OMAP Bootloader Overview
There are several stages of bootloaders that perform different levels of initialization on an OMAP platform, in order to eventually load and run the filesystem. This figure shows the booting sequence of the ROM code, x-loader, u-boot, and kernel, with each stage performing enough configuration in order to load and execute the next.
The Bootloader Project contains information regarding the building and flashing procedures for u-boot and x-loader. These two items are necessary to bring up your OMAP board's environment.
Why are there two bootloaders? The first bootloader, x-loader, is a stripped-down version of the u-boot, designed to run in OMAP's small on-chip SRAM. It initializes the main off-chip memory of the system and other necessary device drivers, and then loads the larger bootloader for Linux, u-boot.
OMAP Boot Sequence
Detailed documentation on the OMAP boot sequence is available in the OMAP TRM, Initialization chapter. The TRMs for multiple OMAP platforms (OMAP4430, OMAP4460, OMAP34xx, etc) are available here: http://www.ti.com/general/docs/wtbu/wtbudocumentcenter.tsp?templateId=6123&navigationId=12037
The internal ROM Code can attempt to boot from several different peripheral and memory devices, including, but not limited to: Serial (UART3), SD Card, eMMC, NAND, and USB. The order in which these devices are searched for a valid first-stage booting image (x-loader) is determine by a set of GPIO configuration pins referred to as SYSBOOT. The TRM includes a table that shows the booting device list that each combination of the SYSBOOT pins refers to.
The SYSBOOT value can be read from physical address 0x480022f0, either using JTAG, or if you have linux running, use devmem2:
# devmem2 0x480022f0 b /dev/mem opened. Memory mapped at address 0x40020000. Value at address 0x480022F0 (0x400202f0): 0x2F
For OMAP34xx, there is a tool to show the boot order: omap34xx-boot-order
First Stage Boot
For example, the SYSBOOT pins could be set such that booting device list consists of 1) serial (UART3), 2) SD card (MMC1), and 3) NAND flash. In this case, the ROM code would first look for a valid x-loader over the serial port, then in the SD card, then in the NAND flash. Whenever it finds a valid x-loader, it proceeds with execution of that binary.
For serial boot, a simple ID is written out of the serial port. If the host responds correctly within a short window of time, the ROM will read from the serial port and transfer the data to the internal SRAM. Control is passed to the start of SDRAM if no errors are detected. UART3 is the only uart for which the ROM will attempt to load from.
SD Card Boot
If MMC is included in the booting device list, the ROM looks for an SD Card on the first MMC controller. If a card is found, the ROM then looks for the first FAT32 partition within the partition table. Once the partition is found, the root directory is scanned for a special signed file called "MLO" (which is the x-loader binary with a header containing the memory location to load the file to and the size of the file). Assuming all is well with the file, it is transfered into the internal SRAM and control is passed to it. Both MMC1 and MMC2 can be used for booting.
NAND / eMMC Boot
If NAND is included in the booting device list, the ROM attempts to load the first sector of NAND. If the sector is bad, corrupt, or blank, the ROM will try the next sector (up to 4) before exiting. Once a good sector is found, the ROM transfers the contents to SRAM and transfers control to it. (The same steps are performed for eMMC if eMMC is included in the booting device list instead of NAND.)
Second Stage Boot
It is the job of the x-loader to transfer the 2nd stage loader into main memory, which we call the u-boot. Typically both the x-loader and u-boot come from the same storage medium. For example, typically if the x-loader is transferred via USB, the u-boot will also be transferred via USB, and if the x-loader is transferred via SD card, the u-boot will also be transferred via SD card. However, this is not required. For example, you could flash the serial x-loader into the NAND. The ROM code will load the x-loader from NAND and transfer control to the x-loader, which will wait for the u-boot to be downloaded from the serial port.
In the case of loading the u-boot over the serial port, the serial x-loader waits for the host to initiate a kermit connection to reliably transfer large files into main memory. Once the file transfer is completed successfully, control is transfered.
SD Card Boot
In the case of loading the u-boot via the SD card, the SD card x-loader looks for a FAT32 partition on the first MMC controller and scans the top level directory for a file named "u-boot.bin". It then transfers the file into main memory and transfers control to it.
NAND / eMMC Boot
The NAND x-loader expects u-boot to be loaded at the 5th sector (offset 0x00800000). It transfers the image from NAND into main memory and transfers control to it.
The x-loader is a small first stage bootloader derived from the u-boot base code. It is loaded into the internal static RAM by the OMAP ROM code. Due to the small size of the internal static RAM, the x-loader is stripped down to the essentials. The x-loader configures the pin muxing, clocks, DDR, and serial console, so that it can access and load the second stage bootloader (u-boot) into the DDR. This figure shows the code flow in the x-loader, beginning in start.S.
The u-boot is a second stage bootloader that is loaded by the x-loader into DDR. It comes from Das U-Boot. The u-boot can perform CPU dependent and board dependent initialization and configuration not done in the x-loader. The u-boot also includes fastboot functionality for partitioning and flashing the eMMC. The u-boot runs on the Master CPU (CPU ID 0), which is responsible for the initialization and booting; at the same time, the Slave CPU (CPU ID 1) is held in the “wait for event” state. This figure shows the code flow in the u-boot, beginning in start.S.
Building the Bootloaders
The following steps will build the x-loader and u-boot.
Note: In order to build the x-loader or u-boot, it is recommended that you use the omappedia Release Notes page corresponding to the platform you are using (Blaze, Blaze Tablet, Panda, or Zoom) and follow the instructions for that release: http://www.omappedia.com/wiki/Release_Notes In general, the instructions will be very similar with the following exceptions:
- The commit ID used to pull the code from the u-boot and x-loader git trees varies based on the release. It is recommended that you use the correct commit ID corresponding to your release, as this is how the validation is performed.
- The config file used to setup the configuration when building the u-boot and x-loader varies based on the platform you are using. See the tables below.
Also, ensure that you have installed the pre-requisite packages for building the bootloaders, namely Git and the CodeSourcery ARM Compiler. These instructions are included in the omappedia Release Notes corresponding to your release. You can also find the information here:
- Git is available for download at git-scm.com. For more details on installing and using git please see this wiki page.
- For the CodeSourcery ARM Compiler, visit Support_Tools/Cross_Compilers for further information.
Downloading u-boot source
Clone the u-boot git tree and checkout the source code corresponding to this release:
# mkdir u-boot # cd u-boot # git clone git://git.omapzoom.org/repo/u-boot.git # git checkout <<commit_ID_for_uboot_in_this_release>>
Make the u-boot source code using the correct config file depending on your platform:
# cd u-boot # make distclean # make CROSS_COMPILE=arm-none-linux-gnueabi- <<config_file_depending_on_your_platform>> # make CROSS_COMPILE=arm-none-linux-gnueabi-
U-boot config files
|44xx Blaze Tablet||omap44XXtablet_config|
The resulting u-boot.bin is located in the top-level of your u-boot directory.
Accessing x-loader source
Clone the x-loader git tree and checkout the source code corresponding to this release:
# cd x-loader # git clone git://git.omapzoom.org/repo/x-loader.git # git checkout <<commit_ID_for_xloader_in_this_release>>
Make the x-loader source code using the correct config file depending on your platform:
# cd x-loader # make distclean # make CROSS_COMPILE=arm-none-linux-gnueabi- <<config_file_depending_on_your_platform>> # make CROSS_COMPILE=arm-none-linux-gnueabi- ift
x-loader config files
|44xx Blaze Tablet||omap44XXtablet_config|
To build x-loader to boot over a serial connection for OMAP3, use "omap3430labradordownload_config"
The resulting MLO is located in the top-level of your x-loader directory.
Note: If you are using an HS (High Security) OMAP device, an extra step is required. First, build x-load.bin using the steps above. Then, download the MShield signing tool and use the commands below. Contact your TI representative to get access to this tool.
# cd mshield-dk-root-folder # ./generate_MLO <<OMAP type>> x-load.bin
For example, for an ES2.3 OMAP4430 device, use the command:
# ./generate_MLO OMAP4430 ES2.3 x-load.bin
The resulting MLO is located in the top-level of your mshield-dk directory.
Refer to Android_Fastboot to review flashing methods for Blaze and Blaze Tablet using Fastboot.
Refer to Zoom Flashing to review flashing methods for Zoom1/Zoom2/Zoom3.
Refer to 3630SDP Flashing to review flashing methods for 3630SDP board.
Refer to eMMC Boot to review flashing and boot methods from eMMC device.
Refer to [] for a USB downloader for PandaBoard.
U-boot upstreaming is being carried out as a separate project. Please refer to the following link for more details.