Hello,
I'm working on a board based on MPC8308 and trying to configure the RTC. The problem is that when I try to write into these registers, their value is always 0x0 !!! I've tried a lot of things such as debugging step by step with gdb/JTAG/Peedi, writing to these registers with the Peedi, desabling address translation, using writel/out_le32/..., they are always 0x0 ! However, writing in RAM works perfectly !
Here is a sample of code about what I'd like to do : --------------------------------------- #define RTC_BASE CONFIG_SYS_IMMR + 0x300 #define RTC_CNR RTC_BASE #define RTC_LDR RTC_BASE + 0x4 #define RTC_PSR RTC_BASE + 0x8 #define RTC_TR RTC_BASE + 0xC #define RTC_EVR RTC_BASE + 0x10 #define RTC_ALR RTC_BASE + 0x14
void initRTC() {
u32* volatile psr = RTC_PSR; u32* volatile ldr = RTC_LDR; u32* volatile cnr = RTC_CNR; u32* volatile loadaddr = 0x800000;
u32 psr_value = 0x0; u32 ldr_value = 0xABCD1234; u32 cnr_value = (0x1 << 7);// | (1 << 24); u32 loadValue = 0xBEEFABCD;
*psr = psr_value; // Divide clock by 1 *ldr = ldr_value; // Initial time *cnr = cnr_value;// | (1 << 24);// Clock enable use, CSB clock *loadaddr = loadValue;
printf("PSR pointer is 0x%08X, put 0x%08X and actually is 0x%08X\n", psr, psr_value, *psr); printf("LDR pointer is 0x%08X, put 0x%08X and actually is 0x%08X\n", ldr, ldr_value, *ldr); printf("CNR pointer is 0x%08X, put 0x%08X and actually is 0x%08X\n", cnr, cnr_value, *cnr); printf("loadaddr pointer is 0x%08X, put 0x%08X and actually is 0x%08X\n", loadaddr, loadValue, *loadaddr); } --------------------------------------- And the output : --------------------------------------- PSR pointer is 0xE0000308, put 0x00000000 and actually is 0x00000000 LDR pointer is 0xE0000304, put 0xABCD1234 and actually is 0x00000000 CNR pointer is 0xE0000300, put 0x00000080 and actually is 0x00000000 loadaddr pointer is 0x00800000, put 0xBEEFABCD and actually is 0xBEEFABCD ---------------------------------------
I have the same problem with ANY register ! I also noticed a register that should be initialized to 0xFFFFFFFF is actually 0x0 at reset ! More it goes, more I think I'm reading the wrong datasheet or the wrong number has been printed on the chip :-p ! Or a bug ... what I doubt ... !
I've been stuck here all the week, without finding any clue, please give me something to dig into !
Thanks for your precious advices !
Guy Morand