＆QUOT; EXC_BAD_ACCESS：无法恢复previously选择框架＆QUOT;错误，数组大小？ [英] "EXC_BAD_ACCESS: Unable to restore previously selected frame" Error, Array size?

问题描述

我对创建埃拉托色尼的筛子，并从中拉素数的算法。它可以让你的筛子输入最大值和算​​法给你的素数低于该值，并存储这些在C风格的数组。

I have an algorithm for creating the sieve of Eratosthenes and pulling primes from it. It lets you enter a max value for the sieve and the algorithm gives you the primes below that value and stores these in a c-style array.

问题：
一切工作正常值高达500.000，但是当我进入一个较大的值助益，而在运行 - 它给了我以x code以下错误信息：

Problem: Everything works fine with values up to 500.000, however when I enter a large value -while running- it gives me the following error message in xcode:

Program received signal:  "EXC_BAD_ACCESS".
warning: Unable to restore previously selected frame.
Data Formatters temporarily unavailable, will re-try after a 'continue'. (Not safe to call dlopen at this time.)

我的第一个想法是，我并没有用足够大的变数，但我使用'无符号长长整型，这不应该成为问题。此外调试器指向我在我的code的点阵列中的点会被分配一个值。
因此，我想知道是否有一个最大限制到一个数组？如果是的话：我应该使用NSArray中呢？如果没有，那么什么是基于这一信息引起该错误？

My first idea was that I didn't use large enough variables, but as I am using 'unsigned long long int', this should not be the problem. Also the debugger points me to a point in my code where a point in the array get assigned a value. Therefore I wonder is there a maximum limit to an array? If yes: should I use NSArray instead? If no, then what is causing this error based on this information?

编辑：
这是code样子（这是不完全的，因为它未能在公布的最后一行）。我使用的垃圾收集。

This is what the code looks like (it's not complete, for it fails at the last line posted). I'm using garbage collection.

/*--------------------------SET UP--------------------------*/
    unsigned long long int upperLimit = 550000;             // 
    unsigned long long int sieve[upperLimit];
    unsigned long long int primes[upperLimit];
    unsigned long long int indexCEX;
    unsigned long long int primesCounter = 0;

// Fill sieve with 2 to upperLimit
for(unsigned long long int indexA = 0; indexA < upperLimit-1; ++indexA) {
        sieve[indexA] = indexA+2;
}


unsigned long long int prime = 2;

/*-------------------------CHECK & FIND----------------------------*/
while(!((prime*prime) > upperLimit)) {

    //check off all multiples of prime
    for(unsigned long long int indexB = prime-2; indexB < upperLimit-1; ++indexB) {

        // Multiple of prime = 0
        if(sieve[indexB] != 0) {
            if(sieve[indexB] % prime == 0) {
                sieve[indexB] = 0;
            }
        }
    }

    /*---------------- Search for next prime ---------------*/
    // index of current prime + 1
    unsigned long long int indexC = prime - 1;

    while(sieve[indexC] == 0) {
        ++indexC;
    }
    prime = sieve[indexC];

    // Store prime in primes[]
    primes[primesCounter] = prime; // This is where the code fails if upperLimit > 500000
    ++primesCounter;

    indexCEX = indexC + 1;

}

，因为你可能会或可能不会看到，我是 - 非常much-初学者。其他任何建议，欢迎当然是：）

As you may or may not see, is that I am -very much- a beginner. Any other suggestions are welcome of course :)

推荐答案

您不是四溢的变量;你溢出堆栈。当你创建像在栈上 INT myArray的[500] ，你就宣告500 INT 秒的数组。正常的堆栈大小为8 MB。只有你的两个数组约8.4 MB（8字节* 55万/（1024 ^ 2）= 4.2 MB）。您应该使用的堆内存（从的malloc（）），而不是在这里。因此，这将是这样的：

You're not overflowing the variables; you're overflowing the stack. When you create an array like int myArray[500], you're declaring 500 ints on the stack. The normal stack size is 8 MB. Your two arrays alone are about 8.4 MB (8 bytes * 550000 / (1024^2) = 4.2 MB). You should be using heap memory (from malloc()) here instead. So it would be like this:

int upperLimit = 550000;
unsigned long long *sieve = malloc(sizeof(long long) * upperLimit);
unsigned long long *primes = malloc(sizeof(long long) * upperLimit);
unsigned long long indexCEX;
unsigned long long primesCounter = 0;

不要忘了，你需要免费（）内存当你用它做，否则你会最终泄漏。

Don't forget that you'll need to free() the memory when you're done with it or you'll end up leaking.

这篇关于＆QUOT; EXC_BAD_ACCESS：无法恢复previously选择框架＆QUOT;错误，数组大小？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！