Binman Entry Documentation =========================== This file describes the entry types supported by binman. These entry types can be placed in an image one by one to build up a final firmware image. It is fairly easy to create new entry types. Just add a new file to the 'etype' directory. You can use the existing entries as examples. Note that some entries are subclasses of others, using and extending their features to produce new behaviours. Entry: atf-bl31: ARM Trusted Firmware (ATF) BL31 blob ----------------------------------------------------- Properties / Entry arguments: - atf-bl31-path: Filename of file to read into entry. This is typically called bl31.bin or bl31.elf This entry holds the run-time firmware, typically started by U-Boot SPL. See the U-Boot README for your architecture or board for how to use it. See https://github.com/ARM-software/arm-trusted-firmware for more information about ATF. Entry: atf-fip: ARM Trusted Firmware's Firmware Image Package (FIP) ------------------------------------------------------------------- A FIP_ provides a way to group binaries in a firmware image, used by ARM's Trusted Firmware A (TF-A) code. It is a simple format consisting of a table of contents with information about the type, offset and size of the binaries in the FIP. It is quite similar to FMAP, with the major difference that it uses UUIDs to indicate the type of each entry. Note: It is recommended to always add an fdtmap to every image, as well as any FIPs so that binman and other tools can access the entire image correctly. The UUIDs correspond to useful names in `fiptool`, provided by ATF to operate on FIPs. Binman uses these names to make it easier to understand what is going on, although it is possible to provide a UUID if needed. The contents of the FIP are defined by subnodes of the atf-fip entry, e.g.:: atf-fip { soc-fw { filename = "bl31.bin"; }; scp-fwu-cfg { filename = "bl2u.bin"; }; u-boot { fip-type = "nt-fw"; }; }; This describes a FIP with three entries: soc-fw, scp-fwu-cfg and nt-fw. You can use normal (non-external) binaries like U-Boot simply by adding a FIP type, with the `fip-type` property, as above. Since FIP exists to bring blobs together, Binman assumes that all FIP entries are external binaries. If a binary may not exist, you can use the `--allow-missing` flag to Binman, in which case the image is still created, even though it will not actually work. The size of the FIP depends on the size of the binaries. There is currently no way to specify a fixed size. If the `atf-fip` node has a `size` entry, this affects the space taken up by the `atf-fip` entry, but the FIP itself does not expand to use that space. Some other FIP features are available with Binman. The header and the entries have 64-bit flag works. The flag flags do not seem to be defined anywhere, but you can use `fip-hdr-flags` and fip-flags` to set the values of the header and entries respectively. FIP entries can be aligned to a particular power-of-two boundary. Use fip-align for this. Binman only understands the entry types that are included in its implementation. It is possible to specify a 16-byte UUID instead, using the fip-uuid property. In this case Binman doesn't know what its type is, so just uses the UUID. See the `u-boot` node in this example:: binman { atf-fip { fip-hdr-flags = /bits/ 64 <0x123>; fip-align = <16>; soc-fw { fip-flags = /bits/ 64 <0x456>; filename = "bl31.bin"; }; scp-fwu-cfg { filename = "bl2u.bin"; }; u-boot { fip-uuid = [fc 65 13 92 4a 5b 11 ec 94 35 ff 2d 1c fc 79 9c]; }; }; fdtmap { }; }; Binman allows reading and updating FIP entries after the image is created, provided that an FDPMAP is present too. Updates which change the size of a FIP entry will cause it to be expanded or contracted as needed. Properties for top-level atf-fip node ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ fip-hdr-flags (64 bits) Sets the flags for the FIP header. Properties for subnodes ~~~~~~~~~~~~~~~~~~~~~~~ fip-type (str) FIP type to use for this entry. This is needed if the entry name is not a valid type. Value types are defined in `fip_util.py`. The FIP type defines the UUID that is used (they map 1:1). fip-uuid (16 bytes) If there is no FIP-type name defined, or it is not supported by Binman, this property sets the UUID. It should be a 16-byte value, following the hex digits of the UUID. fip-flags (64 bits) Set the flags for a FIP entry. Use in one of the subnodes of the 7atf-fip entry. fip-align Set the alignment for a FIP entry, FIP entries can be aligned to a particular power-of-two boundary. The default is 1. Adding new FIP-entry types ~~~~~~~~~~~~~~~~~~~~~~~~~~ When new FIP entries are defined by TF-A they appear in the `TF-A source tree`_. You can use `fip_util.py` to update Binman to support new types, then `send a patch`_ to the U-Boot mailing list. There are two source files that the tool examples: - `include/tools_share/firmware_image_package.h` has the UUIDs - `tools/fiptool/tbbr_config.c` has the name and descripion for each UUID To run the tool:: $ tools/binman/fip_util.py -s /path/to/arm-trusted-firmware Warning: UUID 'UUID_NON_TRUSTED_WORLD_KEY_CERT' is not mentioned in tbbr_config.c file Existing code in 'tools/binman/fip_util.py' is up-to-date If it shows there is an update, it writes a new version of `fip_util.py` to `fip_util.py.out`. You can change the output file using the `-i` flag. If you have a problem, use `-D` to enable traceback debugging. FIP commentary ~~~~~~~~~~~~~~ As a side effect of use of UUIDs, FIP does not support multiple entries of the same type, such as might be used to store fonts or graphics icons, for example. For verified boot it could be used for each part of the image (e.g. separate FIPs for A and B) but cannot describe the whole firmware image. As with FMAP there is no hierarchy defined, although FMAP works around this by having 'section' areas which encompass others. A similar workaround would be possible with FIP but is not currently defined. It is recommended to always add an fdtmap to every image, as well as any FIPs so that binman and other tools can access the entire image correctly. .. _FIP: https://trustedfirmware-a.readthedocs.io/en/latest/design/firmware-design.html#firmware-image-package-fip .. _`TF-A source tree`: https://git.trustedfirmware.org/TF-A/trusted-firmware-a.git .. _`send a patch`: https://www.denx.de/wiki/U-Boot/Patches Entry: blob: Arbitrary binary blob ---------------------------------- Note: This should not be used by itself. It is normally used as a parent class by other entry types. Properties / Entry arguments: - filename: Filename of file to read into entry - compress: Compression algorithm to use: none: No compression lz4: Use lz4 compression (via 'lz4' command-line utility) This entry reads data from a file and places it in the entry. The default filename is often specified specified by the subclass. See for example the 'u-boot' entry which provides the filename 'u-boot.bin'. If compression is enabled, an extra 'uncomp-size' property is written to the node (if enabled with -u) which provides the uncompressed size of the data. Entry: blob-dtb: A blob that holds a device tree ------------------------------------------------ This is a blob containing a device tree. The contents of the blob are obtained from the list of available device-tree files, managed by the 'state' module. Entry: blob-ext: Externally built binary blob --------------------------------------------- Note: This should not be used by itself. It is normally used as a parent class by other entry types. If the file providing this blob is missing, binman can optionally ignore it and produce a broken image with a warning. See 'blob' for Properties / Entry arguments. Entry: blob-ext-list: List of externally built binary blobs ----------------------------------------------------------- This is like blob-ext except that a number of blobs can be provided, typically with some sort of relationship, e.g. all are DDC parameters. If any of the external files needed by this llist is missing, binman can optionally ignore it and produce a broken image with a warning. Args: filenames: List of filenames to read and include Entry: blob-named-by-arg: A blob entry which gets its filename property from its subclass ----------------------------------------------------------------------------------------- Properties / Entry arguments: - -path: Filename containing the contents of this entry (optional, defaults to None) where is the blob_fname argument to the constructor. This entry cannot be used directly. Instead, it is used as a parent class for another entry, which defined blob_fname. This parameter is used to set the entry-arg or property containing the filename. The entry-arg or property is in turn used to set the actual filename. See cros_ec_rw for an example of this. Entry: blob-phase: Section that holds a phase binary ---------------------------------------------------- This is a base class that should not normally be used directly. It is used when converting a 'u-boot' entry automatically into a 'u-boot-expanded' entry; similarly for SPL. Entry: cbfs: Coreboot Filesystem (CBFS) --------------------------------------- A CBFS provides a way to group files into a group. It has a simple directory structure and allows the position of individual files to be set, since it is designed to support execute-in-place in an x86 SPI-flash device. Where XIP is not used, it supports compression and storing ELF files. CBFS is used by coreboot as its way of orgnanising SPI-flash contents. The contents of the CBFS are defined by subnodes of the cbfs entry, e.g.:: cbfs { size = <0x100000>; u-boot { cbfs-type = "raw"; }; u-boot-dtb { cbfs-type = "raw"; }; }; This creates a CBFS 1MB in size two files in it: u-boot.bin and u-boot.dtb. Note that the size is required since binman does not support calculating it. The contents of each entry is just what binman would normally provide if it were not a CBFS node. A blob type can be used to import arbitrary files as with the second subnode below:: cbfs { size = <0x100000>; u-boot { cbfs-name = "BOOT"; cbfs-type = "raw"; }; dtb { type = "blob"; filename = "u-boot.dtb"; cbfs-type = "raw"; cbfs-compress = "lz4"; cbfs-offset = <0x100000>; }; }; This creates a CBFS 1MB in size with u-boot.bin (named "BOOT") and u-boot.dtb (named "dtb") and compressed with the lz4 algorithm. Properties supported in the top-level CBFS node: cbfs-arch: Defaults to "x86", but you can specify the architecture if needed. Properties supported in the CBFS entry subnodes: cbfs-name: This is the name of the file created in CBFS. It defaults to the entry name (which is the node name), but you can override it with this property. cbfs-type: This is the CBFS file type. The following are supported: raw: This is a 'raw' file, although compression is supported. It can be used to store any file in CBFS. stage: This is an ELF file that has been loaded (i.e. mapped to memory), so appears in the CBFS as a flat binary. The input file must be an ELF image, for example this puts "u-boot" (the ELF image) into a 'stage' entry:: cbfs { size = <0x100000>; u-boot-elf { cbfs-name = "BOOT"; cbfs-type = "stage"; }; }; You can use your own ELF file with something like:: cbfs { size = <0x100000>; something { type = "blob"; filename = "cbfs-stage.elf"; cbfs-type = "stage"; }; }; As mentioned, the file is converted to a flat binary, so it is equivalent to adding "u-boot.bin", for example, but with the load and start addresses specified by the ELF. At present there is no option to add a flat binary with a load/start address, similar to the 'add-flat-binary' option in cbfstool. cbfs-offset: This is the offset of the file's data within the CBFS. It is used to specify where the file should be placed in cases where a fixed position is needed. Typical uses are for code which is not relocatable and must execute in-place from a particular address. This works because SPI flash is generally mapped into memory on x86 devices. The file header is placed before this offset so that the data start lines up exactly with the chosen offset. If this property is not provided, then the file is placed in the next available spot. The current implementation supports only a subset of CBFS features. It does not support other file types (e.g. payload), adding multiple files (like the 'files' entry with a pattern supported by binman), putting files at a particular offset in the CBFS and a few other things. Of course binman can create images containing multiple CBFSs, simply by defining these in the binman config:: binman { size = <0x800000>; cbfs { offset = <0x100000>; size = <0x100000>; u-boot { cbfs-type = "raw"; }; u-boot-dtb { cbfs-type = "raw"; }; }; cbfs2 { offset = <0x700000>; size = <0x100000>; u-boot { cbfs-type = "raw"; }; u-boot-dtb { cbfs-type = "raw"; }; image { type = "blob"; filename = "image.jpg"; }; }; }; This creates an 8MB image with two CBFSs, one at offset 1MB, one at 7MB, both of size 1MB. Entry: collection: An entry which contains a collection of other entries ------------------------------------------------------------------------ Properties / Entry arguments: - content: List of phandles to entries to include This allows reusing the contents of other entries. The contents of the listed entries are combined to form this entry. This serves as a useful base class for entry types which need to process data from elsewhere in the image, not necessarily child entries. Entry: cros-ec-rw: A blob entry which contains a Chromium OS read-write EC image -------------------------------------------------------------------------------- Properties / Entry arguments: - cros-ec-rw-path: Filename containing the EC image This entry holds a Chromium OS EC (embedded controller) image, for use in updating the EC on startup via software sync. Entry: fdtmap: An entry which contains an FDT map ------------------------------------------------- Properties / Entry arguments: None An FDT map is just a header followed by an FDT containing a list of all the entries in the image. The root node corresponds to the image node in the original FDT, and an image-name property indicates the image name in that original tree. The header is the string _FDTMAP_ followed by 8 unused bytes. When used, this entry will be populated with an FDT map which reflects the entries in the current image. Hierarchy is preserved, and all offsets and sizes are included. Note that the -u option must be provided to ensure that binman updates the FDT with the position of each entry. Example output for a simple image with U-Boot and an FDT map:: / { image-name = "binman"; size = <0x00000112>; image-pos = <0x00000000>; offset = <0x00000000>; u-boot { size = <0x00000004>; image-pos = <0x00000000>; offset = <0x00000000>; }; fdtmap { size = <0x0000010e>; image-pos = <0x00000004>; offset = <0x00000004>; }; }; If allow-repack is used then 'orig-offset' and 'orig-size' properties are added as necessary. See the binman README. When extracting files, an alternative 'fdt' format is available for fdtmaps. Use `binman extract -F fdt ...` to use this. It will export a devicetree, without the fdtmap header, so it can be viewed with `fdtdump`. Entry: files: A set of files arranged in a section -------------------------------------------------- Properties / Entry arguments: - pattern: Filename pattern to match the files to include - files-compress: Compression algorithm to use: none: No compression lz4: Use lz4 compression (via 'lz4' command-line utility) - files-align: Align each file to the given alignment This entry reads a number of files and places each in a separate sub-entry within this entry. To access these you need to enable device-tree updates at run-time so you can obtain the file positions. Entry: fill: An entry which is filled to a particular byte value ---------------------------------------------------------------- Properties / Entry arguments: - fill-byte: Byte to use to fill the entry Note that the size property must be set since otherwise this entry does not know how large it should be. You can often achieve the same effect using the pad-byte property of the overall image, in that the space between entries will then be padded with that byte. But this entry is sometimes useful for explicitly setting the byte value of a region. Entry: fit: Flat Image Tree (FIT) --------------------------------- This calls mkimage to create a FIT (U-Boot Flat Image Tree) based on the input provided. Nodes for the FIT should be written out in the binman configuration just as they would be in a file passed to mkimage. For example, this creates an image containing a FIT with U-Boot SPL:: binman { fit { description = "Test FIT"; fit,fdt-list = "of-list"; images { kernel@1 { description = "SPL"; os = "u-boot"; type = "rkspi"; arch = "arm"; compression = "none"; load = <0>; entry = <0>; u-boot-spl { }; }; }; }; }; More complex setups can be created, with generated nodes, as described below. Properties (in the 'fit' node itself) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Special properties have a `fit,` prefix, indicating that they should be processed but not included in the final FIT. The top-level 'fit' node supports the following special properties: fit,external-offset Indicates that the contents of the FIT are external and provides the external offset. This is passed to mkimage via the -E and -p flags. fit,fdt-list Indicates the entry argument which provides the list of device tree files for the gen-fdt-nodes operation (as below). This is often `of-list` meaning that `-a of-list="dtb1 dtb2..."` should be passed to binman. Substitutions ~~~~~~~~~~~~~ Node names and property values support a basic string-substitution feature. Available substitutions for '@' nodes (and property values) are: SEQ: Sequence number of the generated fdt (1, 2, ...) NAME Name of the dtb as provided (i.e. without adding '.dtb') The `default` property, if present, will be automatically set to the name if of configuration whose devicetree matches the `default-dt` entry argument, e.g. with `-a default-dt=sun50i-a64-pine64-lts`. Available substitutions for property values in these nodes are: DEFAULT-SEQ: Sequence number of the default fdt, as provided by the 'default-dt' entry argument Available operations ~~~~~~~~~~~~~~~~~~~~ You can add an operation to an '@' node to indicate which operation is required:: @fdt-SEQ { fit,operation = "gen-fdt-nodes"; ... }; Available operations are: gen-fdt-nodes Generate FDT nodes as above. This is the default if there is no `fit,operation` property. Generating nodes from an FDT list (gen-fdt-nodes) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ U-Boot supports creating fdt and config nodes automatically. To do this, pass an `of-list` property (e.g. `-a of-list=file1 file2`). This tells binman that you want to generates nodes for two files: `file1.dtb` and `file2.dtb`. The `fit,fdt-list` property (see above) indicates that `of-list` should be used. If the property is missing you will get an error. Then add a 'generator node', a node with a name starting with '@':: images { @fdt-SEQ { description = "fdt-NAME"; type = "flat_dt"; compression = "none"; }; }; This tells binman to create nodes `fdt-1` and `fdt-2` for each of your two files. All the properties you specify will be included in the node. This node acts like a template to generate the nodes. The generator node itself does not appear in the output - it is replaced with what binman generates. A 'data' property is created with the contents of the FDT file. You can create config nodes in a similar way:: configurations { default = "@config-DEFAULT-SEQ"; @config-SEQ { description = "NAME"; firmware = "atf"; loadables = "uboot"; fdt = "fdt-SEQ"; }; }; This tells binman to create nodes `config-1` and `config-2`, i.e. a config for each of your two files. Note that if no devicetree files are provided (with '-a of-list' as above) then no nodes will be generated. Entry: fmap: An entry which contains an Fmap section ---------------------------------------------------- Properties / Entry arguments: None FMAP is a simple format used by flashrom, an open-source utility for reading and writing the SPI flash, typically on x86 CPUs. The format provides flashrom with a list of areas, so it knows what it in the flash. It can then read or write just a single area, instead of the whole flash. The format is defined by the flashrom project, in the file lib/fmap.h - see www.flashrom.org/Flashrom for more information. When used, this entry will be populated with an FMAP which reflects the entries in the current image. Note that any hierarchy is squashed, since FMAP does not support this. Sections are represented as an area appearing before its contents, so that it is possible to reconstruct the hierarchy from the FMAP by using the offset information. This convention does not seem to be documented, but is used in Chromium OS. CBFS entries appear as a single entry, i.e. the sub-entries are ignored. Entry: gbb: An entry which contains a Chromium OS Google Binary Block --------------------------------------------------------------------- Properties / Entry arguments: - hardware-id: Hardware ID to use for this build (a string) - keydir: Directory containing the public keys to use - bmpblk: Filename containing images used by recovery Chromium OS uses a GBB to store various pieces of information, in particular the root and recovery keys that are used to verify the boot process. Some more details are here: https://www.chromium.org/chromium-os/firmware-porting-guide/2-concepts but note that the page dates from 2013 so is quite out of date. See README.chromium for how to obtain the required keys and tools. Entry: image-header: An entry which contains a pointer to the FDT map --------------------------------------------------------------------- Properties / Entry arguments: location: Location of header ("start" or "end" of image). This is optional. If omitted then the entry must have an offset property. This adds an 8-byte entry to the start or end of the image, pointing to the location of the FDT map. The format is a magic number followed by an offset from the start or end of the image, in twos-compliment format. This entry must be in the top-level part of the image. NOTE: If the location is at the start/end, you will probably need to specify sort-by-offset for the image, unless you actually put the image header first/last in the entry list. Entry: intel-cmc: Intel Chipset Micro Code (CMC) file ----------------------------------------------------- Properties / Entry arguments: - filename: Filename of file to read into entry This file contains microcode for some devices in a special format. An example filename is 'Microcode/C0_22211.BIN'. See README.x86 for information about x86 binary blobs. Entry: intel-descriptor: Intel flash descriptor block (4KB) ----------------------------------------------------------- Properties / Entry arguments: filename: Filename of file containing the descriptor. This is typically a 4KB binary file, sometimes called 'descriptor.bin' This entry is placed at the start of flash and provides information about the SPI flash regions. In particular it provides the base address and size of the ME (Management Engine) region, allowing us to place the ME binary in the right place. With this entry in your image, the position of the 'intel-me' entry will be fixed in the image, which avoids you needed to specify an offset for that region. This is useful, because it is not possible to change the position of the ME region without updating the descriptor. See README.x86 for information about x86 binary blobs. Entry: intel-fit: Intel Firmware Image Table (FIT) -------------------------------------------------- This entry contains a dummy FIT as required by recent Intel CPUs. The FIT contains information about the firmware and microcode available in the image. At present binman only supports a basic FIT with no microcode. Entry: intel-fit-ptr: Intel Firmware Image Table (FIT) pointer -------------------------------------------------------------- This entry contains a pointer to the FIT. It is required to be at address 0xffffffc0 in the image. Entry: intel-fsp: Intel Firmware Support Package (FSP) file ----------------------------------------------------------- Properties / Entry arguments: - filename: Filename of file to read into entry This file contains binary blobs which are used on some devices to make the platform work. U-Boot executes this code since it is not possible to set up the hardware using U-Boot open-source code. Documentation is typically not available in sufficient detail to allow this. An example filename is 'FSP/QUEENSBAY_FSP_GOLD_001_20-DECEMBER-2013.fd' See README.x86 for information about x86 binary blobs. Entry: intel-fsp-m: Intel Firmware Support Package (FSP) memory init -------------------------------------------------------------------- Properties / Entry arguments: - filename: Filename of file to read into entry This file contains a binary blob which is used on some devices to set up SDRAM. U-Boot executes this code in SPL so that it can make full use of memory. Documentation is typically not available in sufficient detail to allow U-Boot do this this itself.. An example filename is 'fsp_m.bin' See README.x86 for information about x86 binary blobs. Entry: intel-fsp-s: Intel Firmware Support Package (FSP) silicon init --------------------------------------------------------------------- Properties / Entry arguments: - filename: Filename of file to read into entry This file contains a binary blob which is used on some devices to set up the silicon. U-Boot executes this code in U-Boot proper after SDRAM is running, so that it can make full use of memory. Documentation is typically not available in sufficient detail to allow U-Boot do this this itself. An example filename is 'fsp_s.bin' See README.x86 for information about x86 binary blobs. Entry: intel-fsp-t: Intel Firmware Support Package (FSP) temp ram init ---------------------------------------------------------------------- Properties / Entry arguments: - filename: Filename of file to read into entry This file contains a binary blob which is used on some devices to set up temporary memory (Cache-as-RAM or CAR). U-Boot executes this code in TPL so that it has access to memory for its stack and initial storage. An example filename is 'fsp_t.bin' See README.x86 for information about x86 binary blobs. Entry: intel-ifwi: Intel Integrated Firmware Image (IFWI) file -------------------------------------------------------------- Properties / Entry arguments: - filename: Filename of file to read into entry. This is either the IFWI file itself, or a file that can be converted into one using a tool - convert-fit: If present this indicates that the ifwitool should be used to convert the provided file into a IFWI. This file contains code and data used by the SoC that is required to make it work. It includes U-Boot TPL, microcode, things related to the CSE (Converged Security Engine, the microcontroller that loads all the firmware) and other items beyond the wit of man. A typical filename is 'ifwi.bin' for an IFWI file, or 'fitimage.bin' for a file that will be converted to an IFWI. The position of this entry is generally set by the intel-descriptor entry. The contents of the IFWI are specified by the subnodes of the IFWI node. Each subnode describes an entry which is placed into the IFWFI with a given sub-partition (and optional entry name). Properties for subnodes: - ifwi-subpart: sub-parition to put this entry into, e.g. "IBBP" - ifwi-entry: entry name t use, e.g. "IBBL" - ifwi-replace: if present, indicates that the item should be replaced in the IFWI. Otherwise it is added. See README.x86 for information about x86 binary blobs. Entry: intel-me: Intel Management Engine (ME) file -------------------------------------------------- Properties / Entry arguments: - filename: Filename of file to read into entry This file contains code used by the SoC that is required to make it work. The Management Engine is like a background task that runs things that are not clearly documented, but may include keyboard, display and network access. For platform that use ME it is not possible to disable it. U-Boot does not directly execute code in the ME binary. A typical filename is 'me.bin'. The position of this entry is generally set by the intel-descriptor entry. See README.x86 for information about x86 binary blobs. Entry: intel-mrc: Intel Memory Reference Code (MRC) file -------------------------------------------------------- Properties / Entry arguments: - filename: Filename of file to read into entry This file contains code for setting up the SDRAM on some Intel systems. This is executed by U-Boot when needed early during startup. A typical filename is 'mrc.bin'. See README.x86 for information about x86 binary blobs. Entry: intel-refcode: Intel Reference Code file ----------------------------------------------- Properties / Entry arguments: - filename: Filename of file to read into entry This file contains code for setting up the platform on some Intel systems. This is executed by U-Boot when needed early during startup. A typical filename is 'refcode.bin'. See README.x86 for information about x86 binary blobs. Entry: intel-vbt: Intel Video BIOS Table (VBT) file --------------------------------------------------- Properties / Entry arguments: - filename: Filename of file to read into entry This file contains code that sets up the integrated graphics subsystem on some Intel SoCs. U-Boot executes this when the display is started up. See README.x86 for information about Intel binary blobs. Entry: intel-vga: Intel Video Graphics Adaptor (VGA) file --------------------------------------------------------- Properties / Entry arguments: - filename: Filename of file to read into entry This file contains code that sets up the integrated graphics subsystem on some Intel SoCs. U-Boot executes this when the display is started up. This is similar to the VBT file but in a different format. See README.x86 for information about Intel binary blobs. Entry: mkimage: Binary produced by mkimage ------------------------------------------ Properties / Entry arguments: - datafile: Filename for -d argument - args: Other arguments to pass The data passed to mkimage is collected from subnodes of the mkimage node, e.g.:: mkimage { args = "-n test -T imximage"; u-boot-spl { }; }; This calls mkimage to create an imximage with u-boot-spl.bin as the input file. The output from mkimage then becomes part of the image produced by binman. To use CONFIG options in the arguments, use a string list instead, as in this example which also produces four arguments:: mkimage { args = "-n", CONFIG_SYS_SOC, "-T imximage"; u-boot-spl { }; }; Entry: opensbi: RISC-V OpenSBI fw_dynamic blob ---------------------------------------------- Properties / Entry arguments: - opensbi-path: Filename of file to read into entry. This is typically called fw_dynamic.bin This entry holds the run-time firmware, typically started by U-Boot SPL. See the U-Boot README for your architecture or board for how to use it. See https://github.com/riscv/opensbi for more information about OpenSBI. Entry: powerpc-mpc85xx-bootpg-resetvec: PowerPC mpc85xx bootpg + resetvec code for U-Boot ----------------------------------------------------------------------------------------- Properties / Entry arguments: - filename: Filename of u-boot-br.bin (default 'u-boot-br.bin') This entry is valid for PowerPC mpc85xx cpus. This entry holds 'bootpg + resetvec' code for PowerPC mpc85xx CPUs which needs to be placed at offset 'RESET_VECTOR_ADDRESS - 0xffc'. Entry: scp: System Control Processor (SCP) firmware blob -------------------------------------------------------- Properties / Entry arguments: - scp-path: Filename of file to read into the entry, typically scp.bin This entry holds firmware for an external platform-specific coprocessor. Entry: section: Entry that contains other entries ------------------------------------------------- A section is an entry which can contain other entries, thus allowing hierarchical images to be created. See 'Sections and hierarchical images' in the binman README for more information. The base implementation simply joins the various entries together, using various rules about alignment, etc. Subclassing ~~~~~~~~~~~ This class can be subclassed to support other file formats which hold multiple entries, such as CBFS. To do this, override the following functions. The documentation here describes what your function should do. For example code, see etypes which subclass `Entry_section`, or `cbfs.py` for a more involved example:: $ grep -l \(Entry_section tools/binman/etype/*.py ReadNode() Call `super().ReadNode()`, then read any special properties for the section. Then call `self.ReadEntries()` to read the entries. Binman calls this at the start when reading the image description. ReadEntries() Read in the subnodes of the section. This may involve creating entries of a particular etype automatically, as well as reading any special properties in the entries. For each entry, entry.ReadNode() should be called, to read the basic entry properties. The properties should be added to `self._entries[]`, in the correct order, with a suitable name. Binman calls this at the start when reading the image description. BuildSectionData(required) Create the custom file format that you want and return it as bytes. This likely sets up a file header, then loops through the entries, adding them to the file. For each entry, call `entry.GetData()` to obtain the data. If that returns None, and `required` is False, then this method must give up and return None. But if `required` is True then it should assume that all data is valid. Binman calls this when packing the image, to find out the size of everything. It is called again at the end when building the final image. SetImagePos(image_pos): Call `super().SetImagePos(image_pos)`, then set the `image_pos` values for each of the entries. This should use the custom file format to find the `start offset` (and `image_pos`) of each entry. If the file format uses compression in such a way that there is no offset available (other than reading the whole file and decompressing it), then the offsets for affected entries can remain unset (`None`). The size should also be set if possible. Binman calls this after the image has been packed, to update the location that all the entries ended up at. ReadChildData(child, decomp, alt_format): The default version of this may be good enough, if you are able to implement SetImagePos() correctly. But that is a bit of a bypass, so you can override this method to read from your custom file format. It should read the entire entry containing the custom file using `super().ReadData(True)`, then parse the file to get the data for the given child, then return that data. If your file format supports compression, the `decomp` argument tells you whether to return the compressed data (`decomp` is False) or to uncompress it first, then return the uncompressed data (`decomp` is True). This is used by the `binman extract -U` option. If your entry supports alternative formats, the alt_format provides the alternative format that the user has selected. Your function should return data in that format. This is used by the 'binman extract -l' option. Binman calls this when reading in an image, in order to populate all the entries with the data from that image (`binman ls`). WriteChildData(child): Binman calls this after `child.data` is updated, to inform the custom file format about this, in case it needs to do updates. The default version of this does nothing and probably needs to be overridden for the 'binman replace' command to work. Your version should use `child.data` to update the data for that child in the custom file format. Binman calls this when updating an image that has been read in and in particular to update the data for a particular entry (`binman replace`) Properties / Entry arguments ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ See :ref:`develop/package/binman:Image description format` for more information. align-default Default alignment for this section, if no alignment is given in the entry pad-byte Pad byte to use when padding sort-by-offset True if entries should be sorted by offset, False if they must be in-order in the device tree description end-at-4gb Used to build an x86 ROM which ends at 4GB (2^32) name-prefix Adds a prefix to the name of every entry in the section when writing out the map skip-at-start Number of bytes before the first entry starts. These effectively adjust the starting offset of entries. For example, if this is 16, then the first entry would start at 16. An entry with offset = 20 would in fact be written at offset 4 in the image file, since the first 16 bytes are skipped when writing. Since a section is also an entry, it inherits all the properies of entries too. Note that the `allow_missing` member controls whether this section permits external blobs to be missing their contents. The option will produce an image but of course it will not work. It is useful to make sure that Continuous Integration systems can build without the binaries being available. This is set by the `SetAllowMissing()` method, if `--allow-missing` is passed to binman. Entry: tee-os: Entry containing an OP-TEE Trusted OS (TEE) blob --------------------------------------------------------------- Properties / Entry arguments: - tee-os-path: Filename of file to read into entry. This is typically called tee-pager.bin This entry holds the run-time firmware, typically started by U-Boot SPL. See the U-Boot README for your architecture or board for how to use it. See https://github.com/OP-TEE/optee_os for more information about OP-TEE. Entry: text: An entry which contains text ----------------------------------------- The text can be provided either in the node itself or by a command-line argument. There is a level of indirection to allow multiple text strings and sharing of text. Properties / Entry arguments: text-label: The value of this string indicates the property / entry-arg that contains the string to place in the entry (actual name is the value of text-label): contains the string to place in the entry. : The text to place in the entry (overrides the above mechanism). This is useful when the text is constant. Example node:: text { size = <50>; text-label = "message"; }; You can then use: binman -amessage="this is my message" and binman will insert that string into the entry. It is also possible to put the string directly in the node:: text { size = <8>; text-label = "message"; message = "a message directly in the node" }; or just:: text { size = <8>; text = "some text directly in the node" }; The text is not itself nul-terminated. This can be achieved, if required, by setting the size of the entry to something larger than the text. Entry: u-boot: U-Boot flat binary --------------------------------- Properties / Entry arguments: - filename: Filename of u-boot.bin (default 'u-boot.bin') This is the U-Boot binary, containing relocation information to allow it to relocate itself at runtime. The binary typically includes a device tree blob at the end of it. U-Boot can access binman symbols at runtime. See: 'Access to binman entry offsets at run time (fdt)' in the binman README for more information. Note that this entry is automatically replaced with u-boot-expanded unless --no-expanded is used or the node has a 'no-expanded' property. Entry: u-boot-dtb: U-Boot device tree ------------------------------------- Properties / Entry arguments: - filename: Filename of u-boot.dtb (default 'u-boot.dtb') This is the U-Boot device tree, containing configuration information for U-Boot. U-Boot needs this to know what devices are present and which drivers to activate. Note: This is mostly an internal entry type, used by others. This allows binman to know which entries contain a device tree. Entry: u-boot-dtb-with-ucode: A U-Boot device tree file, with the microcode removed ----------------------------------------------------------------------------------- Properties / Entry arguments: - filename: Filename of u-boot.dtb (default 'u-boot.dtb') See Entry_u_boot_ucode for full details of the three entries involved in this process. This entry provides the U-Boot device-tree file, which contains the microcode. If the microcode is not being collated into one place then the offset and size of the microcode is recorded by this entry, for use by u-boot-with-ucode_ptr. If it is being collated, then this entry deletes the microcode from the device tree (to save space) and makes it available to u-boot-ucode. Entry: u-boot-elf: U-Boot ELF image ----------------------------------- Properties / Entry arguments: - filename: Filename of u-boot (default 'u-boot') This is the U-Boot ELF image. It does not include a device tree but can be relocated to any address for execution. Entry: u-boot-env: An entry which contains a U-Boot environment --------------------------------------------------------------- Properties / Entry arguments: - filename: File containing the environment text, with each line in the form var=value Entry: u-boot-expanded: U-Boot flat binary broken out into its component parts ------------------------------------------------------------------------------ This is a section containing the U-Boot binary and a devicetree. Using this entry type automatically creates this section, with the following entries in it: u-boot-nodtb u-boot-dtb Having the devicetree separate allows binman to update it in the final image, so that the entries positions are provided to the running U-Boot. Entry: u-boot-img: U-Boot legacy image -------------------------------------- Properties / Entry arguments: - filename: Filename of u-boot.img (default 'u-boot.img') This is the U-Boot binary as a packaged image, in legacy format. It has a header which allows it to be loaded at the correct address for execution. You should use FIT (Flat Image Tree) instead of the legacy image for new applications. Entry: u-boot-nodtb: U-Boot flat binary without device tree appended -------------------------------------------------------------------- Properties / Entry arguments: - filename: Filename to include (default 'u-boot-nodtb.bin') This is the U-Boot binary, containing relocation information to allow it to relocate itself at runtime. It does not include a device tree blob at the end of it so normally cannot work without it. You can add a u-boot-dtb entry after this one, or use a u-boot entry instead, normally expands to a section containing u-boot and u-boot-dtb Entry: u-boot-spl: U-Boot SPL binary ------------------------------------ Properties / Entry arguments: - filename: Filename of u-boot-spl.bin (default 'spl/u-boot-spl.bin') This is the U-Boot SPL (Secondary Program Loader) binary. This is a small binary which loads before U-Boot proper, typically into on-chip SRAM. It is responsible for locating, loading and jumping to U-Boot. Note that SPL is not relocatable so must be loaded to the correct address in SRAM, or written to run from the correct address if direct flash execution is possible (e.g. on x86 devices). SPL can access binman symbols at runtime. See: 'Access to binman entry offsets at run time (symbols)' in the binman README for more information. The ELF file 'spl/u-boot-spl' must also be available for this to work, since binman uses that to look up symbols to write into the SPL binary. Note that this entry is automatically replaced with u-boot-spl-expanded unless --no-expanded is used or the node has a 'no-expanded' property. Entry: u-boot-spl-bss-pad: U-Boot SPL binary padded with a BSS region --------------------------------------------------------------------- Properties / Entry arguments: None This holds the padding added after the SPL binary to cover the BSS (Block Started by Symbol) region. This region holds the various variables used by SPL. It is set to 0 by SPL when it starts up. If you want to append data to the SPL image (such as a device tree file), you must pad out the BSS region to avoid the data overlapping with U-Boot variables. This entry is useful in that case. It automatically pads out the entry size to cover both the code, data and BSS. The contents of this entry will a certain number of zero bytes, determined by __bss_size The ELF file 'spl/u-boot-spl' must also be available for this to work, since binman uses that to look up the BSS address. Entry: u-boot-spl-dtb: U-Boot SPL device tree --------------------------------------------- Properties / Entry arguments: - filename: Filename of u-boot.dtb (default 'spl/u-boot-spl.dtb') This is the SPL device tree, containing configuration information for SPL. SPL needs this to know what devices are present and which drivers to activate. Entry: u-boot-spl-elf: U-Boot SPL ELF image ------------------------------------------- Properties / Entry arguments: - filename: Filename of SPL u-boot (default 'spl/u-boot-spl') This is the U-Boot SPL ELF image. It does not include a device tree but can be relocated to any address for execution. Entry: u-boot-spl-expanded: U-Boot SPL flat binary broken out into its component parts -------------------------------------------------------------------------------------- Properties / Entry arguments: - spl-dtb: Controls whether this entry is selected (set to 'y' or '1' to select) This is a section containing the U-Boot binary, BSS padding if needed and a devicetree. Using this entry type automatically creates this section, with the following entries in it: u-boot-spl-nodtb u-boot-spl-bss-pad u-boot-dtb Having the devicetree separate allows binman to update it in the final image, so that the entries positions are provided to the running U-Boot. This entry is selected based on the value of the 'spl-dtb' entryarg. If this is non-empty (and not 'n' or '0') then this expanded entry is selected. Entry: u-boot-spl-nodtb: SPL binary without device tree appended ---------------------------------------------------------------- Properties / Entry arguments: - filename: Filename to include (default 'spl/u-boot-spl-nodtb.bin') This is the U-Boot SPL binary, It does not include a device tree blob at the end of it so may not be able to work without it, assuming SPL needs a device tree to operate on your platform. You can add a u-boot-spl-dtb entry after this one, or use a u-boot-spl entry instead' which normally expands to a section containing u-boot-spl-dtb, u-boot-spl-bss-pad and u-boot-spl-dtb SPL can access binman symbols at runtime. See: 'Access to binman entry offsets at run time (symbols)' in the binman README for more information. The ELF file 'spl/u-boot-spl' must also be available for this to work, since binman uses that to look up symbols to write into the SPL binary. Entry: u-boot-spl-with-ucode-ptr: U-Boot SPL with embedded microcode pointer ---------------------------------------------------------------------------- This is used when SPL must set up the microcode for U-Boot. See Entry_u_boot_ucode for full details of the entries involved in this process. Entry: u-boot-tpl: U-Boot TPL binary ------------------------------------ Properties / Entry arguments: - filename: Filename of u-boot-tpl.bin (default 'tpl/u-boot-tpl.bin') This is the U-Boot TPL (Tertiary Program Loader) binary. This is a small binary which loads before SPL, typically into on-chip SRAM. It is responsible for locating, loading and jumping to SPL, the next-stage loader. Note that SPL is not relocatable so must be loaded to the correct address in SRAM, or written to run from the correct address if direct flash execution is possible (e.g. on x86 devices). SPL can access binman symbols at runtime. See: 'Access to binman entry offsets at run time (symbols)' in the binman README for more information. The ELF file 'tpl/u-boot-tpl' must also be available for this to work, since binman uses that to look up symbols to write into the TPL binary. Note that this entry is automatically replaced with u-boot-tpl-expanded unless --no-expanded is used or the node has a 'no-expanded' property. Entry: u-boot-tpl-bss-pad: U-Boot TPL binary padded with a BSS region --------------------------------------------------------------------- Properties / Entry arguments: None This holds the padding added after the TPL binary to cover the BSS (Block Started by Symbol) region. This region holds the various variables used by TPL. It is set to 0 by TPL when it starts up. If you want to append data to the TPL image (such as a device tree file), you must pad out the BSS region to avoid the data overlapping with U-Boot variables. This entry is useful in that case. It automatically pads out the entry size to cover both the code, data and BSS. The contents of this entry will a certain number of zero bytes, determined by __bss_size The ELF file 'tpl/u-boot-tpl' must also be available for this to work, since binman uses that to look up the BSS address. Entry: u-boot-tpl-dtb: U-Boot TPL device tree --------------------------------------------- Properties / Entry arguments: - filename: Filename of u-boot.dtb (default 'tpl/u-boot-tpl.dtb') This is the TPL device tree, containing configuration information for TPL. TPL needs this to know what devices are present and which drivers to activate. Entry: u-boot-tpl-dtb-with-ucode: U-Boot TPL with embedded microcode pointer ---------------------------------------------------------------------------- This is used when TPL must set up the microcode for U-Boot. See Entry_u_boot_ucode for full details of the entries involved in this process. Entry: u-boot-tpl-elf: U-Boot TPL ELF image ------------------------------------------- Properties / Entry arguments: - filename: Filename of TPL u-boot (default 'tpl/u-boot-tpl') This is the U-Boot TPL ELF image. It does not include a device tree but can be relocated to any address for execution. Entry: u-boot-tpl-expanded: U-Boot TPL flat binary broken out into its component parts -------------------------------------------------------------------------------------- Properties / Entry arguments: - tpl-dtb: Controls whether this entry is selected (set to 'y' or '1' to select) This is a section containing the U-Boot binary, BSS padding if needed and a devicetree. Using this entry type automatically creates this section, with the following entries in it: u-boot-tpl-nodtb u-boot-tpl-bss-pad u-boot-dtb Having the devicetree separate allows binman to update it in the final image, so that the entries positions are provided to the running U-Boot. This entry is selected based on the value of the 'tpl-dtb' entryarg. If this is non-empty (and not 'n' or '0') then this expanded entry is selected. Entry: u-boot-tpl-nodtb: TPL binary without device tree appended ---------------------------------------------------------------- Properties / Entry arguments: - filename: Filename to include (default 'tpl/u-boot-tpl-nodtb.bin') This is the U-Boot TPL binary, It does not include a device tree blob at the end of it so may not be able to work without it, assuming TPL needs a device tree to operate on your platform. You can add a u-boot-tpl-dtb entry after this one, or use a u-boot-tpl entry instead, which normally expands to a section containing u-boot-tpl-dtb, u-boot-tpl-bss-pad and u-boot-tpl-dtb TPL can access binman symbols at runtime. See: 'Access to binman entry offsets at run time (symbols)' in the binman README for more information. The ELF file 'tpl/u-boot-tpl' must also be available for this to work, since binman uses that to look up symbols to write into the TPL binary. Entry: u-boot-tpl-with-ucode-ptr: U-Boot TPL with embedded microcode pointer ---------------------------------------------------------------------------- See Entry_u_boot_ucode for full details of the entries involved in this process. Entry: u-boot-ucode: U-Boot microcode block ------------------------------------------- Properties / Entry arguments: None The contents of this entry are filled in automatically by other entries which must also be in the image. U-Boot on x86 needs a single block of microcode. This is collected from the various microcode update nodes in the device tree. It is also unable to read the microcode from the device tree on platforms that use FSP (Firmware Support Package) binaries, because the API requires that the microcode is supplied before there is any SRAM available to use (i.e. the FSP sets up the SRAM / cache-as-RAM but does so in the call that requires the microcode!). To keep things simple, all x86 platforms handle microcode the same way in U-Boot (even non-FSP platforms). This is that a table is placed at _dt_ucode_base_size containing the base address and size of the microcode. This is either passed to the FSP (for FSP platforms), or used to set up the microcode (for non-FSP platforms). This all happens in the build system since it is the only way to get the microcode into a single blob and accessible without SRAM. There are two cases to handle. If there is only one microcode blob in the device tree, then the ucode pointer it set to point to that. This entry (u-boot-ucode) is empty. If there is more than one update, then this entry holds the concatenation of all updates, and the device tree entry (u-boot-dtb-with-ucode) is updated to remove the microcode. This last step ensures that that the microcode appears in one contiguous block in the image and is not unnecessarily duplicated in the device tree. It is referred to as 'collation' here. Entry types that have a part to play in handling microcode: Entry_u_boot_with_ucode_ptr: Contains u-boot-nodtb.bin (i.e. U-Boot without the device tree). It updates it with the address and size of the microcode so that U-Boot can find it early on start-up. Entry_u_boot_dtb_with_ucode: Contains u-boot.dtb. It stores the microcode in a 'self.ucode_data' property, which is then read by this class to obtain the microcode if needed. If collation is performed, it removes the microcode from the device tree. Entry_u_boot_ucode: This class. If collation is enabled it reads the microcode from the Entry_u_boot_dtb_with_ucode entry, and uses it as the contents of this entry. Entry: u-boot-with-ucode-ptr: U-Boot with embedded microcode pointer -------------------------------------------------------------------- Properties / Entry arguments: - filename: Filename of u-boot-nodtb.bin (default 'u-boot-nodtb.bin') - optional-ucode: boolean property to make microcode optional. If the u-boot.bin image does not include microcode, no error will be generated. See Entry_u_boot_ucode for full details of the three entries involved in this process. This entry updates U-Boot with the offset and size of the microcode, to allow early x86 boot code to find it without doing anything complicated. Otherwise it is the same as the u-boot entry. Entry: vblock: An entry which contains a Chromium OS verified boot block ------------------------------------------------------------------------ Properties / Entry arguments: - content: List of phandles to entries to sign - keydir: Directory containing the public keys to use - keyblock: Name of the key file to use (inside keydir) - signprivate: Name of provide key file to use (inside keydir) - version: Version number of the vblock (typically 1) - kernelkey: Name of the kernel key to use (inside keydir) - preamble-flags: Value of the vboot preamble flags (typically 0) Output files: - input. - input file passed to futility - vblock. - output file generated by futility (which is used as the entry contents) Chromium OS signs the read-write firmware and kernel, writing the signature in this block. This allows U-Boot to verify that the next firmware stage and kernel are genuine. Entry: x86-reset16: x86 16-bit reset code for U-Boot ---------------------------------------------------- Properties / Entry arguments: - filename: Filename of u-boot-x86-reset16.bin (default 'u-boot-x86-reset16.bin') x86 CPUs start up in 16-bit mode, even if they are 32-bit CPUs. This code must be placed at a particular address. This entry holds that code. It is typically placed at offset CONFIG_RESET_VEC_LOC. The code is responsible for jumping to the x86-start16 code, which continues execution. For 64-bit U-Boot, the 'x86_reset16_spl' entry type is used instead. Entry: x86-reset16-spl: x86 16-bit reset code for U-Boot -------------------------------------------------------- Properties / Entry arguments: - filename: Filename of u-boot-x86-reset16.bin (default 'u-boot-x86-reset16.bin') x86 CPUs start up in 16-bit mode, even if they are 32-bit CPUs. This code must be placed at a particular address. This entry holds that code. It is typically placed at offset CONFIG_RESET_VEC_LOC. The code is responsible for jumping to the x86-start16 code, which continues execution. For 32-bit U-Boot, the 'x86_reset_spl' entry type is used instead. Entry: x86-reset16-tpl: x86 16-bit reset code for U-Boot -------------------------------------------------------- Properties / Entry arguments: - filename: Filename of u-boot-x86-reset16.bin (default 'u-boot-x86-reset16.bin') x86 CPUs start up in 16-bit mode, even if they are 32-bit CPUs. This code must be placed at a particular address. This entry holds that code. It is typically placed at offset CONFIG_RESET_VEC_LOC. The code is responsible for jumping to the x86-start16 code, which continues execution. For 32-bit U-Boot, the 'x86_reset_tpl' entry type is used instead. Entry: x86-start16: x86 16-bit start-up code for U-Boot ------------------------------------------------------- Properties / Entry arguments: - filename: Filename of u-boot-x86-start16.bin (default 'u-boot-x86-start16.bin') x86 CPUs start up in 16-bit mode, even if they are 32-bit CPUs. This code must be placed in the top 64KB of the ROM. The reset code jumps to it. This entry holds that code. It is typically placed at offset CONFIG_SYS_X86_START16. The code is responsible for changing to 32-bit mode and jumping to U-Boot's entry point, which requires 32-bit mode (for 32-bit U-Boot). For 64-bit U-Boot, the 'x86_start16_spl' entry type is used instead. Entry: x86-start16-spl: x86 16-bit start-up code for SPL -------------------------------------------------------- Properties / Entry arguments: - filename: Filename of spl/u-boot-x86-start16-spl.bin (default 'spl/u-boot-x86-start16-spl.bin') x86 CPUs start up in 16-bit mode, even if they are 32-bit CPUs. This code must be placed in the top 64KB of the ROM. The reset code jumps to it. This entry holds that code. It is typically placed at offset CONFIG_SYS_X86_START16. The code is responsible for changing to 32-bit mode and jumping to U-Boot's entry point, which requires 32-bit mode (for 32-bit U-Boot). For 32-bit U-Boot, the 'x86-start16' entry type is used instead. Entry: x86-start16-tpl: x86 16-bit start-up code for TPL -------------------------------------------------------- Properties / Entry arguments: - filename: Filename of tpl/u-boot-x86-start16-tpl.bin (default 'tpl/u-boot-x86-start16-tpl.bin') x86 CPUs start up in 16-bit mode, even if they are 32-bit CPUs. This code must be placed in the top 64KB of the ROM. The reset code jumps to it. This entry holds that code. It is typically placed at offset CONFIG_SYS_X86_START16. The code is responsible for changing to 32-bit mode and jumping to U-Boot's entry point, which requires 32-bit mode (for 32-bit U-Boot). If TPL is not being used, the 'x86-start16-spl or 'x86-start16' entry types may be used instead.