tfalkstub is a simple stub firmware that allows using the 32-bit "Little Kernel" (LK) bootloader from Qualcomm on top of 64-bit Trusted Firmware-A (TF-A). It switches the CPU to 32-bit mode and emulates required Secure Monitor Calls (SMC) to make LK work properly.
The following Qualcomm SoCs are supported by tfalkstub (and TF-A) so far:
tfalkstub is loaded at address 0x86400000, while LK should be loaded at 0x8f600000.
All control is handed over to the actual operating system after LK has finished running.
A 64-bit kernel should be loaded from LK; the emulation layer for 32-bit kernels in tfalkstub
is not designed to be used for anything except LK. Booting e.g. 32-bit Linux (instead of 64-bit)
may produce unexpected results.
Build and install TF-A for msm8916
with the following configuration (this will make it jump to tfalkstub instead
of directly to the aboot firmware):
$ make CROSS_COMPILE=aarch64-linux-gnu- PLAT=msm8916 PRELOADED_BL33_BASE=0x86400000
Build tfalkstub:
$ make CROSS_COMPILE=aarch64-linux-gnu-
The resulting ELF file must be "signed" before flashing it, even if the board has secure boot disabled. In this case the signature does not provide any security, but it provides the firmware with required metadata.
On boards without secure board you can use qtestsign to sign the firmware with automatically generated test keys:
$ ./qtestsign.py hyp tfalkstub.elf
Tip: If you clone qtestsign directly into your tfalkstub clone,
running make will also automatically sign the binary!
Install the resulting tfalkstub-test-signed.mbn to the hyp partition.
WARNING: Do not flash incorrectly signed firmware on devices that have secure boot enabled! Make sure that you have a way to recover the board in case of problems (e.g. using EDL).
tfalkstub is based on qhypstub and is also licensed under the GNU General Public License, version 2.