ARM Cortex M4处理器中经过时间的测试C代码中的错误SEGV

Question

I'm writing a code for embedded OS for my stm32f429 board. I'm testing this code for elapsed time:

#include <stdio.h>
#include <stdlib.h>
#include <sys/stat.h> 
#include <fcntl.h>
#include <time.h>
#include <stdint.h>


#define DEFAULT_DELAY   1

uint32_t m_nStart;               //DEBUG Stopwatch start cycle counter value
uint32_t m_nStop;   //DEBUG Stopwatch stop cycle counter value             
#define DEMCR_TRCENA    0x01000000

/* Core Debug registers */
#define DEMCR           (*((volatile uint32_t *)0xE000EDFC))
#define DWT_CTRL        (*(volatile uint32_t *)0xE0001000)
#define CYCCNTENA       (1<<0)
#define DWT_CYCCNT      ((volatile uint32_t *)0xE0001004)
#define CPU_CYCLES      *DWT_CYCCNT



#define STOPWATCH_START { m_nStart = *(*(volatile unsigned int*)0xE0001004);}//DWT_CYCCNT;}
#define STOPWATCH_STOP  { m_nStop = *(*(volatile unsigned int *)0xE0001004);}




static inline void stopwatch_reset(void)
{
    /* Enable DWT */
    DEMCR |= DEMCR_TRCENA; 
    *DWT_CYCCNT = 0;             
    /* Enable CPU cycle counter */
    DWT_CTRL |= CYCCNTENA;
}

static inline uint32_t stopwatch_getticks()
{
    return CPU_CYCLES;
}

static inline void stopwatch_delay(uint32_t ticks)
{
    stopwatch_reset();
    while(1)
    {
        if (stopwatch_getticks() >= ticks)
                break;
    }
}

uint32_t CalcNanosecondsFromStopwatch(uint32_t nStart, uint32_t nStop)
{
    uint32_t nTemp;
    uint32_t n;
    uint32_t SystemCoreClock = 180000000;
    nTemp = nStop - nStart;

    nTemp *= 1000;                          // Scale cycles by 1000.
    n = SystemCoreClock / 1000000;          // Convert Hz to MHz
   nTemp = nTemp / n;                      // nanosec = (Cycles * 1000) / (Cycles/microsec)

   return nTemp;
} 



int main( int argc, char **argv )
{
    int delay = DEFAULT_DELAY;  // Initial value for the delay

    int timeDiff = 0;

    STOPWATCH_START;
    printf("Try\n\n");
    STOPWATCH_STOP;
    timeDiff = CalcNanosecondsFromStopwatch(m_nStart, m_nStop);
    printf("My function took %d nanoseconds\n", timeDiff);  


   return( 0 );
}

It compiles without errors, but when I run this program on my stm32f429, I obtain a SEGV error, probably in the #define STOPWATCH_START. Maybe I have problems on registers (?).

The code is http://pastebin.com/qr6sF9eU (It deleted the calling to system call that I use)

The output of make is: http://pastebin.com/Q14xTaXH

The output when I run my test on stm32f429 board is: http://pastebin.com/sGmjZjxj

Can you help me?

Answer 1

You cannot get a SEGV in a #define as the #define is a preprocessor directive and doesn't form part of the output. It may be in the code produced by the #define. Are you deliberately reading an address from 0xE0001004 , reading what's at that address as another address, then reading the contents of that? That sounds a pretty unlikely way to operate a clock.

Answer 2

The macros STOPWATCH_START; and STOPWATCH_STOP; are causing "illegal indirection" errors from my compiler. The double pointer reference is causing the counter value read to be used as a pointer.

#define STOPWATCH_START { m_nStart = *(*(volatile unsigned int*)0xE0001004);}
#define STOPWATCH_STOP  { m_nStop = *(*(volatile unsigned int *)0xE0001004);}

Based on your other #define statements and the addresses therein, should these be

#define STOPWATCH_START { m_nStart = CPU_CYCLES;}
#define STOPWATCH_STOP  { m_nStop = CPU_CYCLES;}

as you have used elsewhere in the code?

Answer 3

NVIC registers are most likely protected by MPU and not accessible to user code. You cannot mess with everything when using an operating system.