Tool/software:
Hello,
I`m currently developing software on a TMS570LC4357-EP board and am trying to run the SCM self-test (per TRM section 3.3.2).
I am running the code to start the test early in the execution, right after the registers are initialized (_coreInitRegisters_ and _coreInitStackPointer_ ), as part of a sequence of various self-tests in order to validate system stability.
This setup is similar to the one mentioned in this issue:
TMS570LC4357: SCM Self Test issue - Arm-based microcontrollers forum - Arm-based microcontrollers - TI E2E support forums
I'm discovering something strange in that the self-test fails to initiate due to pending transactions on the CPU PP-AXI bus master when I try to start from a power cycle of the board.
The self-test successfully initiates if I run it after commanding a system reset from a debugger (Lauterbach Trace32 Debugger, JLInk Debugger, etc.) connected to the HDK board.
But when I unplug the HDK from power supply, re-plug it and re-attach the probe to the board, all I see in the call stack is that I get perpetually stuck in the loop that starts the self-test (_gotoCPUIdle_())
I was wondering if anyone else had experience running this self-test from a hardware power-cycled machine and if a workaround was known?
