HSE Firmware Flashing

[Z0091608]

Hello,
I have flashed HSE without FBL or APP (cmm script from IVP “IVP\\debug
S32K388_HSE_INSTALL.cmm“, “SADM_s32k388_hse_fw_28.16.12_2.69.4.pink“). I was successfull, but now I understand it was not a good idea
Previous value of register “HSE_GPR_addr = SD:0x4039C028“ was 0xC0, after HSE SBAF *** value is 0xE1. This based on copilot’s description seems like enabled HSE.
Unfortunately my only experience is Tricore TC233, where Infineon firmware was simple - but we have also bricked some of the ECUs during development - enabled HSM without firmware in place.
I have used copilot to get a little bit oriented in the situation.
I have learned that for NXP K388, it shouldn’t be the end, however by enabling HSE, situation has changed - eg. we have flashed APP before without issues to 0x00400000, now I cannot freely flashed to that memory (memory filled with 0xFF and access secured by HSE?).
Some memory dump attached in “HSE_issue_trace32_dump_1.png “.
My questions are:
(1) is the evalboard now “lost“ for further GM related development?
(2) copilot described something like NXP recovery procedure, with S32 studio, using UART cable - would this be a good direction to get control over ECU again?
I don’t fully understand the situation and it’s impact.
Please, do you have any suggestions/ideas? Thank you for any information you might have. I can provide additional information of course.

[lukaszadrapa]

Hi @Z0091608 

It’s not wrong step to install the firmware. It won’t brick the device. The device is still in CUST_DEL life cycle, so it does not secure the debug port. The only noticeable difference at the beginning is that certain memory areas are exclusively allocated for HSE firmware. It’s removed from memory map, user can’t access it anymore.

The memory map is described in following tables in S32K3 manual:

Table 196. Configuration details when the HSE firmware usage feature flag is disabled

Table 198. Configuration details when the HSE_B firmware usage feature flag is enabled

Table 199. AB swap configuration

https://www.nxp.com/webapp/Download?colCode=S32K3XXRM

Note: Table 198 shows the memory when HSE usage feature flag is enabled – that’s the same like when FULL_MEM firmware is installed.

If you can’t program the memory now, it may be caused by your script which is used for programming. The script should not touch the memory areas which are allocated for HSE. You can easily check it – just run the script in T32 which is supposed to initialize the flash programming. Then use command flash.list and watch which memory areas are defined for programming and compare it with the tables mentioned above. If reserved areas are included, it is necessary to change the flash initialization in your script.

And, of course, you should also adjust linker file of your application to make sure that no data or code is put to reserved memory.

If 0x4039C028 is 0xE1 – yes, HSE firmware is successfully installed. It also says that the device entered recovery mode. This also corresponds to the screenshot because it hangs in endless loop in RAM memory at 0x2040012C. This is RAM recovery address.

The device enters recovery mode when there’s no valid IVT in flash (that’s your case) or when secure boot was not successful (you have not configured secure boot yet).

Because you use premium HSE firmware, I recommend to use dedicated support channel – either direct FAE  support or enter a ticket here (https://support.nxp.com/s/?language=en_US), so it is assigned to regional FAE team or maybe there’s private community space for your company, I’m not sure because I don’t have access.

Hope this helps.

Regards,

Lukas