emWIN GUI_Init() creates Hard fault on LPC56428 using Amazon FreeRTOS

Kernel
1

anealca wrote on Wednesday, June 26, 2019:

HELP:
I’m trying to port a working LPC54608 example to the LPC54628 mcu.
I have the RTOS working correctly with LWIP and can create threads, Queues, Semafor’s, etc.
I can get emWIN working without the RTOS on the LPC54628 dev board (OM13098).
I’m using NXP’s MCUXpresso v10 IDE with v2.5 SDK for this new mcu.
BUT
I get a forced bus fault (hard fault) in GUI_Init() when porting to use the RTOS.
The LCD init and I2C (TFT LCD) returns success.

What can I do to isolate/resolve this issue?

2

rtel wrote on Wednesday, June 26, 2019:

A couple of suggestions to start with.

First if you have not already done so then read through the following
page of the FAQ: https://www.freertos.org/FAQHelp.html

Second, step into the offending function to see which line causes the
fault, then report back to see if we can suggest anything further (could
just be a linker script error, or something basic like that).

3

anealca wrote on Wednesday, June 26, 2019:

Thanks Richard…
I have read that doc but it never hurts to re-read it…
The function (GUI_Init()) that causes the hard fault is in the emWIN Library (libemWIN_M4F.a) and does not allow me to step into it.
I have set and used assert but I’m get nothing back from it during the LOAD_X… Ftns. They all execute without error.
There must be some way to enable internal error checking in the library ftn’s ?

4

rtel wrote on Wednesday, June 26, 2019:

There must be some way to enable internal error checking in the library
ftn’s ?

I would guess so, but I’m not familiar with the emWin library I’m afraid.

5

anealca wrote on Wednesday, June 26, 2019:

Thanks Richard…
It just seems funny that it runs OK ON THE 54628 when the MCU is set to 54608 but when it is set to 54628 it fails… These 2 processors are supposed to be plug-n-play replacible…
The 54628 is just faster…
I’ll continue writting code using the older processor for now…