Retrieve MAC address from EEPROM

Add the CONFIG_PHY_GIGE option as a hidden Kconfig symbol so that we can select it from the menu as a dependency.

Signed-off-by Olliver Schinagl <[email protected]>

Add the CONFIG_PHY_GIGE option as a hidden Kconfig symbol so that we can select it from the menu as a dependency.

Signed-off-by Olliver Schinagl <[email protected]>

We currently have a few dependencies explicitly set in the sunxi defconfigs. Things such as, RGMII, DESIGNWARE_ETH in combination with SUN7I_GMAC. When selecting SUN7I_GMAC we already imply DESIGNWARE_ETH for example.

This patch puts this logic into the Kconfig thus simplifying the defconfigs. For a user it is also no more logical when enabling one of the drivers in Kconfig to have the proper dependencies automatically selected.

One thing to note, the sun7i driver can be connected in two modes, RGMII and MII mode and we use both throughout the boards. To make this easy we split up the CONFIG_SUNXI_GMAC symbol into the SUN7I_GMAC and the SUN7I_EMAC symbol, where the SUN7I_EMAC indicates a MII connected designware IP.

We currently have a few dependencies explicitly set in the sunxi defconfigs. Things such as, RGMII, DESIGNWARE_ETH in combination with SUN7I_GMAC. When selecting SUN7I_GMAC we already imply DESIGNWARE_ETH for example.

This patch puts this logic into the Kconfig thus simplifying the defconfigs. For a user it is also no more logical when enabling one of the drivers in Kconfig to have the proper dependencies automatically selected.

One thing to note, the sun7i driver can be connected in two modes, RGMII and MII mode and we use both throughout the boards. To make this easy we split up the CONFIG_SUNXI_GMAC symbol into the SUN7I_GMAC and the SUN7I_EMAC symbol, where the SUN7I_EMAC indicates a MII connected designware IP.

Signed-off-by Olliver Schinagl <[email protected]>

---
 configs/A20-OLinuXino-Lime2-eMMC_defconfig |  2 --
 configs/A20-OLinuXino-Lime2_defconfig      |  2 --
 configs/A20-OLinuXino-Lime_defconfig       |  3 +-
 configs/A20-OLinuXino_MICRO_defconfig      |  3 +-
 configs/A20-Olimex-SOM-EVB_defconfig       |  2 --
 configs/Bananapi_defconfig                 |  2 --
 configs/Bananapro_defconfig                |  2 --
 configs/CSQ_CS908_defconfig                |  3 +-
 configs/Colombus_defconfig                 |  2 --
 configs/Cubieboard2_defconfig              |  3 +-
 configs/Cubietruck_defconfig               |  2 --
 configs/Hummingbird_A31_defconfig          |  2 --
 configs/Itead_Ibox_A20_defconfig           |  3 +-
 configs/Lamobo_R1_defconfig                |  2 --
 configs/Linksprite_pcDuino3_Nano_defconfig |  2 --
 configs/Linksprite_pcDuino3_defconfig      |  3 +-
 configs/Mele_A1000G_quad_defconfig         |  3 +-
 configs/Mele_I7_defconfig                  |  3 +-
 configs/Mele_M3_defconfig                  |  3 +-
 configs/Mele_M5_defconfig                  |  3 +-
 configs/Mele_M9_defconfig                  |  3 +-
 configs/Orangepi_defconfig                 |  2 --
 configs/Orangepi_mini_defconfig            |  2 --
 configs/Sinlinx_SinA31s_defconfig          |  3 +-
 configs/Sinovoip_BPI_M2_defconfig          |  2 --
 configs/Wits_Pro_A20_DKT_defconfig         |  2 --
 configs/i12-tvbox_defconfig                |  3 +-
 configs/icnova-a20-swac_defconfig          |  3 +-
 configs/mixtile_loftq_defconfig            |  2 --
 drivers/net/Kconfig                        | 54 ++++++++++++++++++++++++------
 include/configs/sunxi-common.h             |  3 --
 31 files changed, 57 insertions(+), 72 deletions(-)

drivers/net/Kconfig +24 -7

 config PHY_GIGE
        bool

@@ -170,17 +170,49 @@ config RTL8169
          This driver supports Realtek 8169 series gigabit ethernet family of
          PCI/PCIe chipsets/adapters.
 
-config SUN7I_GMAC
-       bool "Enable Allwinner GMAC Ethernet support"
-       select PHY_GIGE
-       help
-         Enable the support for Sun7i GMAC Ethernet controller
-
 config SUN4I_EMAC
        bool "Allwinner Sun4i Ethernet MAC support"
        depends on DM_ETH
+       select MII
+       select PHYLIB
+       help
+         This driver provides the Allwinner based SoCs with 100 Megabit
+         network support as it is found on the sun4i and sun7i. This driver
+         is known to have performance issues and should only be used on
+         sun4i hardware. Newer hardware supports the SUN7I_EMAC driver,
+         which also works with 100 Megabit PHY's.
+
+choice
+       prompt "Allwinner Sun7i GMAC support"
        help
-         This driver supports the Allwinner based SUN4I Ethernet MAC.
+         This driver provides the Allwinner based SoCs network support based
+         on the Synopsys Designware gigabit driver. The driver supports both
+         RGMII and MII communications and can thus be used as a drop in
+         replacement of the SUN4I_EMAC driver.
+
+config SUN7I_NONE
+       bool "Disable"
+
+config SUN7I_EMAC
+       bool "MII"
+       select ETH_DESIGNWARE
+       select PHYLIB
+       select MII

drivers/net/Kconfig +5 -0

+       help
+         If connecting the sun7i MAC to a gigabit ethernet PHY, select
+         this option.
+
+endchoice
 
 config SUN8I_EMAC
         bool "Allwinner Sun8i Ethernet MAC support"

include/configs/sunxi-common.h +0 -4

index 3d8a130f38..8f97b80619 100644
@@ -289,13 +289,10 @@ extern int soft_i2c_gpio_scl;
 /* Ethernet support */
 #ifdef CONFIG_SUNXI_EMAC
 #define CONFIG_PHY_ADDR                1
-#define CONFIG_MII                     /* MII PHY management           */
-#define CONFIG_PHYLIB
 #endif
 
 #ifdef CONFIG_SUNXI_GMAC
 #define CONFIG_PHY_ADDR                1
-#define CONFIG_MII                     /* MII PHY management           */
 #define CONFIG_PHY_REALTEK
 #endif
 
--

2.11.0

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://lists.denx.de/listinfo/u-boot

Commit 8728c97eff5bd9 (" configs: Re-sync") potentially broke the lime2 as it removed the RTL8211X_PHY_FORCE_MASTER flag.

Re-add this flag.

Signed-off-by Olliver Schinagl <[email protected]>

Commit 4d43d065db326 ("sunxi: Move SUNXI_GMAC to Kconfig") renamed SUNXI_[EG]MAC but did not update include/configs/sunxi-common.h where based on these two symbols other config symbols where being set, such as CONFIG_PHY_REALTEK for SUNXI_GMAC boards and the CONFIG_PHY_ADDR being set to a safe default.

This patch restores that behavior by adding the PHY_REALTEK to the defconfigs, where they belong and by setting the address based on the new config symbols.

Additionally, we use the new renamed symbol in the Makefiles to actually compile the drivers.

Signed-off-by Olliver Schinagl <[email protected]>

---
 arch/arm/include/asm/arch-sunxi/sys_proto.h | 2 +-
 board/sunxi/Makefile                        | 2 +-
 configs/A20-OLinuXino-Lime2_defconfig       | 1 +
 configs/A20-OLinuXino-Lime_defconfig        | 1 +
 configs/A20-OLinuXino_MICRO_defconfig       | 1 +
 configs/A20-Olimex-SOM-EVB_defconfig        | 1 +
 configs/Bananapi_defconfig                  | 1 +
 configs/Bananapro_defconfig                 | 1 +
 configs/CSQ_CS908_defconfig                 | 1 +
 configs/Colombus_defconfig                  | 1 +
 configs/Cubieboard2_defconfig               | 1 +
 configs/Cubietruck_defconfig                | 1 +
 configs/Hummingbird_A31_defconfig           | 1 +
 configs/Itead_Ibox_A20_defconfig            | 1 +
 configs/Lamobo_R1_defconfig                 | 1 +
 configs/Linksprite_pcDuino3_Nano_defconfig  | 1 +
 configs/Linksprite_pcDuino3_defconfig       | 1 +
 configs/Mele_A1000G_quad_defconfig          | 1 +
 configs/Mele_I7_defconfig                   | 1 +
 configs/Mele_M3_defconfig                   | 1 +
 configs/Mele_M5_defconfig                   | 1 +
 configs/Mele_M9_defconfig                   | 1 +
 configs/Orangepi_defconfig                  | 1 +
 configs/Orangepi_mini_defconfig             | 1 +
 configs/Sinlinx_SinA31s_defconfig           | 1 +
 configs/Sinovoip_BPI_M2_defconfig           | 1 +
 configs/Wits_Pro_A20_DKT_defconfig          | 1 +
 configs/i12-tvbox_defconfig                 | 1 +
 configs/icnova-a20-swac_defconfig           | 1 +
 configs/mixtile_loftq_defconfig             | 1 +
 drivers/net/Kconfig                         | 5 +++++
 drivers/net/Makefile                        | 2 +-
 include/configs/sunxi-common.h              | 9 +++------
 scripts/config_whitelist.txt                | 2 --
 34 files changed, 39 insertions(+), 11 deletions(-)

arch/arm/include/asm/arch-sunxi/sys_proto.h +1 -1

index a373319a2b..850236ed42 100644
@@ -24,7 +24,7 @@ void sdelay(unsigned long);
 void return_to_fel(uint32_t lr, uint32_t sp);
 
 /* Board / SoC level designware gmac init */
-#if !defined CONFIG_SPL_BUILD && defined CONFIG_SUNXI_GMAC
+#if !defined CONFIG_SPL_BUILD && defined CONFIG_SUN7I_MAC
 void eth_init_board(void);
 #else
 static inline void eth_init_board(void) {}

board/sunxi/Makefile +1 -1

index 43766e0ef4..b4768b9b9b 100644
@@ -9,7 +9,7 @@
 # SPDX-License-Identifier:     GPL-2.0+
 #
 obj-y  += board.o
-obj-$(CONFIG_SUNXI_GMAC)       += gmac.o
+obj-$(CONFIG_SUN7I_MAC)                += gmac.o
 obj-$(CONFIG_SUNXI_AHCI)       += ahci.o
 obj-$(CONFIG_MACH_SUN4I)       += dram_sun4i_auto.o
 obj-$(CONFIG_MACH_SUN5I)       += dram_sun5i_auto.o

configs/A20-OLinuXino-Lime2_defconfig +0 -0

index 14d1159ead..14a51ef3f9 100644

configs/A20-OLinuXino-Lime2_defconfig +0 -0

 CONFIG_RTL8211X_PHY_FORCE_MASTER=y
 CONFIG_AXP_ALDO3_VOLT=2800
 CONFIG_AXP_ALDO4_VOLT=2800

drivers/net/Kconfig +5 -0

index c0d141754f..04e8cf39c8 100644
@@ -182,6 +182,9 @@ config SUN4I_EMAC
          sun4i hardware. Newer hardware supports the SUN7I_EMAC driver,
          which also works with 100 Megabit PHY's.
 
+config SUN7I_MAC
+       bool
+
 choice
        prompt "Allwinner Sun7i GMAC support"
        help

@@ -196,6 +199,7 @@ config SUN7I_NONE
 config SUN7I_EMAC
        bool "MII"
        select ETH_DESIGNWARE
+       select SUN7I_MAC
        select PHYLIB
        select MII
        help

@@ -205,6 +209,7 @@ config SUN7I_EMAC
 config SUN7I_GMAC
        bool "RGMII"
        select ETH_DESIGNWARE
+       select SUN7I_MAC
        select PHYLIB
        select RGMII
        select PHY_GIGE

drivers/net/Makefile +1 -1

index aedb2cc90d..c5dd52bee2 100644
@@ -22,7 +22,7 @@ obj-$(CONFIG_DNET) += dnet.o
 obj-$(CONFIG_E1000) += e1000.o
 obj-$(CONFIG_E1000_SPI) += e1000_spi.o
 obj-$(CONFIG_EEPRO100) += eepro100.o
-obj-$(CONFIG_SUNXI_EMAC) += sunxi_emac.o
+obj-$(CONFIG_SUN4I_EMAC) += sunxi_emac.o
 obj-$(CONFIG_SUN8I_EMAC) += sun8i_emac.o
 obj-$(CONFIG_ENC28J60) += enc28j60.o
 obj-$(CONFIG_EP93XX) += ep93xx_eth.o

include/configs/sunxi-common.h +3 -6

index 8f97b80619..5d2c5b9040 100644
@@ -287,15 +287,12 @@ extern int soft_i2c_gpio_scl;
 #endif /* CONFIG_VIDEO */
 
 /* Ethernet support */
-#ifdef CONFIG_SUNXI_EMAC
+#if defined(CONFIG_SUN4I_EMAC) || \
+    defined(CONFIG_SUN7I_EMAC) || \
+    defined(CONFIG_SUN7I_GMAC)
 #define CONFIG_PHY_ADDR                1
 #endif
 
-#ifdef CONFIG_SUNXI_GMAC
-#define CONFIG_PHY_ADDR                1
-#define CONFIG_PHY_REALTEK
-#endif
-
 #ifdef CONFIG_USB_EHCI_HCD
 #define CONFIG_USB_OHCI_NEW
 #define CONFIG_USB_OHCI_SUNXI

scripts/config_whitelist.txt +0 -3

index cd0165f934..1e2bf24d5a 100644
@@ -2796,8 +2796,6 @@ CONFIG_STV0991_HZ
 CONFIG_STV0991_HZ_CLOCK
 CONFIG_ST_SMI
 CONFIG_SUNXI_AHCI
-CONFIG_SUNXI_EMAC
-CONFIG_SUNXI_GMAC
 CONFIG_SUNXI_GPIO
 CONFIG_SUNXI_MAX_FB_SIZE
 CONFIG_SUNXI_USB_PHYS
--

2.11.0

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://lists.denx.de/listinfo/u-boot

The GMAC_TX_DELAY symbol sets and is dependent on the SUN7I_GMAC driver. Move it from the generic ARCH section to the driver specific section.

Signed-off-by Olliver Schinagl <[email protected]>

Previously overlooked magic value in commit a40db6d51171 ("net: cosmetic: Do not use magic values for ARP_HLEN")

Signed-off-by Olliver Schinagl <[email protected]>

Previously overlooked magic value in commit a40db6d51171 ("net: cosmetic: Do not use magic values for ARP_HLEN")

Signed-off-by Olliver Schinagl <[email protected]>

In the current net stack, we have a few functions to get and set the "ethaddr" and "ethNaddr" environment variables, which use magic values to get and set these environment variables. Remove the magicness of the buffer by defining it proper and also check the input for its length.

Additionally use the define in fdt parser where the ethaddr variables are also used.

Signed-off-by Olliver Schinagl <[email protected]>

In the current net stack, we have a few functions to get and set the "ethaddr" and "ethNaddr" environment variables, which use magic values to get and set these environment variables. Remove the magicness of the buffer by defining it proper and also check the input for its length.

Additionally use the define in fdt parser where the ethaddr variables are also used.

Signed-off-by Olliver Schinagl <[email protected]>

In certain conditions we currently print the MAC address. For example a warning when a random mac address is in use or a missmatch between HW and ENV.

If all things went well however (but even if there is a miss-match) we do not inform the user what the final MAC address of the device is.

Lets print the final MAC address of the device with which it has been setup.

Signed-off-by Olliver Schinagl <[email protected]>

Universally administered and locally administered addresses are distinguished by setting the second-least-significant bit of the first octet of the address. Having a function to check and set this U/L bit from a function makes it nice for boards that want to generate their own mac address to ensure they are locally administered.

Unicast and multicast addresses are distinguised by setting the least-significant bit of the first octet of the address. Having a function to check and set this U/M bit from a function it nice to make a generated mac address a unicast address.

This patch introduces both these helper functions

Signed-off-by Olliver Schinagl <[email protected]>

Universally administered and locally administered addresses are distinguished by setting the second-least-significant bit of the first octet of the address. Having a function to check and set this U/L bit from a function makes it nice for boards that want to generate their own mac address to ensure they are locally administered.

Unicast and multicast addresses are distinguised by setting the least-significant bit of the first octet of the address. Having a function to check and set this U/M bit from a function it nice to make a generated mac address a unicast address.

This patch introduces both these helper functions

Signed-off-by Olliver Schinagl <[email protected]>

This patch allows Kconfig to enable and set parameters to make it possible to read the MAC address from an EEPROM. The net core layer then uses this information to read MAC addresses from this EEPROM.

Besides the various tuneables as to how to access the eeprom (bus, address, addressing mode/length, 2 configurable that are EEPROM generic (e.g. SPI or some other form of access) which are:

NET_ETHADDR_EEPROM_OFFSET, indicating where in the EEPROM the start of the MAC address is. The default is 8 allowing for 8 bytes before the MAC for other purposes (header MAGIC for example).

NET_ETHADDR_EEPROM_CRC8, indicating the MAC is appended with a CRC8-CCIT checksum that should be verified.

Currently only I2C eeproms have been tested and thus only those options are available, but shouldn't be a limit. NET_ETHADDR_EEPROM_SPI can be just as created and added.

The code currently first checks if there is a non-zero MAC address in the eeprom. If that fails to be the case, the read_rom_hwaddr can be used by a board to supply the MAC in other ways.

If both these fails, the other code is still in place to query the environent, which then can be used to override the hardware supplied data.

Signed-off-by Olliver Schinagl <[email protected]>

---
 arch/arm/mach-sunxi/Kconfig                |  8 +--
 configs/A10-OLinuXino-Lime_defconfig       |  1 -
 configs/A20-OLinuXino-Lime2-eMMC_defconfig |  1 -
 configs/A20-OLinuXino-Lime2_defconfig      |  1 -
 configs/A20-OLinuXino-Lime_defconfig       |  1 -
 configs/A20-OLinuXino_MICRO_defconfig      |  1 -
 doc/README.enetaddr                        | 95 ++++++++++++++++++++++++++++++
 include/net.h                              | 14 +++++
 net/Kconfig                                | 59 +++++++++++++++++++
 net/eth-uclass.c                           | 42 +++++++------
 net/eth_common.c                           | 37 ++++++++++++
 net/eth_legacy.c                           |  2 +
 12 files changed, 231 insertions(+), 31 deletions(-)

arch/arm/mach-sunxi/Kconfig +4 -4

index be060b65dc..cf5af4fe18 100644
@@ -479,14 +479,14 @@ config I2C0_ENABLE
 
 config I2C1_ENABLE
 	bool "Enable I2C/TWI controller 1"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 1)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.
 
 config I2C2_ENABLE
 	bool "Enable I2C/TWI controller 2"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 2)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.

@@ -494,7 +494,7 @@ config I2C2_ENABLE
 if MACH_SUN6I || MACH_SUN7I
 config I2C3_ENABLE
 	bool "Enable I2C/TWI controller 3"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 3)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.

@@ -513,7 +513,7 @@ endif
 if MACH_SUN7I
 config I2C4_ENABLE
 	bool "Enable I2C/TWI controller 4"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 4)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.

configs/A10-OLinuXino-Lime_defconfig +0 -0

index ec16a44fbd..d4513144eb 100644

Quoted 86 lines · Olliver Schinagl

This patch allows Kconfig to enable and set parameters to make it possible to read the MAC address from an EEPROM. The net core layer then uses this information to read MAC addresses from this EEPROM.

Besides the various tuneables as to how to access the eeprom (bus, address, addressing mode/length, 2 configurable that are EEPROM generic (e.g. SPI or some other form of access) which are:

NET_ETHADDR_EEPROM_OFFSET, indicating where in the EEPROM the start of the MAC address is. The default is 8 allowing for 8 bytes before the MAC for other purposes (header MAGIC for example).

NET_ETHADDR_EEPROM_CRC8, indicating the MAC is appended with a CRC8-CCIT checksum that should be verified.

Currently only I2C eeproms have been tested and thus only those options are available, but shouldn't be a limit. NET_ETHADDR_EEPROM_SPI can be just as created and added.

The code currently first checks if there is a non-zero MAC address in the eeprom. If that fails to be the case, the read_rom_hwaddr can be used by a board to supply the MAC in other ways.

If both these fails, the other code is still in place to query the environent, which then can be used to override the hardware supplied data.

Signed-off-by Olliver Schinagl <[email protected]>

---
 arch/arm/mach-sunxi/Kconfig                |  8 +--
 configs/A10-OLinuXino-Lime_defconfig       |  1 -
 configs/A20-OLinuXino-Lime2-eMMC_defconfig |  1 -
 configs/A20-OLinuXino-Lime2_defconfig      |  1 -
 configs/A20-OLinuXino-Lime_defconfig       |  1 -
 configs/A20-OLinuXino_MICRO_defconfig      |  1 -
 doc/README.enetaddr                        | 95 ++++++++++++++++++++++++++++++
 include/net.h                              | 14 +++++
 net/Kconfig                                | 59 +++++++++++++++++++
 net/eth-uclass.c                           | 42 +++++++------
 net/eth_common.c                           | 37 ++++++++++++
 net/eth_legacy.c                           |  2 +
 12 files changed, 231 insertions(+), 31 deletions(-)

arch/arm/mach-sunxi/Kconfig +4 -4

index be060b65dc..cf5af4fe18 100644
@@ -479,14 +479,14 @@ config I2C0_ENABLE
 
 config I2C1_ENABLE
 	bool "Enable I2C/TWI controller 1"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 1)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.
 
 config I2C2_ENABLE
 	bool "Enable I2C/TWI controller 2"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 2)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.

@@ -494,7 +494,7 @@ config I2C2_ENABLE
 if MACH_SUN6I || MACH_SUN7I
 config I2C3_ENABLE
 	bool "Enable I2C/TWI controller 3"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 3)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.

@@ -513,7 +513,7 @@ endif
 if MACH_SUN7I
 config I2C4_ENABLE
 	bool "Enable I2C/TWI controller 4"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 4)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.

configs/A10-OLinuXino-Lime_defconfig +0 -0

index ec16a44fbd..d4513144eb 100644

configs/A10-OLinuXino-Lime_defconfig CONFIG_DRAM_CLK=480 · line 8

Quoted 5 lines

configs/A10-OLinuXino-Lime_defconfig +0 -1

@@ -5,7 +5,6 @@ CONFIG_DRAM_CLK=480
 CONFIG_DRAM_EMR1=4
 CONFIG_SYS_CLK_FREQ=912000000
 CONFIG_MMC0_CD_PIN="PH1"
-CONFIG_I2C1_ENABLE=y

Didn't that just remove the option if NET_ETHADDR_EEPROM is disabled (which is its default) ?

Maxime

Hey Maxime,

Quoted 87 lines · Maxime Ripard

Quoted 86 lines · Olliver Schinagl

This patch allows Kconfig to enable and set parameters to make it possible to read the MAC address from an EEPROM. The net core layer then uses this information to read MAC addresses from this EEPROM.

Besides the various tuneables as to how to access the eeprom (bus, address, addressing mode/length, 2 configurable that are EEPROM generic (e.g. SPI or some other form of access) which are:

NET_ETHADDR_EEPROM_OFFSET, indicating where in the EEPROM the start of the MAC address is. The default is 8 allowing for 8 bytes before the MAC for other purposes (header MAGIC for example).

NET_ETHADDR_EEPROM_CRC8, indicating the MAC is appended with a CRC8-CCIT checksum that should be verified.

Currently only I2C eeproms have been tested and thus only those options are available, but shouldn't be a limit. NET_ETHADDR_EEPROM_SPI can be just as created and added.

The code currently first checks if there is a non-zero MAC address in the eeprom. If that fails to be the case, the read_rom_hwaddr can be used by a board to supply the MAC in other ways.

If both these fails, the other code is still in place to query the environent, which then can be used to override the hardware supplied data.

Signed-off-by Olliver Schinagl <[email protected]>

---
 arch/arm/mach-sunxi/Kconfig                |  8 +--
 configs/A10-OLinuXino-Lime_defconfig       |  1 -
 configs/A20-OLinuXino-Lime2-eMMC_defconfig |  1 -
 configs/A20-OLinuXino-Lime2_defconfig      |  1 -
 configs/A20-OLinuXino-Lime_defconfig       |  1 -
 configs/A20-OLinuXino_MICRO_defconfig      |  1 -
 doc/README.enetaddr                        | 95 ++++++++++++++++++++++++++++++
 include/net.h                              | 14 +++++
 net/Kconfig                                | 59 +++++++++++++++++++
 net/eth-uclass.c                           | 42 +++++++------
 net/eth_common.c                           | 37 ++++++++++++
 net/eth_legacy.c                           |  2 +
 12 files changed, 231 insertions(+), 31 deletions(-)

arch/arm/mach-sunxi/Kconfig +4 -4

index be060b65dc..cf5af4fe18 100644
@@ -479,14 +479,14 @@ config I2C0_ENABLE
 
 config I2C1_ENABLE
 	bool "Enable I2C/TWI controller 1"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 1)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.
 
 config I2C2_ENABLE
 	bool "Enable I2C/TWI controller 2"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 2)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.

@@ -494,7 +494,7 @@ config I2C2_ENABLE
 if MACH_SUN6I || MACH_SUN7I
 config I2C3_ENABLE
 	bool "Enable I2C/TWI controller 3"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 3)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.

@@ -513,7 +513,7 @@ endif
 if MACH_SUN7I
 config I2C4_ENABLE
 	bool "Enable I2C/TWI controller 4"
-	default n
+	default y if (NET_ETHADDR_EEPROM_I2C_BUS = 4)
 	select CMD_I2C
 	---help---
 	See I2C0_ENABLE help text.

configs/A10-OLinuXino-Lime_defconfig +0 -0

index ec16a44fbd..d4513144eb 100644

Quoted 8 lines

configs/A10-OLinuXino-Lime_defconfig CONFIG_DRAM_CLK=480 · line 8

Quoted 5 lines

configs/A10-OLinuXino-Lime_defconfig +0 -1

@@ -5,7 +5,6 @@ CONFIG_DRAM_CLK=480
 CONFIG_DRAM_EMR1=4
 CONFIG_SYS_CLK_FREQ=912000000
 CONFIG_MMC0_CD_PIN="PH1"
-CONFIG_I2C1_ENABLE=y

Didn't that just remove the option if NET_ETHADDR_EEPROM is disabled (which is its default) ?

I initially forced I2C1_ENABLE on all the Olimex boards before finding that I could do:

default y if (NET_ETHADDR_EEPROM_I2C_BUS = 1)

So by defining NET_ETHADDR_EEPROM_I2C_BUS = x in the defconfig, the correct I2C1_ENABLE is being set, so we do not need to hardcode this in the defconfig anymore.

Or am I miss-understanding you?

This patch allows Kconfig to enable and set parameters to make it possible to read the MAC address from an EEPROM. The net core layer then uses this information to read MAC addresses from this EEPROM.

Besides the various tuneables as to how to access the eeprom (bus, address, addressing mode/length, 2 configurable that are EEPROM generic (e.g. SPI or some other form of access) which are:

NET_ETHADDR_EEPROM_OFFSET, indicating where in the EEPROM the start of the MAC address is. The default is 8 allowing for 8 bytes before the MAC for other purposes (header MAGIC for example).

NET_ETHADDR_EEPROM_CRC8, indicating the MAC is appended with a CRC8-CCIT checksum that should be verified.

Currently only I2C eeproms have been tested and thus only those options are available, but shouldn't be a limit. NET_ETHADDR_EEPROM_SPI can be just as created and added.

The code currently first checks if there is a non-zero MAC address in the eeprom. If that fails to be the case, the read_rom_hwaddr can be used by a board to supply the MAC in other ways.

If both these fails, the other code is still in place to query the environent, which then can be used to override the hardware supplied data.

Signed-off-by Olliver Schinagl <[email protected]>

---
 arch/arm/mach-sunxi/Kconfig                |  8 +--
 configs/A10-OLinuXino-Lime_defconfig       |  1 -
 configs/A20-OLinuXino-Lime2-eMMC_defconfig |  1 -
 configs/A20-OLinuXino-Lime2_defconfig      |  1 -
 configs/A20-OLinuXino-Lime_defconfig       |  1 -
 configs/A20-OLinuXino_MICRO_defconfig      |  1 -
 doc/README.enetaddr                        | 95 ++++++++++++++++++++++++++++++
 include/net.h                              | 14 +++++
 net/Kconfig                                | 59 +++++++++++++++++++
 net/eth-uclass.c                           | 42 +++++++------
 net/eth_common.c                           | 37 ++++++++++++
 net/eth_legacy.c                           |  2 +
 12 files changed, 231 insertions(+), 31 deletions(-)

This patch allows Kconfig to enable and set parameters to make it possible to read the MAC address from an EEPROM. The net core layer then uses this information to read MAC addresses from this EEPROM.

Besides the various tuneables as to how to access the eeprom (bus, address, addressing mode/length, 2 configurable that are EEPROM generic (e.g. SPI or some other form of access) which are:

NET_ETHADDR_EEPROM_OFFSET, indicating where in the EEPROM the start of the MAC address is. The default is 8 allowing for 8 bytes before the MAC for other purposes (header MAGIC for example).

NET_ETHADDR_EEPROM_CRC8, indicating the MAC is appended with a CRC8-CCIT checksum that should be verified.

Currently only I2C eeproms have been tested and thus only those options are available, but shouldn't be a limit. NET_ETHADDR_EEPROM_SPI can be just as created and added.

The code currently first checks if there is a non-zero MAC address in the eeprom. If that fails to be the case, the read_rom_hwaddr can be used by a board to supply the MAC in other ways.

If both these fails, the other code is still in place to query the environent, which then can be used to override the hardware supplied data.

Signed-off-by Olliver Schinagl <[email protected]>

With many potential places where a MAC address can be read from, the user may not know where the MAC address originated from. Print the MAC source after initializing the Ethernet device.

Signed-off-by Olliver Schinagl <[email protected]>

Currently, we print that the MAC from the SROM does not match. It can be many forms of ROM, so lets drop the S.

Signed-off-by Olliver Schinagl <[email protected]>

---
 net/eth_legacy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

net/eth_legacy.c +0 -0

index 4276058800..800b91c327 100644

Quoted 8 lines · Olliver Schinagl

Currently, we print that the MAC from the SROM does not match. It can be many forms of ROM, so lets drop the S.

Signed-off-by Olliver Schinagl <[email protected]>

---
 net/eth_legacy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

net/eth_legacy.c int eth_write_hwaddr(struct eth_device *dev, const char *base_name, · line 152

Quoted 13 lines

net/eth_legacy.c +1 -1

index 4276058800..800b91c327 100644
@@ -146,7 +146,7 @@ int eth_write_hwaddr(struct eth_device *dev, const char *base_name,
 		    memcmp(dev->enetaddr, env_enetaddr, ARP_HLEN)) {
 			printf("\nWarning: %s MAC addresses don't match:\n",
 			       dev->name);
-			printf("Address in SROM is         %pM\n",
+			printf("Address in ROM is         %pM\n",
 			       dev->enetaddr);
 			printf("Address in environment is  %pM\n",
 			       env_enetaddr);

We need to add a space in the print then as the output is manually aligned.

Quoted 8 lines · Olliver Schinagl

Currently, we print that the MAC from the SROM does not match. It can be many forms of ROM, so lets drop the S.

Signed-off-by Olliver Schinagl <[email protected]>

---
 net/eth_legacy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

net/eth_legacy.c int eth_write_hwaddr(struct eth_device *dev, const char *base_name, · line 152

Quoted 13 lines

net/eth_legacy.c +1 -1

index 4276058800..800b91c327 100644
@@ -146,7 +146,7 @@ int eth_write_hwaddr(struct eth_device *dev, const char *base_name,
                  memcmp(dev->enetaddr, env_enetaddr, ARP_HLEN)) {
                      printf("\nWarning: %s MAC addresses don't match:\n",
                             dev->name);
-                     printf("Address in SROM is         %pM\n",
+                     printf("Address in ROM is         %pM\n",
                             dev->enetaddr);
                      printf("Address in environment is  %pM\n",
                             env_enetaddr);

We need to add a space in the print then as the output is manually aligned.

Currently, we silently ignore the return value of netops->read_rom_hwaddr(). This naturally is bad and we should check if the code ran successfully.

Signed-off-by Olliver Schinagl <[email protected]>

Currently, we silently ignore the return value of netops->read_rom_hwaddr(). This naturally is bad and we should check if the code ran successfully.

Signed-off-by Olliver Schinagl <[email protected]>

In fdt_support.c we use a loop to parse the mac address string from the fdt blob, net.h has a function for this however, so lets use it.

Also, rename the variable from tmp to something more descriptive.

Signed-off-by Olliver Schinagl <[email protected]>

Currently when checking for an error in ethernet aliases in the fdt, we only check for the error case -1. It is safer to ignore anything < 0.

By rearranging logic a bit we can now also reduce identation.

Signed-off-by Olliver Schinagl <[email protected]>

Expose the function to generate a MAC adddress based on the serial number. This can then later be moved completly out of the sunxi board specific stuff.

The setup_environment quirky function still exists as it is still used to fixup the FDT for drivers that need the MAC address available in the FDT. Once that is changed, we can clean up some more.

Signed-off-by Olliver Schinagl <[email protected]>

Expose the write_hwaddr net_ops hook to write the Ethernet address.

Signed-off-by Olliver Schinagl <[email protected]>

The sunxi_emac driver is very specific to the IP used in the sun4i. Thus rename it to what it actually is, the MAC driver for the sun4i MAC.

The rename also changes the variable and function names inside the driver. There are no code changes in this patch however.

While here, use the BIT() macro for bitfields as well.

Signed-off-by Olliver Schinagl <[email protected]>

---
 drivers/net/Makefile                      |   2 +-
 drivers/net/{sunxi_emac.c => sun4i_mac.c} | 260 +++++++++++++++---------------

The sun[8x]i network drivers have some common functions. Let's introduce a common file with the reading of the MAC address as a start.

In the future, we can move more sunxi shared/common code into this file.

Signed-off-by Olliver Schinagl <[email protected]>

The sun[8x]i network drivers have some common functions. Let's introduce a common file with the reading of the MAC address as a start.

In the future, we can move more sunxi shared/common code into this file.

Signed-off-by Olliver Schinagl <[email protected]>

---
 drivers/net/sunxi_common.c | 33 +++++++++++++++++++++++++++++++++
 drivers/net/sunxi_common.h | 13 +++++++++++++
 2 files changed, 46 insertions(+)
 create mode 100644 drivers/net/sunxi_common.c

 create mode 100644 drivers/net/sunxi_common.h

With this patch sun4i_mac can now get the MAC address from the board in a predetermined board specific manner.

This patch removes the old sunxi_gmac glue layer and turns it into a proper sub-class driver.

Signed-off-by Olliver Schinagl <[email protected]>

---
 arch/arm/mach-sunxi/board.c                   |  1 -
 board/sunxi/Makefile                          |  1 -
 drivers/net/Makefile                          |  1 +
 drivers/net/designware.c                      |  1 -
 board/sunxi/gmac.c => drivers/net/sun7i_mac.c | 52 +++++++++++++++++++++++----
 5 files changed, 47 insertions(+), 9 deletions(-)
 rename board/sunxi/gmac.c => drivers/net/sun7i_mac.c (69%)

Commit e72ced234045f ("net: designware: Export the operation functions") started to expose some of the net_ops. The sunxi_gmac glue driver also needs the start function, so let us expose that as well.

Signed-off-by Olliver Schinagl <[email protected]>

---
 drivers/net/designware.c | 2 +-
 drivers/net/designware.h | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

Commit e72ced234045f ("net: designware: Export the operation functions") started to expose some of the net_ops. The sunxi_gmac glue driver also needs the start function, so let us expose that as well.

Signed-off-by Olliver Schinagl <[email protected]>

With this patch sun7i_mac can now get the MAC address from the board in a predetermined board specific manner.

Signed-off-by Olliver Schinagl <[email protected]>

---
 drivers/net/Makefile    | 2 +-
 drivers/net/sun7i_mac.c | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

drivers/net/Makefile +0 -0

index 86dee7d746..5df5567536 100644

Quoted 9 lines · Olliver Schinagl

With this patch sun7i_mac can now get the MAC address from the board in a predetermined board specific manner.

Signed-off-by Olliver Schinagl <[email protected]>

---
 drivers/net/Makefile    | 2 +-
 drivers/net/sun7i_mac.c | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

drivers/net/Makefile obj-$(CONFIG_E1000) += e1000.o · line 26

Quoted 10 lines

drivers/net/Makefile +1 -1

index 86dee7d746..5df5567536 100644
@@ -23,7 +23,7 @@ obj-$(CONFIG_E1000) += e1000.o
 obj-$(CONFIG_E1000_SPI) += e1000_spi.o
 obj-$(CONFIG_EEPRO100) += eepro100.o
 obj-$(CONFIG_SUN4I_EMAC) += sunxi_common.o sun4i_mac.o
-obj-$(CONFIG_SUN7I_MAC) += sun7i_mac.o
+obj-$(CONFIG_SUN7I_MAC) += sunxi_common.o sun7i_mac.o

Shouldn't that be turned in a Kconfig option selected by both drivers?

Maxime

Hey Maxime,

Quoted 10 lines · Maxime Ripard

Quoted 9 lines · Olliver Schinagl

With this patch sun7i_mac can now get the MAC address from the board in a predetermined board specific manner.

Signed-off-by Olliver Schinagl <[email protected]>

---
 drivers/net/Makefile    | 2 +-
 drivers/net/sun7i_mac.c | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

Quoted 12 lines

drivers/net/Makefile obj-$(CONFIG_E1000) += e1000.o · line 26

Quoted 10 lines

drivers/net/Makefile +1 -1

index 86dee7d746..5df5567536 100644
@@ -23,7 +23,7 @@ obj-$(CONFIG_E1000) += e1000.o
 obj-$(CONFIG_E1000_SPI) += e1000_spi.o
 obj-$(CONFIG_EEPRO100) += eepro100.o
 obj-$(CONFIG_SUN4I_EMAC) += sunxi_common.o sun4i_mac.o
-obj-$(CONFIG_SUN7I_MAC) += sun7i_mac.o
+obj-$(CONFIG_SUN7I_MAC) += sunxi_common.o sun7i_mac.o

Shouldn't that be turned in a Kconfig option selected by both drivers?

What do you mean? There are some common functions that all of them share, such as reading the MAC address.

Or do you mean we should make the net_op rom_read_hwaddr() a configurable Kconfig?

Fix a few whitespaces errors in the sun8i driver.

Signed-off-by Olliver Schinagl <[email protected]>

With commit 2681e78a5ee ("configs: Re-sync") we lost the EEPROM settings from the defconfig files. Let us re-add them.

Signed-off-by Olliver Schinagl <[email protected]>

This patch-series introduces methods to retrieve the MAC address from an onboard EEPROM. The series does a few small cleanups at the start, as either I ran into them while doing this series and fixed them along the way or actually depended on them.

A manufacturer wants to produce boards and may even have MAC addresses for boards. Maintaining unique environments on a per-board basis however is horrible. Also this data should be very persistent, and not easily deletable by simply wiping the environment or device tree. Finally there are chips available on the market with a pre-programmed MAC address chips (proms)
that a board manufacturer wants to use. Because of this, the MAC needs to be stored be able to read from such an 'external' source.

The current idea of the eeprom layout, is to skip the first 8 bytes, so that other information can be stored there if needed, for example a header with some magic to identify the EEPROM. Or equivalent purposes.

After those 8 bytes the MAC address follows the first macaddress. The macaddress is appended by a CRC8 byte and then padded to make for nice 8 bytes. Following the first macaddress one can store a second, or a third etc etc macaddress.

The CRC8 is optional (via a define) but is strongly recommended to have. It helps preventing user error and more importantly, checks if the bytes read are actually a user inserted address. E.g. only writing 1 macaddress into the eeprom but trying to consume 2.

These patches where tested on Olimex OLinuXino Lime1 (A10/A20), Lime2 (NAND and eMMC) and A20-OLinuXino-MICRO-4G variants and have been in use internally on our production systems since v2 of this patch set.

TODO in a follow up series:
* Replace the I2C specific read function to a generic function using Maxime Ripard's EEPROM uclass once merged.
* Upgrade to use the EERPOM layout functionality as suggested by Igor Grinberg
* Add EEPROM/Mac commands suggested by York Sun

The series was initially based ontop of Joe's u-boot_net/next branches, but I rebased it ontop of u-boot/master to get the latest sunxi changes, including:
sun7i: Add support for Olimex A20-OLinuXino-LIME2-eMMC
ARM: dts: sunxi: Change node name for pwrseq pin on Olinuxino-lime2-emmc

Which are on their way to being merged in master. So once those land into master this all should apply cleanly. Sorry Joe!

=======
Changes since v5:
* Removed changes to eth_legacy as per Simon's request
* Adapted, improved and fixed the emac/gmac Kconfig conversion
* Renamed sunxi_[eg]mac to sun[47]_mac to be more clear and allowing preparing to merge common parts into sunxi_common
* Added a few more MAC helpers
* Added PHY_GIGE as Kconfig symbol

Changes since v4:
* Some patches have been merged, so those are no longer included in the set.
* Reworked review comments
* Added the source of a MAC address
* Made sunxi-gmac a subclassed driver
* Switch sunxi-[eg]mac to Kconfig
* Drop the FDT injection code for now, this needs more thought/discussion

Changes since v3:
* Split off board specific stuff and only modify the generic functions
* Make reading of an eeprom available to every board. By default this is unconfigure and thus should just fall through
* Clean some minor bits up (ARP_HLEN) and use it more generically
* Update the gen_ethaddr_crc as suggested by simon
* Let the fixup_ethernet from fdt_common insert mac addresses to the environment
for unconfigured devices. There is a small caveat here however as
described
in the TODO above.
* Print the mac address that u-boot assigned to each device.

Changes since v2:
* Drop the id byte altogether and just mark it as reserved. The 'index' can be
used to indicate the interface instead
* Addopt the read_rom_hwaddr hooks
* Renamed crc8 tool to gen_ethaddr_crc
* Improved the layout EEPROM example
* Made a function out of the hwaddress writing function in sunxi_emac so it can be re-used as the write_hwaddr net_ops hook.
* No longer handle fdt parameters in board.c

Changes since v1:
* Do not CRC the id byte, move it just after the crc byte.
One of the reasons I decided to move it after the crc8 was mostly due to mass
generation of MAC + CRC combo's where the ID is still unknown. Also not crc-ing
the ID means that it is much easier for a user to change it (via the u-boot i2c
cmd line or from within linux) without having to worry about the crc.
* Add a generator to convert a MAC address from the input to a MAC + CRC8 on the output.

Emmanuel Vadot (1):
ARM: dts: sunxi: Change node name for pwrseq pin on
Olinuxino-lime2-emmc

Olliver Schinagl (27):
sun7i: Add support for Olimex A20-OLinuXino-LIME2-eMMC
net: core: Add PHY_GIGE as a Kconfig symbol
net: sunxi simplify defconfig
net: sunxi: Re-add RTL8211X_PHY_FORCE_MASTER
net: sunxi: Restore sunxi_[eg]mac behavior
net: sunxi: Move GMAC_TX_DELAY to the driver
net: cosmetic: Do not use magic values for ARP_HLEN
net: core: Sanitize get/set operations for enetaddr
net: core: Inform the user of the device MAC address
net: core: Add MAC address helper functions
net: Add ability to set MAC address via EEPROM
net: core: print the source of the MAC address
net: cosmetic: A MAC address is not limited to SROM
net: core: Check return value of read_rom_hwaddr
fdt: fixup_eth: Remove code duplication with a function
fdt: fixup_eth: improve error catching/reduce identation
arm: sunxi: Expose function to generate sunxi-specific a MAC address
net: sunxi_emac: Write HW address via net_ops hook
net: sunxi: Rename sunxi_emac to a more specific sun4i_mac
net: sunxi: Have sunxi common functions together
net: sun4i_mac: Add read_rom_hwaddr hook
net: sun7i_mac: Make the sun7i a dw driver subclass
net: dw: Expose designware_eth_start
net: sun7i_mac: Add read_rom_hwaddr hook
net: sun8i_gmac: Add read_rom_hwaddr hook
net: sun8i: fix whitespace
net: sunxi: Enable eeprom on OLinuXino Lime boards (again)

 arch/arm/dts/sun7i-a20-olinuxino-lime2-emmc.dts |   2 +-
 arch/arm/include/asm/arch-sunxi/sys_proto.h     |   4 +-
 arch/arm/mach-sunxi/Kconfig                     |  14 +-
 arch/arm/mach-sunxi/board.c                     |   1 -
 board/sunxi/MAINTAINERS                         |   5 +
 board/sunxi/Makefile                            |   1 -
 board/sunxi/board.c                             | 146 ++++++++-----
 common/fdt_support.c                            |  33 ++-
 configs/A10-OLinuXino-Lime_defconfig            |   4 +-
 configs/A20-OLinuXino-Lime2-eMMC_defconfig      |  36 ++++
 configs/A20-OLinuXino-Lime2_defconfig           |   8 +-
 configs/A20-OLinuXino-Lime_defconfig            |   8 +-
 configs/A20-OLinuXino_MICRO_defconfig           |   8 +-
 configs/A20-Olimex-SOM-EVB_defconfig            |   4 +-
 configs/Bananapi_defconfig                      |   3 +-
 configs/Bananapro_defconfig                     |   3 +-
 configs/CSQ_CS908_defconfig                     |   4 +-
 configs/Colombus_defconfig                      |   3 +-
 configs/Cubieboard2_defconfig                   |   4 +-
 configs/Cubietruck_defconfig                    |   3 +-
 configs/Hummingbird_A31_defconfig               |   3 +-
 configs/Itead_Ibox_A20_defconfig                |   4 +-
 configs/Lamobo_R1_defconfig                     |   3 +-
 configs/Linksprite_pcDuino3_Nano_defconfig      |   3 +-
 configs/Linksprite_pcDuino3_defconfig           |   4 +-
 configs/Mele_A1000G_quad_defconfig              |   4 +-
 configs/Mele_I7_defconfig                       |   4 +-
 configs/Mele_M3_defconfig                       |   4 +-
 configs/Mele_M5_defconfig                       |   4 +-
 configs/Mele_M9_defconfig                       |   4 +-
 configs/Orangepi_defconfig                      |   3 +-
 configs/Orangepi_mini_defconfig                 |   3 +-
 configs/Sinlinx_SinA31s_defconfig               |   4 +-
 configs/Sinovoip_BPI_M2_defconfig               |   3 +-
 configs/Wits_Pro_A20_DKT_defconfig              |   3 +-
 configs/i12-tvbox_defconfig                     |   4 +-
 configs/icnova-a20-swac_defconfig               |   4 +-
 configs/mixtile_loftq_defconfig                 |   3 +-
 doc/README.enetaddr                             |  95 +++++++++
 drivers/net/Kconfig                             |  76 +++++--
 drivers/net/Makefile                            |   5 +-
 drivers/net/designware.c                        |   3 +-
 drivers/net/designware.h                        |   1 +
 drivers/net/{sunxi_emac.c => sun4i_mac.c}       | 267 +++++++++++++-----------
 board/sunxi/gmac.c => drivers/net/sun7i_mac.c   |  54 ++++-
 drivers/net/sun8i_emac.c                        |  15 +-
 drivers/net/sunxi_common.c                      |  33 +++
 drivers/net/sunxi_common.h                      |  13 ++
 include/configs/sunxi-common.h                  |  13 +-
 include/net.h                                   |  76 ++++++-
 net/Kconfig                                     |  59 ++++++
 net/eth-uclass.c                                |  65 +++---
 net/eth_common.c                                |  53 ++++-
 net/eth_legacy.c                                |   4 +-
 scripts/config_whitelist.txt                    |   2 -
 55 files changed, 842 insertions(+), 350 deletions(-)
 create mode 100644 configs/A20-OLinuXino-Lime2-eMMC_defconfig
 rename drivers/net/{sunxi_emac.c => sun4i_mac.c} (59%)
 rename board/sunxi/gmac.c => drivers/net/sun7i_mac.c (67%)
 create mode 100644 drivers/net/sunxi_common.c
 create mode 100644 drivers/net/sunxi_common.h

Hey all,

Quoted 43 lines · Olliver Schinagl

This patch-series introduces methods to retrieve the MAC address from an onboard EEPROM. The series does a few small cleanups at the start, as either I ran into them while doing this series and fixed them along the way or actually depended on them.

A manufacturer wants to produce boards and may even have MAC addresses for boards. Maintaining unique environments on a per-board basis however is horrible. Also this data should be very persistent, and not easily deletable by simply wiping the environment or device tree. Finally there are chips available on the market with a pre-programmed MAC address chips (proms)
that a board manufacturer wants to use. Because of this, the MAC needs to be stored be able to read from such an 'external' source.

The current idea of the eeprom layout, is to skip the first 8 bytes, so that other information can be stored there if needed, for example a header with some magic to identify the EEPROM. Or equivalent purposes.

After those 8 bytes the MAC address follows the first macaddress. The macaddress is appended by a CRC8 byte and then padded to make for nice 8 bytes. Following the first macaddress one can store a second, or a third etc etc macaddress.

The CRC8 is optional (via a define) but is strongly recommended to have. It helps preventing user error and more importantly, checks if the bytes read are actually a user inserted address. E.g. only writing 1 macaddress into the eeprom but trying to consume 2.

These patches where tested on Olimex OLinuXino Lime1 (A10/A20), Lime2 (NAND and eMMC) and A20-OLinuXino-MICRO-4G variants and have been in use internally on our production systems since v2 of this patch set.

TODO in a follow up series:
* Replace the I2C specific read function to a generic function using Maxime Ripard's EEPROM uclass once merged.
* Upgrade to use the EERPOM layout functionality as suggested by Igor Grinberg
* Add EEPROM/Mac commands suggested by York Sun

The series was initially based ontop of Joe's u-boot_net/next branches, but I rebased it ontop of u-boot/master to get the latest sunxi changes, including:
sun7i: Add support for Olimex A20-OLinuXino-LIME2-eMMC
ARM: dts: sunxi: Change node name for pwrseq pin on Olinuxino-lime2-emmc

Quoted 2 lines

Which are on their way to being merged in master. So once those land into master this all should apply cleanly. Sorry Joe!

So I accidentally added those too to this patch series as well (first two patches). Which are technically not part of this series and are on their way as it is. Does patchwork figure out this automatically, or do you prefer a resend with the 2 patches removed?

Quoted 49 lines · Olliver Schinagl

Hey all,

Quoted 45 lines · Olliver Schinagl

This patch-series introduces methods to retrieve the MAC address from an onboard EEPROM. The series does a few small cleanups at the start, as either I ran into them while doing this series and fixed them along the way or actually depended on them.

A manufacturer wants to produce boards and may even have MAC addresses for boards. Maintaining unique environments on a per-board basis however is horrible. Also this data should be very persistent, and not easily deletable by simply wiping the environment or device tree. Finally there are chips available on the market with a pre-programmed MAC address chips (proms)
that a board manufacturer wants to use. Because of this, the MAC needs to be stored be able to read from such an 'external' source.

The current idea of the eeprom layout, is to skip the first 8 bytes, so that other information can be stored there if needed, for example a header with some magic to identify the EEPROM. Or equivalent purposes.

After those 8 bytes the MAC address follows the first macaddress. The macaddress is appended by a CRC8 byte and then padded to make for nice 8 bytes. Following the first macaddress one can store a second, or a third etc etc macaddress.

The CRC8 is optional (via a define) but is strongly recommended to have. It helps preventing user error and more importantly, checks if the bytes read are actually a user inserted address. E.g. only writing 1 macaddress into the eeprom but trying to consume 2.

These patches where tested on Olimex OLinuXino Lime1 (A10/A20), Lime2 (NAND and eMMC) and A20-OLinuXino-MICRO-4G variants and have been in use internally on our production systems since v2 of this patch set.

TODO in a follow up series:
* Replace the I2C specific read function to a generic function using Maxime Ripard's EEPROM uclass once merged.
* Upgrade to use the EERPOM layout functionality as suggested by Igor Grinberg
* Add EEPROM/Mac commands suggested by York Sun

The series was initially based ontop of Joe's u-boot_net/next branches, but I rebased it ontop of u-boot/master to get the latest sunxi changes, including:
sun7i: Add support for Olimex A20-OLinuXino-LIME2-eMMC
ARM: dts: sunxi: Change node name for pwrseq pin on Olinuxino-lime2-emmc

Which are on their way to being merged in master. So once those land into master this all should apply cleanly. Sorry Joe!

Quoted 4 lines

So I accidentally added those too to this patch series as well (first two patches). Which are technically not part of this series and are on their way as it is. Does patchwork figure out this automatically, or do you prefer a resend with the 2 patches removed?

I've made patchwork reflect things correctly.

* Renamed sunxi_[eg]mac to sun[47]_mac to be more clear and allowing preparing to merge common parts into sunxi_common