mirror of
https://github.com/AsahiLinux/u-boot
synced 2024-11-14 00:47:26 +00:00
4fce6554dc
Add a section describing the secure boot image used on Keystone2 secure devices. Signed-off-by: Madan Srinivas <madans@ti.com> Signed-off-by: Andrew F. Davis <afd@ti.com> Reviewed-by: Tom Rini <trini@konsulko.com> Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
212 lines
9.7 KiB
Text
212 lines
9.7 KiB
Text
README on how boot images are created for secure TI devices
|
|
|
|
CONFIG_TI_SECURE_DEVICE:
|
|
Secure TI devices require a boot image that is authenticated by ROM
|
|
code to function. Without this, even JTAG remains locked and the
|
|
device is essentially useless. In order to create a valid boot image for
|
|
a secure device from TI, the initial public software image must be signed
|
|
and combined with various headers, certificates, and other binary images.
|
|
|
|
Information on the details on the complete boot image format can be obtained
|
|
from Texas Instruments. The tools used to generate boot images for secure
|
|
devices are part of a secure development package (SECDEV) that can be
|
|
downloaded from:
|
|
|
|
http://www.ti.com/mysecuresoftware (login required)
|
|
|
|
The secure development package is access controlled due to NDA and export
|
|
control restrictions. Access must be requested and granted by TI before the
|
|
package is viewable and downloadable. Contact TI, either online or by way
|
|
of a local TI representative, to request access.
|
|
|
|
Booting of U-Boot SPL
|
|
=====================
|
|
|
|
When CONFIG_TI_SECURE_DEVICE is set, the U-Boot SPL build process
|
|
requires the presence and use of these tools in order to create a
|
|
viable boot image. The build process will look for the environment
|
|
variable TI_SECURE_DEV_PKG, which should be the path of the installed
|
|
SECDEV package. If the TI_SECURE_DEV_PKG variable is not defined or
|
|
if it is defined but doesn't point to a valid SECDEV package, a
|
|
warning is issued during the build to indicate that a final secure
|
|
bootable image was not created.
|
|
|
|
Within the SECDEV package exists an image creation script:
|
|
|
|
${TI_SECURE_DEV_PKG}/scripts/create-boot-image.sh
|
|
|
|
This is called as part of the SPL/u-boot build process. As the secure
|
|
boot image formats and requirements differ between secure SOC from TI,
|
|
the purpose of this script is to abstract these details as much as
|
|
possible.
|
|
|
|
The script is basically the only required interface to the TI SECDEV
|
|
package for creating a bootable SPL image for secure TI devices.
|
|
|
|
Invoking the script for AM33xx Secure Devices
|
|
=============================================
|
|
|
|
create-boot-image.sh \
|
|
<IMAGE_FLAG> <INPUT_FILE> <OUTPUT_FILE> <SPL_LOAD_ADDR>
|
|
|
|
<IMAGE_FLAG> is a value that specifies the type of the image to
|
|
generate OR the action the image generation tool will take. Valid
|
|
values are:
|
|
SPI_X-LOADER - Generates an image for SPI flash (byte swapped)
|
|
X-LOADER - Generates an image for non-XIP flash
|
|
MLO - Generates an image for SD/MMC/eMMC media
|
|
2ND - Generates an image for USB, UART and Ethernet
|
|
XIP_X-LOADER - Generates a single stage u-boot for NOR/QSPI XiP
|
|
|
|
<INPUT_FILE> is the full path and filename of the public world boot
|
|
loaderbinary file (depending on the boot media, this is usually
|
|
either u-boot-spl.bin or u-boot.bin).
|
|
|
|
<OUTPUT_FILE> is the full path and filename of the final secure
|
|
image. The output binary images should be used in place of the standard
|
|
non-secure binary images (see the platform-specific user's guides and
|
|
releases notes for how the non-secure images are typically used)
|
|
u-boot-spl_HS_SPI_X-LOADER - byte swapped boot image for SPI flash
|
|
u-boot-spl_HS_X-LOADER - boot image for NAND or SD/MMC/eMMC rawmode
|
|
u-boot-spl_HS_MLO - boot image for SD/MMC/eMMC media
|
|
u-boot-spl_HS_2ND - boot image for USB, UART and Ethernet
|
|
u-boot_HS_XIP_X-LOADER - boot image for NOR or QSPI Xip flash
|
|
|
|
<SPL_LOAD_ADDR> is the address at which SOC ROM should load the
|
|
<INPUT_FILE>
|
|
|
|
Invoking the script for AM43xx Secure Devices
|
|
=============================================
|
|
|
|
create-boot-image.sh \
|
|
<IMAGE_FLAG> <INPUT_FILE> <OUTPUT_FILE> <SPL_LOAD_ADDR>
|
|
|
|
<IMAGE_FLAG> is a value that specifies the type of the image to
|
|
generate OR the action the image generation tool will take. Valid
|
|
values are:
|
|
SPI_X-LOADER - Generates an image for SPI flash (byte
|
|
swapped)
|
|
XIP_X-LOADER - Generates a single stage u-boot for
|
|
NOR/QSPI XiP
|
|
ISSW - Generates an image for all other boot modes
|
|
|
|
<INPUT_FILE> is the full path and filename of the public world boot
|
|
loaderbinary file (depending on the boot media, this is usually
|
|
either u-boot-spl.bin or u-boot.bin).
|
|
|
|
<OUTPUT_FILE> is the full path and filename of the final secure
|
|
image. The output binary images should be used in place of the standard
|
|
non-secure binary images (see the platform-specific user's guides and
|
|
releases notes for how the non-secure images are typically used)
|
|
u-boot-spl_HS_SPI_X-LOADER - byte swapped boot image for SPI flash
|
|
u-boot_HS_XIP_X-LOADER - boot image for NOR or QSPI flash
|
|
u-boot-spl_HS_ISSW - boot image for all other boot media
|
|
|
|
<SPL_LOAD_ADDR> is the address at which SOC ROM should load the
|
|
<INPUT_FILE>
|
|
|
|
Invoking the script for DRA7xx/AM57xx Secure Devices
|
|
====================================================
|
|
|
|
create-boot-image.sh <IMAGE_TYPE> <INPUT_FILE> <OUTPUT_FILE>
|
|
|
|
<IMAGE_TYPE> is a value that specifies the type of the image to
|
|
generate OR the action the image generation tool will take. Valid
|
|
values are:
|
|
X-LOADER - Generates an image for NOR or QSPI boot modes
|
|
MLO - Generates an image for SD/MMC/eMMC boot modes
|
|
ULO - Generates an image for USB/UART peripheral boot modes
|
|
Note: ULO is not yet used by the u-boot build process
|
|
|
|
<INPUT_FILE> is the full path and filename of the public world boot
|
|
loader binary file (for this platform, this is always u-boot-spl.bin).
|
|
|
|
<OUTPUT_FILE> is the full path and filename of the final secure image.
|
|
The output binary images should be used in place of the standard
|
|
non-secure binary images (see the platform-specific user's guides
|
|
and releases notes for how the non-secure images are typically used)
|
|
u-boot-spl_HS_MLO - boot image for SD/MMC/eMMC. This image is
|
|
copied to a file named MLO, which is the name that
|
|
the device ROM bootloader requires for loading from
|
|
the FAT partition of an SD card (same as on
|
|
non-secure devices)
|
|
u-boot-spl_HS_X-LOADER - boot image for all other flash memories
|
|
including QSPI and NOR flash
|
|
|
|
Invoking the script for Keystone2 Secure Devices
|
|
=============================================
|
|
|
|
create-boot-image.sh \
|
|
<UNUSED> <INPUT_FILE> <OUTPUT_FILE> <UNUSED>
|
|
|
|
<UNUSED> is currently ignored and reserved for future use.
|
|
|
|
<INPUT_FILE> is the full path and filename of the public world boot
|
|
loader binary file (only u-boot.bin is currently supported on
|
|
Keystone2 devices, u-boot-spl.bin is not currently supported).
|
|
|
|
<OUTPUT_FILE> is the full path and filename of the final secure image.
|
|
The output binary images should be used in place of the standard
|
|
non-secure binary images (see the platform-specific user's guides
|
|
and releases notes for how the non-secure images are typically used)
|
|
u-boot_HS_MLO - signed and encrypted boot image that can be used to
|
|
boot from all media. Secure boot from SPI NOR flash is not
|
|
currently supported.
|
|
|
|
Booting of Primary U-Boot (u-boot.img)
|
|
======================================
|
|
|
|
The SPL image is responsible for loading the next stage boot loader,
|
|
which is the main u-boot image. For secure TI devices, the SPL will
|
|
be authenticated, as described above, as part of the particular
|
|
device's ROM boot process. In order to continue the secure boot
|
|
process, the authenticated SPL must authenticate the main u-boot
|
|
image that it loads.
|
|
|
|
The configurations for secure TI platforms are written to make the boot
|
|
process use the FIT image format for the u-boot.img (CONFIG_SPL_FRAMEWORK
|
|
and CONFIG_SPL_LOAD_FIT). With these configurations the binary
|
|
components that the SPL loads include a specific DTB image and u-boot
|
|
image. These DTB image may be one of many available to the boot
|
|
process. In order to secure these components so that they can be
|
|
authenticated by the SPL as they are loaded from the FIT image, the
|
|
build procedure for secure TI devices will secure these images before
|
|
they are integrated into the FIT image. When those images are extracted
|
|
from the FIT image at boot time, they are post-processed to verify that
|
|
they are still secure. The outlined security-related SPL post-processing
|
|
is enabled through the CONFIG_SPL_FIT_IMAGE_POST_PROCESS option which
|
|
must be enabled for the secure boot scheme to work. In order to allow
|
|
verifying proper operation of the secure boot chain in case of successful
|
|
authentication messages like "Authentication passed: CERT_U-BOOT-NOD" are
|
|
output by the SPL to the console for each blob that got extracted from the
|
|
FIT image. Note that the last part of this log message is the (truncated)
|
|
name of the signing certificate embedded into the blob that got processed.
|
|
|
|
The exact details of the how the images are secured is handled by the
|
|
SECDEV package. Within the SECDEV package exists a script to process
|
|
an input binary image:
|
|
|
|
${TI_SECURE_DEV_PKG}/scripts/secure-binary-image.sh
|
|
|
|
This is called as part of the u-boot build process. As the secure
|
|
image formats and requirements can differ between the various secure
|
|
SOCs from TI, this script in the SECDEV package abstracts these
|
|
details. This script is essentially the only required interface to the
|
|
TI SECDEV package for creating a u-boot.img image for secure TI
|
|
devices.
|
|
|
|
The SPL/u-boot code contains calls to dedicated secure ROM functions
|
|
to perform the validation on the secured images. The details of the
|
|
interface to those functions is shown in the code. The summary
|
|
is that they are accessed by invoking an ARM secure monitor call to
|
|
the device's secure ROM (fixed read-only-memory that is secure and
|
|
only accessible when the ARM core is operating in the secure mode).
|
|
|
|
Invoking the secure-binary-image script for Secure Devices
|
|
==========================================================
|
|
|
|
secure-binary-image.sh <INPUT_FILE> <OUTPUT_FILE>
|
|
|
|
<INPUT_FILE> is the full path and filename of the input binary image
|
|
|
|
<OUTPUT_FILE> is the full path and filename of the output secure image.
|