U-Boot makes use of the so-called environment variables to define various aspects of the target functionality. The U-Boot environment is stored in the SD Card memory and is persistent across power or reset cycles. Parameters defined by the U-boot environment variables include: target IP address, target MAC address, location in RAM where a Linux bootable image will be loaded, and many others.

In various use case scenarios, it may be needed to read or update U-Boot environment variables from Linux. Probably, the most obvious scenario would be when a remote target is accessible only through an Ethernet or WiFi link and no access to the serial console is available. In that scenario, it is impossible to get access to the U-Boot serial line command monitor, however the target still can be connected to over a telnet or ssh link providing access to the Linux shell interface. This application note explains how to read and update U-Boot environment variables from Linux.

The Linux utilities are called fw_printenv (read U-Boot environment) and fw_setenv (modify U-Boot environment). The two utitlies are in fact implemented as a single utility, and the name of the executable file is used to distinguish between the fw_printenv and fw_setenv use cases on the target.

When you boot the newly installed Linux image on the target, you will be able to read and update the U-Boot environment. Here are some examples.

The following command reads the whole U-Boot environment:

/ # fw_printenv
addip=setenv bootargs ${bootargs} ip=${ipaddr}:${serverip}:${gatewayip}:${netmask}:${hostname}:eth0:off
baudrate=115200
bootargs=console=ttyLP0,115200 consoleblank=0 ignore_loglevel
...
/ #

Here is how you can read a specific environment variable:

/ # fw_printenv bootdelay
bootdelay=0
/ #

Here is how you change the value of an environment variable:

/ # fw_setenv bootdelay 5
/ #

Having updated bootdelay as shown above, on the next reset U-Boot will wait for 5 seconds for you to press a keyboard and enter the U-Boot command monitor before proceeding to boot Linux.

The source files of the fw_printenv utility reside in the U-boot source tree, which is located in the u-boot/ sub-directory in the Emcraft software distribution.

fw_setenv and fw_printenv are included in the default rootfs project. The utilities are built during the rootfs compilation according to the rules specified in the projects/rootfs/env/Makefile:

• The U-Boot source tree is configured for the NXP i.MX RT1060 EVK board;
• The fw_printenv utility is built:

When run on the target, the utility makes use of a configuration file that defines the geometry of the device and where the U-Boot environment is located in it. Since we keep the environment in the uboot.env file on SD Card here is what the configuration file looks like for the NXP i.MX RT1060 EVK board:

-bash-4.2$ cat projects/rootfs/etc/fw_env.config.nxpevk
# Configuration file for fw_(printenv/saveenv) utility.
# Up to two entries are valid, in this case the redundant
# environment sector is assumed present.

# Configuration for the iMX RT106x NXP-EVK board (CONFIG_ENV_IS_IN_FAT)
#
# MTD device name       Device offset   Env. size
/mnt/sdcard/uboot.env   0x0000          0x2000

In the rootfs project, the fw_setenv and the configuration file are added to rootfs.initramfs:

...
file /bin/fw_printenv ${INSTALL_ROOT}/u-boot/tools/env/fw_printenv 755 0 0
slink /bin/fw_setenv fw_printenv 755 0 0
ifarch IMXRT106X_NXPEVK file /etc/fw_env.config \
${INSTALL_ROOT}/projects/${SAMPLE}/etc/fw_env.config.nxpevk 644 0 0
...

Several U-Boot environment variables are used very early during the Linux bootstrap, at the /etc/rc stage.