S32DS start debug session fails 1'st time

[ofer-bahar-guardknox]

When starting to debug on S32G2 on the RDB2 (NXP GoldBox) using the NXP S32 Debug Probe, it fails with the following message on the first try

oferbaharguardknox_0-1688294318808.png

doing relaunch always succeeds and the debug goes well

what is error 102? regarding the GDB

here are the debug configuration 

oferbaharguardknox_1-1688294350235.png

[Daniel-Aguirre]

Hi,

Error 102 is defined as a connection error, as shown below:

DanielAguirre_0-1688405729319.png

It can be either that your board was not energized prior to the debugging session or that the cable was not seeded correctly.

Please, let us know.

[ofer-bahar-guardknox]

Hi

It's definitely not the power nor any cabling issue, the GoldBox is powered all the time, nothing moves in terms of the hardware, the board is brand new NXP GoldBox RDB2, the S32 Debug Probe is brand new, all cables are NXP

The problem is deterministic 100% 

1'st DEBUG run is successful, meaning starting the debug => debug session stops at the start (i.e. main.c first line) =>  Resume (F8) or Step over (F6) going perfect. Since no 'go to start' or RESET is available in the S32DS 3.5, a relaunch has to be done, and then

2'nd DEBUG after relaunch fails all the time

3'rd DEBUG relaunch goes OK as the 1'st

4'th fails

5'th OK

and so on... every 2'nd launch fails 

looks like a RESET issue or GDB server reloading sequence logic bug 

[ofer-bahar-guardknox · Accepted solution]

Update on this one

it looks like the BootROM is polling and interfere with the loading to RAM

As written in the AN12422

oferbaharguardknox_0-1688484151990.png

In my case (GoldBox RDB2), all on board dip switches are in default position

Changing SW4 / 7'th was to OFF, meaning boot not from SD Card but from NOR Flash, solves the issue, and terminate and launch works all the time and does not fail

looks like booting from the SD Card somehow interfere with the loadind to the RAM