 13/18 arm: K3: am654: Map common EEPROM data into SRAM scratch space
Quoted 21 lines · Lokesh Vutla
Quoted 18 lines · Andreas Dannenberg
The board detection scheme employed on various TI EVMs makes use of SRAM scratch space to share data read from an on-board EEPROM between the different bootloading stages. Map the associated definition that's used to locate this data into the SRAM scratch space we use on AM654x.
Signed-off-by Andreas Dannenberg <[email protected]>--- arch/arm/mach-k3/include/mach/am6_hardware.h | 3 +++ 1 file changed, 3 insertions(+)arch/arm/mach-k3/include/mach/am6_hardware.h +0
-0@@ -44,4 +44,7 @@#define CTRLMMR_LOCK_KICK1 0x0100c#define CTRLMMR_LOCK_KICK1_UNLOCK_VAL 0xd172bc5a
Quoted 5 lines
arch/arm/mach-k3/include/mach/am6_hardware.h line 48
Quoted 2 lines
arch/arm/mach-k3/include/mach/am6_hardware.h +2
-0+/* MCU SCRATCHPAD usage */ +#define TI_SRAM_SCRATCH_BOARD_EEPROM_START CONFIG_SYS_K3_MCU_SCRATCHPAD_BASEWon't HS devices fail while accessing this region? We should drop it altogether.
HS devices cannot read this before SYSFW is loaded as by default it is left firewalled by ROM. This read happens after SYSFW loading so it does work currently, but no guarantee SYSFW will always remove this firewall by default, we may need a driver that does an explicit device get for this region to be sure it is powered and accessible (it is on a different reset domain, it may need special handling).
I think we should avoid using this scratchpad for a couple other reasons. After R5 SPL has finished bootloading is handled by A53 cores, the R5 will be repurposed and other software will run on it, possibly wiping out the memory here. Anything we want to pass form R5 to A53 should use a non-R5-local memory, not this scratchpad. We need the same for passing the original boot media info also.
A lot of pitfalls for just 512 bytes of RAM..
Andrew
Quoted 11 lines
Thanks and regards,
Lokesharch/arm/mach-k3/include/mach/am6_hardware.h line 50
Quoted 3 lines
arch/arm/mach-k3/include/mach/am6_hardware.h +1
-0+#endif /* __ASM_ARCH_AM6_HARDWARE_H */ _______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://lists.denx.de/listinfo/u-boot
PATCH 13/18 arm: K3: am654: Map common EEPROM data into SRAM scratch space
Andrew,
Quoted 28 lines · Andrew F. Davis
Quoted 21 lines · Lokesh Vutla
Quoted 18 lines · Andreas Dannenberg
The board detection scheme employed on various TI EVMs makes use of SRAM scratch space to share data read from an on-board EEPROM between the different bootloading stages. Map the associated definition that's used to locate this data into the SRAM scratch space we use on AM654x.
Signed-off-by Andreas Dannenberg <[email protected]>--- arch/arm/mach-k3/include/mach/am6_hardware.h | 3 +++ 1 file changed, 3 insertions(+)arch/arm/mach-k3/include/mach/am6_hardware.h +0
-0@@ -44,4 +44,7 @@#define CTRLMMR_LOCK_KICK1 0x0100c#define CTRLMMR_LOCK_KICK1_UNLOCK_VAL 0xd172bc5aQuoted 5 lines
arch/arm/mach-k3/include/mach/am6_hardware.h line 48
Quoted 2 lines
arch/arm/mach-k3/include/mach/am6_hardware.h +2
-0+/* MCU SCRATCHPAD usage */ +#define TI_SRAM_SCRATCH_BOARD_EEPROM_START CONFIG_SYS_K3_MCU_SCRATCHPAD_BASEWon't HS devices fail while accessing this region? We should drop it altogether.
Quoted 6 lines
HS devices cannot read this before SYSFW is loaded as by default it is left firewalled by ROM. This read happens after SYSFW loading so it does work currently, but no guarantee SYSFW will always remove this firewall by default, we may need a driver that does an explicit device get for this region to be sure it is powered and accessible (it is on a different reset domain, it may need special handling).
Yes this is understood. For others reading this, this is how it is done in our current production U-Boot tree (upstream U-Boot does not yet have SYSFW loading capability).
Quoted 8 lines
I think we should avoid using this scratchpad for a couple other reasons. After R5 SPL has finished bootloading is handled by A53 cores, the R5 will be repurposed and other software will run on it, possibly wiping out the memory here. Anything we want to pass form R5 to A53 should use a non-R5-local memory, not this scratchpad. We need the same for passing the original boot media info also.
A lot of pitfalls for just 512 bytes of RAM..
I don't disagree, this approach is limited, it's just the "easiest" we started using when creating the initial solution. Let's find/align on a different memory region.
While at it, there is a need to pass additional data across our three boot stages (R5 SPL -> A53 SPL -> A53 U-Boot); for example I'd at some point like to carry forward peripheral initialization state (so we don't have to re-probe stuff 3 times), amongst other things, for which I was eyeing to use the new CONFIG_BLOBLIST* feature. So if we can identify a new/dedicated memory region for such data it would be a good opportunity to also start making use of the BLOBLIST feature at the same time, creating a more scalable solution.
--
Andreas Dannenberg
Texas Instruments Inc
Quoted 13 lines
Andrew
Quoted 11 lines
Thanks and regards,
Lokesharch/arm/mach-k3/include/mach/am6_hardware.h line 50
Quoted 3 lines
arch/arm/mach-k3/include/mach/am6_hardware.h +1
-0+#endif /* __ASM_ARCH_AM6_HARDWARE_H */ _______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://lists.denx.de/listinfo/u-boot
PATCH 13/18 arm: K3: am654: Map common EEPROM data into SRAM scratch space
Tom et al.,
On Thu, May 09, 2019 at 11:27:25AM -0500, Andreas Dannenberg wrote:
<snip>
Quoted 21 lines
Quoted 9 lines
Quoted 2 lines
Won't HS devices fail while accessing this region? We should drop it altogether.
HS devices cannot read this before SYSFW is loaded as by default it is left firewalled by ROM. This read happens after SYSFW loading so it does work currently, but no guarantee SYSFW will always remove this firewall by default, we may need a driver that does an explicit device get for this region to be sure it is powered and accessible (it is on a different reset domain, it may need special handling).
Yes this is understood. For others reading this, this is how it is done in our current production U-Boot tree (upstream U-Boot does not yet have SYSFW loading capability).
Quoted 7 lines
I think we should avoid using this scratchpad for a couple other reasons. After R5 SPL has finished bootloading is handled by A53 cores, the R5 will be repurposed and other software will run on it, possibly wiping out the memory here. Anything we want to pass form R5 to A53 should use a non-R5-local memory, not this scratchpad. We need the same for passing the original boot media info also.
Quoted 5 lines
Quoted 1 line
A lot of pitfalls for just 512 bytes of RAM..
I don't disagree, this approach is limited, it's just the "easiest" we started using when creating the initial solution. Let's find/align on a different memory region.
After additional alignment internally with Lokesh and Andrew we agreed on that for the time being we would like to move forward with the solution proposed here as it relates to the use of the scratchpad SRAM. It works for all current use cases including "high security" type devices as it stands. We may re-visit and evolve the solution should a need arise when adding support for a future device (so far there is no such need on our immediate radar screen).
I also just rebased the solution on the latest upstream/master tree and confirmed that it still works as expected.
--
Andreas Dannenberg
Texas Instruments Inc
Quoted 8 lines
While at it, there is a need to pass additional data across our three boot stages (R5 SPL -> A53 SPL -> A53 U-Boot); for example I'd at some point like to carry forward peripheral initialization state (so we don't have to re-probe stuff 3 times), amongst other things, for which I was eyeing to use the new CONFIG_BLOBLIST* feature. So if we can identify a new/dedicated memory region for such data it would be a good opportunity to also start making use of the BLOBLIST feature at the same time, creating a more scalable solution.
PATCH 14/18 ti: common: am6: Add support for board description EEPROM
In reply to Andreas DannenbergThe AM654x EVM based on the TI K3 family of SoCs have an updated board detection EEPROM structure, now comprising variable-sized TLV-type records, containing a superset of what is already being provided on earlier platforms such as DRA7. Add basic support for parsing the new data structures contained on the base board into the common TI EEPROM structure while also providing infrastructure that can be used later on to parse data from additional EEPROMs such as the ones that are used on daughtercards for this platform.
--- board/ti/common/board_detect.c | 214 +++++++++++++++++++++++++++++++++ board/ti/common/board_detect.h | 194 ++++++++++++++++++++++++++++++ 2 files changed, 408 insertions(+)
board/ti/common/board_detect.c +214 -0
@@ -8,6 +8,7 @@*/ #include <common.h> +#include <asm/arch/hardware.h> #include <asm/omap_common.h> #include <dm/uclass.h> #include <i2c.h>
@@ -284,6 +285,191 @@ int __maybe_unused ti_i2c_eeprom_dra7_get(int bus_addr, int dev_addr)return 0 ; } +static int ti_i2c_eeprom_am6_parse_record (struct ti_am6_eeprom_record *record, +struct ti_am6_eeprom *ep, +char **mac_addr, +u8 mac_addr_max_cnt, +u8 *mac_addr_cnt) +{ +switch (record->header.id) { +case TI_AM6_EEPROM_RECORD_BOARD_INFO : +if (record->header.len !=sizeof (record->data.board_info)) +return -EINVAL ; + +if (!ep) +break ; + +/* Populate (and clean, if needed) the board name */ +strlcpy (ep->name, record->data.board_info.name, +sizeof (ep->name)); +ti_eeprom_string_cleanup (ep->name); + +/* Populate selected other fields from the board info record */ +strlcpy (ep->version, record->data.board_info.version, +sizeof (ep->version)); +strlcpy (ep->software_revision, + record->data.board_info.software_revision, +sizeof (ep->software_revision)); +strlcpy (ep->serial, record->data.board_info.serial, +sizeof (ep->serial)); +break ; +case TI_AM6_EEPROM_RECORD_MAC_INFO : +if (record->header.len !=sizeof (record->data.mac_info)) +return -EINVAL ; + +if (!mac_addr || !mac_addr_max_cnt) +break ; + + *mac_addr_cnt = ((record->data.mac_info.mac_control & +TI_AM6_EEPROM_MAC_ADDR_COUNT_MASK ) >> +TI_AM6_EEPROM_MAC_ADDR_COUNT_SHIFT ) +1 ; + +/* +* The EEPROM can (but may not) hold a very large amount +* of MAC addresses, by far exceeding what we want/can store +* in the common memory array, so only grab what we can fit. +* Note that a value of 0 means 1 MAC address, and so on. +*/ + *mac_addr_cnt =min (*mac_addr_cnt, mac_addr_max_cnt); + +memcpy (mac_addr, record->data.mac_info.mac_addr, + *mac_addr_cnt *TI_EEPROM_HDR_ETH_ALEN ); +break ; +case 0x00 : +/* Illegal value... Fall through... */ +case 0xFF : +/* Illegal value... Something went horribly wrong... */ +return -EINVAL ; +default : +pr_warn ("%s: Ignoring record id %u\n" , __func__, + record->header.id); + } + +return 0 ; +} + +int __maybe_unused ti_i2c_eeprom_am6_get (int bus_addr,int dev_addr, +struct ti_am6_eeprom *ep, +char **mac_addr, +u8 mac_addr_max_cnt, +u8 *mac_addr_cnt) +{ +struct udevice *dev; +struct udevice *bus; +unsigned int eeprom_addr; +struct ti_am6_eeprom_record_board_id board_id; +struct ti_am6_eeprom_record record; +int rc; + +/* Initialize with a known bad marker for i2c fails.. */ +memset (ep,0 ,sizeof (*ep)); + ep->header =TI_DEAD_EEPROM_MAGIC ; + +/* Read the board ID record which is always the first EEPROM record */ + rc =ti_i2c_eeprom_get (bus_addr, dev_addr,TI_EEPROM_HEADER_MAGIC , +sizeof (board_id), (uint8_t *)&board_id); +if (rc) +return rc; + +if (board_id.header.id !=TI_AM6_EEPROM_RECORD_BOARD_ID ) { +pr_err ("%s: Invalid board ID record!\n" , __func__); +return -EINVAL ; + } + +/* Establish DM handle to board config EEPROM */ + rc =uclass_get_device_by_seq (UCLASS_I2C , bus_addr, &bus); +if (rc) +return rc; + rc =i2c_get_chip (bus, dev_addr,1 , &dev); +if (rc) +return rc; + + ep->header =TI_EEPROM_HEADER_MAGIC ; + +/* Ready to parse TLV structure. Initialize variables... */ + *mac_addr_cnt =0 ; + +/* +* After the all-encompassing board ID record all other records follow +* a TLV-type scheme. Point to the first such record and then start +* parsing those one by one. +*/ + eeprom_addr =sizeof (board_id); + +while (true) { + rc =dm_i2c_read (dev, eeprom_addr, (uint8_t *)&record.header, +sizeof (record.header)); +if (rc) +return rc; + +/* +* Check for end of list marker. If we reached it don't go +* any further and stop parsing right here. +*/ +if (record.header.id ==TI_AM6_EEPROM_RECORD_END_LIST ) +break ; + + eeprom_addr +=sizeof (record.header); + +debug ("%s: dev_addr=0x%02x header.id=%u header.len=%u\n" , + __func__, dev_addr, record.header.id, + record.header.len); + +/* Read record into memory if it fits */ +if (record.header.len <=sizeof (record.data)) { + rc =dm_i2c_read (dev, eeprom_addr, + (uint8_t *)&record.data, + record.header.len); +if (rc) +return rc; + +/* Process record */ + rc =ti_i2c_eeprom_am6_parse_record (&record, ep, + mac_addr, + mac_addr_max_cnt, + mac_addr_cnt); +if (rc) { +pr_err ("%s: EEPROM parsing error!\n" , __func__); +return rc; + } + }else { +/* +* We may get here in case of larger records which +* are not yet understood. +*/ +pr_err ("%s: Ignoring record id %u\n" , __func__, + record.header.id); + } + + eeprom_addr += record.header.len; + } + +return 0 ; +} + +int __maybe_unused ti_i2c_eeprom_am6_get_base (int bus_addr,int dev_addr) +{ +struct ti_am6_eeprom *ep =TI_AM6_EEPROM_DATA ; +int ret; + +/* +* Always execute EEPROM read by not allowing to bypass it during the +* first invocation of SPL which happens on the R5 core. +*/ +#if !(defined(CONFIG_SPL_BUILD ) && defined(CONFIG_CPU_V7R )) +if (ep->header ==TI_EEPROM_HEADER_MAGIC ) { +debug ("%s: EEPROM has already been read\n" , __func__); +return 0 ; + } +#endif + + ret =ti_i2c_eeprom_am6_get (bus_addr, dev_addr, ep, + (char **)ep->mac_addr, +AM6_EEPROM_HDR_NO_OF_MAC_ADDR , + &ep->mac_addr_cnt); +return ret; +} +bool __maybe_unused board_ti_is (char *name_tag) {struct ti_common_eeprom *ep =TI_EEPROM_DATA ;
@@ -391,6 +577,34 @@ void __maybe_unused set_board_info_env(char *name)env_set ("board_serial" , unknown); } +void __maybe_unused set_board_info_env_am6 (char *name) +{ +char *unknown ="unknown" ; +struct ti_am6_eeprom *ep =TI_AM6_EEPROM_DATA ; + +if (name) +env_set ("board_name" , name); +else if (ep->name) +env_set ("board_name" , ep->name); +else +env_set ("board_name" , unknown); + +if (ep->version) +env_set ("board_rev" , ep->version); +else +env_set ("board_rev" , unknown); + +if (ep->software_revision) +env_set ("board_software_revision" , ep->software_revision); +else +env_set ("board_software_revision" , unknown); + +if (ep->serial) +env_set ("board_serial" , ep->serial); +else +env_set ("board_serial" , unknown); +} +static u64 mac_to_u64 (u8 mac[6 ]) {int i;
board/ti/common/board_detect.h +194 -0
218 lines not on this page · raw message
Signature · 1 line
2.17.1
PATCH 15/18 ti: common: am6: Add support for setting MAC addresses
In reply to Andreas DannenbergThe AM654x EVM based on the TI K3 family of SoCs has an updated board detection EEPROM structure that contains a TLV record of dedicated MAC addresses rather than a range of MAC addresses as it was used on earlier platforms such as DRA7. Add a basic function that allows us setting up Ethernet MAC addresses into the U-Boot environment based on the MAC address record contained in the common TI EEPROM structure.
--- board/ti/common/board_detect.c | 32 ++++++++++++++++++++++++++++++++ board/ti/common/board_detect.h | 12 ++++++++++++ 2 files changed, 44 insertions(+)
board/ti/common/board_detect.c +32 -0
@@ -534,6 +534,25 @@ fail:memset (mac_addr,0 ,TI_EEPROM_HDR_ETH_ALEN ); } +void __maybe_unused +board_ti_am6_get_eth_mac_addr (int index, +u8 mac_addr[TI_EEPROM_HDR_ETH_ALEN ]) +{ +struct ti_am6_eeprom *ep =TI_AM6_EEPROM_DATA ; + +if (ep->header ==TI_DEAD_EEPROM_MAGIC ) + goto fail; + +if (index <0 || index >= ep->mac_addr_cnt) + goto fail; + +memcpy (mac_addr, ep->mac_addr[index],TI_EEPROM_HDR_ETH_ALEN ); +return ; + +fail: +memset (mac_addr,0 ,TI_EEPROM_HDR_ETH_ALEN ); +} +u64 __maybe_unused board_ti_get_emif1_size (void ) {struct ti_common_eeprom *ep =TI_EEPROM_DATA ;
@@ -667,6 +686,19 @@ void board_ti_set_ethaddr(int index) } } +void board_ti_am6_set_ethaddr (int index,int count) +{ +u8 mac_addr[6 ]; +int i; + +for (i =0 ; i < count; i++) { +board_ti_am6_get_eth_mac_addr (i, mac_addr); +if (is_valid_ethaddr (mac_addr)) +eth_env_set_enetaddr_by_index ("eth" , i + index, + mac_addr); + } +} +bool __maybe_unused board_ti_was_eeprom_read (void ) {struct ti_common_eeprom *ep =TI_EEPROM_DATA ;
board/ti/common/board_detect.h +12 -0
@@ -398,6 +398,18 @@ void set_board_info_env_am6(char *name);*/ void board_ti_set_ethaddr (int index); +/** +* board_ti_am6_set_ethaddr- Sets the ethaddr environment from EEPROM +* @index: The first eth<index>addr environment variable to set +* @count: The number of MAC addresses to process +* +* EEPROM should be already read before calling this function. The EEPROM +* contains n dedicated MAC addresses. This function sets the ethaddr +* environment variable for all the available MAC addresses starting +* from eth<index>addr. +*/ +void board_ti_am6_set_ethaddr (int index,int count); +/** * board_ti_was_eeprom_read() - Check to see if the eeprom contents have been read *
Signature · 1 line
2.17.1
PATCH 16/18 board: ti: am654: Use EEPROM-based board detection
In reply to Andreas DannenbergThe TI AM654x EVM base board and the associated daughtercards have on- board I2C-based EEPROMs containing board configuration data. Use the board detection infrastructure introduced earlier to do the following:
1) Parse the AM654x EVM base board EEPROM and populate items like board name and MAC addresses into the TI common EEPROM data structure residing in SRAM scratch space
2) Check for presence of daughter card(s) by probing the associated presence signals via an I2C-based GPIO expander. Then, if such a card is found, parse the data such as additional Ethernet MAC addresses from its on-board EEPROM and populate into U-Boot accordingly
3) Dynamically create an U-Boot ENV variable called overlay_files containing a list of daugherboard-specific DTB overlays based on daughercards found.
This patch adds support for the AM654x base board ("AM6-COMPROCEVM") as well as for the IDK ("AM6-IDKAPPEVM"), OLDI LCD ("OLDI-LCD1EVM") PCIe/USB3.0 ("SER-PCIEUSBEVM"), 2 Lane PCIe/USB2.0 ("SER-PCIE2LEVM"), and general purpuse ("AM6-GPAPPEVM") daughtercards.
--- arch/arm/mach-k3/am6_init.c | 4 + arch/arm/mach-k3/include/mach/sys_proto.h | 2 + board/ti/am65x/Kconfig | 7 + board/ti/am65x/evm.c | 241 ++++++++++++++++++++++ include/configs/am65x_evm.h | 9 +- 5 files changed, 257 insertions(+), 6 deletions(-)
arch/arm/mach-k3/am6_init.c +4 -0
@@ -10,6 +10,7 @@#include <asm/io.h> #include <spl.h> #include <asm/arch/hardware.h> +#include <asm/arch/sys_proto.h> #include "common.h" #include <dm.h>
@@ -86,6 +87,9 @@ void board_init_f(ulong dummy)/* Prepare console output */ preloader_console_init (); +/* Perform EEPROM-based board detection */ +do_board_detect (); +#ifdef CONFIG_K3_AM654_DDRSS ret =uclass_get_device (UCLASS_RAM ,0 , &dev);if (ret)
arch/arm/mach-k3/include/mach/sys_proto.h +2 -0
@@ -12,4 +12,6 @@ u32 wait_on_value(u32 read_bit_mask, u32 match_value, void *read_addr, u32 bound);struct ti_sci_handle *get_ti_sci_handle (void );int fdt_fixup_msmc_ram (void *blob,char *parent_path,char *node_name); +int do_board_detect (void ); +#endif
board/ti/am65x/Kconfig +7 -0
@@ -12,6 +12,8 @@ config TARGET_AM654_A53_EVM select ARM64 select SOC_K3_AM6 select SYS_DISABLE_DCACHE_OPS + select BOARD_LATE_INIT + imply TI_I2C_BOARD_DETECT config TARGET_AM654_R5_EVM bool "TI K3 based AM654 EVM running on R5"
@@ -20,6 +22,7 @@ config TARGET_AM654_R5_EVM select SOC_K3_AM6 select K3_AM654_DDRSS imply SYS_K3_SPL_ATF + imply TI_I2C_BOARD_DETECT endchoice
@@ -34,6 +37,8 @@ config SYS_VENDOR config SYS_CONFIG_NAME default "am65x_evm" +source "board/ti/common/Kconfig" + endif if TARGET_AM654_R5_EVM
@@ -50,4 +55,6 @@ config SYS_CONFIG_NAME config SPL_LDSCRIPT default "arch/arm/mach-omap2/u-boot-spl.lds" +source "board/ti/common/Kconfig" + endif
board/ti/am65x/evm.c +241 -0
254 lines not on this page · raw message
include/configs/am65x_evm.h +3 -6
@@ -37,12 +37,9 @@/* U-Boot general configuration */ #define EXTRA_ENV_AM65X_BOARD_SETTINGS \ "findfdt=" \ - "if test $board_name = am65x; then " \ - "setenv name_fdt k3-am654-base-board.dtb; " \ - "else if test $name_fdt = undefined; then " \ - "echo WARNING: Could not determine device tree to use;"\ - "fi; fi; " \ - "setenv fdtfile ${name_fdt}\0" \ + "setenv name_fdt k3-am654-base-board.dtb;" \ + "setenv fdtfile ${name_fdt};" \ + "setenv overlay_files ${name_overlays}\0" \ "loadaddr=0x80080000\0" \ "fdtaddr=0x82000000\0" \ "name_kern=Image\0" \
Signature · 1 line
2.17.1
PATCH 17/18 configs: am65x_evm: Add support for applying overlays
In reply to Andreas DannenbergThis will allow for downloading and applying overlays from an MMC/SD boot media based on the overlay_files ENV variable containing a list of overlay files.
--- configs/am65x_evm_a53_defconfig | 3 ++- include/configs/am65x_evm.h | 9 +++++++++ 2 files changed, 11 insertions(+), 1 deletion(-)
configs/am65x_evm_a53_defconfig +2 -1
@@ -16,7 +16,7 @@ CONFIG_DISTRO_DEFAULTS=y # CONFIG_SYS_MALLOC_CLEAR_ON_INIT is not set CONFIG_SPL_LOAD_FIT=y CONFIG_OF_BOARD_SETUP=y -CONFIG_BOOTCOMMAND="run findfdt; run envboot; run init_${boot}; run get_kern_${boot}; run get_fdt_${boot}; run run_kern" +CONFIG_BOOTCOMMAND="run findfdt; run envboot; run init_${boot}; run get_kern_${boot}; run get_fdt_${boot}; run get_overlay_${boot}; run run_kern" # CONFIG_DISPLAY_CPUINFO is not set CONFIG_SPL_TEXT_BASE=0x80080000 CONFIG_SPL_SYS_MALLOC_SIMPLE=y
@@ -79,3 +79,4 @@ CONFIG_DM_SERIAL=y CONFIG_SOC_TI=y CONFIG_SYSRESET=y CONFIG_SYSRESET_TI_SCI=y +CONFIG_OF_LIBFDT_OVERLAY=y
include/configs/am65x_evm.h +9 -0
@@ -42,6 +42,7 @@"setenv overlay_files ${name_overlays}\0" \"loadaddr=0x80080000\0" \"fdtaddr=0x82000000\0" \ +"overlayaddr=0x83000000\0" \"name_kern=Image\0" \"console=ttyS2,115200n8\0" \"args_all=setenv optargs earlycon=ns16550a,mmio32,0x02800000\0" \
@@ -56,6 +57,14 @@"rd_spec=-\0" \"init_mmc=run args_all args_mmc\0" \"get_fdt_mmc=load mmc ${bootpart} ${fdtaddr} ${bootdir}/${name_fdt}\0" \ +"get_overlay_mmc=" \ +"fdt address ${fdtaddr};" \ +"fdt resize 0x100000;" \ +"for overlay in $overlay_files;" \ +"do;" \ +"load mmc ${bootpart} ${overlayaddr} ${bootdir}/${overlay};" \ +"fdt apply ${overlayaddr};" \ +"done;\0" \"get_kern_mmc=load mmc ${bootpart} ${loadaddr} " \"${bootdir}/${name_kern}\0"
Signature · 1 line
2.17.1
PATCH 18/18 configs: am65x_hs_evm: Add support for applying overlays
In reply to Andreas DannenbergThis will allow for downloading and applying overlays from an MMC/SD boot media based on the overlay_files ENV variable containing a list of overlay files.
--- configs/am65x_hs_evm_a53_defconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
configs/am65x_hs_evm_a53_defconfig +2 -1
@@ -19,7 +19,7 @@ CONFIG_FIT_IMAGE_POST_PROCESS=y CONFIG_SPL_LOAD_FIT=y CONFIG_SPL_FIT_IMAGE_POST_PROCESS=y CONFIG_OF_BOARD_SETUP=y -CONFIG_BOOTCOMMAND="run findfdt; run envboot; run init_${boot}; run get_kern_${boot}; run get_fdt_${boot}; run run_kern" +CONFIG_BOOTCOMMAND="run findfdt; run envboot; run init_${boot}; run get_kern_${boot}; run get_fdt_${boot}; run get_overlay_${boot}; run run_kern" # CONFIG_DISPLAY_CPUINFO is not set CONFIG_SPL_TEXT_BASE=0x80080000 CONFIG_SPL_SYS_MALLOC_SIMPLE=y
@@ -82,3 +82,4 @@ CONFIG_DM_SERIAL=y CONFIG_SOC_TI=y CONFIG_SYSRESET=y CONFIG_SYSRESET_TI_SCI=y +CONFIG_OF_LIBFDT_OVERLAY=y
Signature · 1 line
2.17.1
cover Add EEPROM-based board detect support for TI
In reply to Andreas DannenbergQuoted 10 lines · Andreas Dannenberg
This series adds the pieces needed to fully configure the TI AM654x EVM and its associated (optional) daughtercards using a GPIO-based card presence detection scheme. It then reads out and parses any I2C EEPROM data present on any such boards for example board name, serial #, Ethernet MAC addresses, etc. and incorporates this data into the current U-Boot environment. One important aspect is the application of daughercard-specific DTB overlays right before the Kernel boot happens.
This series also adds basic I2C support for TI K3 family devices which is a prerequisite for the EEPROM access.
For the entire series:
Thanks and regards,
Lokesh
Quoted 51 lines
--
Andreas Dannenberg
Texas Instruments IncAndreas Dannenberg (16):
arm: dts: k3-am65: Move pinctrl nodes out of U-Boot specific dtsi
arm: dts: k3-am65: Add I2C nodes
arm: dts: k3-am654-base-board: Enablewkup_i2c0across all boot stages
configs:am65x_evm_defconfig: Enable I2C support
configs:am65x_hs_evm_defconfig: Enable I2C support
arm64: dts: k3-am654-base-board: Add I2C GPIO expander @0x38
configs:am65x_evm_a53: Enable PCA953X-type GPIO driver
configs:am65x_hs_evm_a53: Enable PCA953X-type GPIO driver
configs:am65x_evm_a53: Increase malloc pool before relocation
configs:am65x_hs_evm_a53: Increase malloc pool before relocation
arm: K3: am654: Map common EEPROM data into SRAM scratch space
ti: common: am6: Add support for board description EEPROM
ti: common: am6: Add support for setting MAC addresses
board: ti: am654: Use EEPROM-based board detection
configs:am65x_evm: Add support for applying overlays
configs:am65x_hs_evm: Add support for applying overlaysVignesh R (2):
i2c:omap24xx_i2c: Adapt driver to support K3 devices
arm:omap_i2c: Remove unwanted header file inclusionarch/arm/dts/k3-am65-main.dtsi | 60 +++++ arch/arm/dts/k3-am65-mcu.dtsi | 11 + arch/arm/dts/k3-am65-wakeup.dtsi | 19 ++ arch/arm/dts/k3-am65.dtsi | 7 + arch/arm/dts/k3-am654-base-board-u-boot.dtsi | 35 +-- arch/arm/dts/k3-am654-base-board.dts | 22 ++ arch/arm/dts/k3-am654-r5-base-board.dts | 13 + arch/arm/include/asm/omap_i2c.h | 2 - arch/arm/mach-k3/am6_init.c | 4 + arch/arm/mach-k3/include/mach/am6_hardware.h | 3 + arch/arm/mach-k3/include/mach/sys_proto.h | 2 + board/ti/am65x/Kconfig | 7 + board/ti/am65x/evm.c | 241 ++++++++++++++++++ board/ti/common/board_detect.c | 246 +++++++++++++++++++ board/ti/common/board_detect.h | 206 ++++++++++++++++ configs/am65x_evm_a53_defconfig | 11 +- configs/am65x_evm_r5_defconfig | 4 + configs/am65x_hs_evm_a53_defconfig | 11 +- configs/am65x_hs_evm_r5_defconfig | 4 + drivers/i2c/Kconfig | 2 +- include/configs/am65x_evm.h | 18 +- 21 files changed, 892 insertions(+), 36 deletions(-)