Part Number: TMS320F280049C
Hey TI-Team,
I want to store a variable to a certain address and get the reset cause. I reset my device by using the driverlib function:
SysCtl_resetDevice();
Now, I can invoke
SysCtl_getResetCause();
in order to get the reset cause. Since I want to distinguish between intended reset (my software triggering the watchdog reset) and unintended reset, the watchdog timing out, I want to store
a intended reset flag in ram. This should be possible, since the tmsf280049c ( this is not the case for every ti uc! ) will not erase ram on watchdog reset.
Now in the linker command file I added:
MEMORY
{
...
RAMLS7 : origin = 0x00B800, length = 0x000800
...
}
SECTIONS
{ ....
.resetCause : > RAMLS7, PAGE = 1
....}
and in my *.c file in global space:
#pragma DATA_SECTION(resetTriggered, ".resetCause")
uint16_t resetTriggered;
compiling result in .map:
.resetCause
* 1 0000b800 00000001 UNINITIALIZED
0000b800 00000001 myTestObj.obj (.resetCause)
The Handbook only states:
After a POR, the boot ROMs will clear the M0/M1, LSx, GSx, and message RAMs to ensure that they contain valid ECC or parity.
Same counts for the boot section.
And in this post the tmsf280049 is mentioned as not resetting ram, but since I use the tmsf280049c.. (https://e2e.ti.com/support/microcontrollers/c2000/f/171/t/742852?TMS320F280049-Which-reset-source-will-cause-the-clear-of-RAM-data-on-F280049
This behavior is recognized in release/debug mode. What do I miss out?
Thanks in advance.
