Open jettr opened 4 months ago
I think this issue is also occurring for the primary boot slot field if you attempt to swap it using boot services. This is what I'm doing:
If I reboot the device again, the primary slot updates to the correct value. Having the primary slot being correctly reported by ROM_EXT is useful to know if an image update was successful or not.
I'm seeing this issue on the earlgrey_es_sival branch using the ROM_EXT 0.4 prebuilt.
Edit: Sorry I think this is probably a different issue... this one is for reporting the booted ROM_EXT slot, but the issue I'm seeing is about the reported primary BL0 slot. I'll file a new issue.
Description
It appears that the boot log that reports the active ROM_EXT is incorrect in some scenarios. Notice the following from the abridge firmware log after performing a firmware update that contains both a new ROM_EXT and new Owner Firmware.
Starting ROM_EXT 0.2
0x406007a8: 5f5f4141 00000000 00000002 00010000
(memory dump of BootLog section start withrom_ext_slot
Starting ROM_EXT 0.3
0x406007a8: 5f5f4141 00000000 00000003 00010000
rom_ext_slot
is reporting SlotA (AA__
), but the version is reporting0.3
, but SlotA contains 0.2, and SlotB contains 0.3. I am pretty sure 0.3 is what actually ran since it printedStarting ROM_EXT 0.3
on the UART.The boot log remains incorrect event after performing soft reboots from FW (via
reboot
console command). Only after I perform a chip reset with theRESET_L
pin, do I see the boot log be correct.Log from a ROM_EXT and owner firmware update:
Log from firmware initiated soft reboot
Log after performing
ott gpio apply RESET && ott gpio remove RESET
Notice that
0x406007a8: 42425f5f 00000000 00000003 00010000
correct indicates that SlotB (__BB
) is executing ROM_EXT version 0.3 now.