HPS/ARM Freeze when reading SDRAM - Cyclone V

Altera_Forum · ‎03-06-2018

Hello,

I'm trying to execute a code on ARM-A9 that's part of the HPS inside the cyclone v (DE10-Nano) and when I try to access some specific SDRAM address the program freezes (program counter gets stucked), the curious aspect is that the freeze only happen when I read the content of specific addresses like (0xc000282). My code is simple, first I write an image from address 0xc00000 to 0xc4b000 and then I make a loop to read each byte address that represents a pixel, the code that I'm running is this:

# include <assert.h># include <stdbool.h># include <stdio.h># include <stdlib.h>
# include "alt_clock_manager.h"# include "alt_generalpurpose_io.h"# include "alt_globaltmr.h"# include "hwlib.h"# include "socal/alt_gpio.h"# include "socal/hps.h"# include "socal/socal.h"# include "hps_0.h"# include "system_crios.h"# include "mser.h"
int main(void) {
  setup_system();
  //mser_init();
  while (1) {
    uint32_t *pt = 0x00C00282;
    uint32_t content;
    content = *pt; // IN THIS MOMENT I LOST CONTROL OF THE PROGRAM, THE PC KEEPS FREEZE
    //uint8_t test = getPixelLevel(641); // PIXEL 641 --> 0x00C00282
    delay_us(ALT_MICROSECS_IN_A_SEC/10);
  }
  return 0;
}

The auxiliary functions like (setup_system) just make some malloc concerns but nothing special that could break my program flow, and if I try to read other addresses the programs returns the right pixel as I expect. The function that I had used to read the memory was this:


uint8_t getPixelLevel(uint32_t pixel){
  uint32_t *ptr_level = pixel+IMG_OFFSET_ADDR;
  uint8_t remain = pixel%4;
  uint32_t mask = (0xff << remain*8);
  uint32_t content = *ptr_level;
  uint8_t data = (content & mask) >> remain*8;
  return data;
}

Does someone has the same problem or know specific rules that I need to follow to read the SDRAM address?