Content originally posted in LPCWare by jokn on Mon Sep 24 07:16:33 MST 2012
I'm testing the LPC43xx CMSIS library Vers 2012-05-31.
Here the dual core example:
/Examples/DUALCORE/Int_Demo.
The SPIFI Version works fine on my custom board as well as on the hitex board.
Then I tried the to run the M4_RAM target configuration and always get a Hard Fault (Vector Table Fault)
I spend al lot of time to find out the reason for such a strange behavior. Finally inspected the debug initialization file:
LPC43xx Internal SRAM.ini
and found some initializations for the M0 core.
When I comment out the Pre_setuo() statement, the Hard Fault vanished an the project works fine including the IPC communication with the M0 core
Does anybody know what is the real intention of the M0 Pres_Setup() and why it causes a Hard Fault?
FUNC void Pre_Setup (void) {
/* configure M0 for infinite loop */
_WDWORD(0x10080000,0x00001F00);/* dummy stack pointer */
_WDWORD(0x10080004,0x000000D5);/* reset handler */
_WDWORD(0x100800D4,0xE7FEE7FE);/* jump to itself instruction for M0a */
_WDWORD(0x40043404,0x10080000);/* M0 shadow pointer. */
/* release M0 from reset to allow Jtag access */
_WDWORD(0x40053104,(~(_RDWORD(0x40053154))) & (~(1<<24)));
}
FUNC void Setup (unsigned int region) {
region &= 0xFF000000;
SP = _RDWORD(region); // Setup Stack Pointer
PC = _RDWORD(region + 4); // Setup Program Counter
_WDWORD(0xE000ED08, region); // Setup Vector Table Offset Register
}
/* prepare M0 for access, if needed */
//Pre_setup();
/* uncomment the following line if using a Jlink, and uncheck the "load application at startup" */
/* checkbox inthe project settings > debug window in uVision */
/*
LOAD "keil_output\\LPC43xx Internal SRAM\\example.axf" INCREMENTAL
*/
Setup(0x10000000); // Get ready to execute image in SRAM
