Linux glibc 检测到 free():下一个大小无效(快速)

声明:本页面是StackOverFlow热门问题的中英对照翻译,遵循CC BY-SA 4.0协议,如果您需要使用它,必须同样遵循CC BY-SA许可,注明原文地址和作者信息,同时你必须将它归于原作者(不是我):StackOverFlow 原文地址: http://stackoverflow.com/questions/9073148/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me): StackOverFlow

提示:将鼠标放在中文语句上可以显示对应的英文。显示中英文
时间:2020-08-06 04:17:49  来源:igfitidea点击:

glibc detected free(): invalid next size (fast)

clinuxfreeglibc

提问by Wade G

This code generates random numbers and then produces a histogram based on input to the functions regarding the intervals. "bins" represents the histogram intervals and "bin_counts" holds the number of random numbers in a given interval.

此代码生成随机数,然后根据有关间隔的函数的输入生成直方图。“bins”表示直方图间隔,“bin_counts”保存给定间隔内随机数的数量。

I've reviewed several of the posts dealing with similiar issues and I understand that I'm out of bounds in the memory somewhere but GBD only points me to the "free(bins);" at the end of the code. I've double-checked my array lengths and I think they are all correct in terms of not accessing elements that don't exist/writing to memory not allocated. The weird thing is that the code works as intended, it produces an accuarate histogram, now I just need helping cleaning up this free() invalid next size error. If anybody has any suggestions I would be much obliged. The whole output is :

我已经查看了一些处理类似问题的帖子,我知道我在某处的内存中超出了界限,但 GBD 只将我指向“free(bins);” 在代码的末尾。我已经仔细检查了我的数组长度,我认为它们在不访问不存在的元素/写入未分配的内存方面都是正确的。奇怪的是代码按预期工作,它产生了一个准确的直方图,现在我只需要帮助清理这个 free() 无效的下一个大小错误。如果有人有任何建议,我将非常感激。整个输出是:

glibc detected./file: free(): invalid next size (fast): 0x8429008

glibc 检测到./file: free(): 下一个大小无效(快速):0x8429008

followed by a bunch of addresses in memory, seperated by Backtrace and Memory Map. The Backtrace only points me towards line 129, which is "free(bins);". Thanks in advance

后面是一堆内存地址,由Backtrace和Memory Map分开。回溯仅将我指向第 129 行,即“free(bins);”。提前致谢

    #include "stdio.h"
    #include "string.h"
    #include "stdlib.h"

    void histo(int N, double m, double M, int nbins, int *bin_counts, double *bins);

     int main(int argc, char* argv[])
     {

     int *ptr_bin_counts;
     double *ptr_bins; 

     histo(5,0.0,11.0,4, ptr_bin_counts, ptr_bins);

     return 0;
     } 

     void histo(int N, double m, double M, int nbins, int *bin_counts, double *bins)
     {

     srand(time(NULL));
     int i,j,k,x,y;
     double interval;
     int randoms[N-1];
     int temp_M = (int)M;
     int temp_m = (int)m;
     interval = (M-m) /((double)nbins);


     //allocating mem to arrays
     bins =(double*)malloc(nbins * sizeof(double));
     bin_counts =(int*)malloc((nbins-1) * sizeof(int));

     //create bins from intervals
     for(j=0; j<=(nbins); j++)
     {
            bins[j] = m + (j*interval); 
     } 

      //generate "bin_counts[]" with all 0's
      for(y=0; y<=(nbins-1); y++)
       {
         bin_counts[y] = 0; 
       }


      //Generate "N" random numbers in "randoms[]" array
      for(k =0; k<=(N-1); k++)
      {
          randoms[k] = rand() % (temp_M + temp_m);
          printf("The random number is %d \n", randoms[k]);
      }

       //histogram code 
       for(i=0; i<=(N-1); i++)
        {
         for(x=0; x<=(nbins-1); x++)
         {
              if( (double)randoms[i]<=bins[x+1] && (double)randoms[i]>=bins[x] )
               {
                    bin_counts[x] = bin_counts[x] + 1; 
               }
         }
         }
         free(bins);
         free(bin_counts);
         }

采纳答案by Daniel Fischer

bins =(double*)malloc(nbins * sizeof(double));
bin_counts =(int*)malloc((nbins-1) * sizeof(int));

//create bins from intervals
for(j=0; j<=(nbins); j++)
{
    bins[j] = m + (j*interval); 
} 

//generate "bin_counts[]" with all 0's
for(y=0; y<=(nbins-1); y++)
{
    bin_counts[y] = 0; 
}

You are overstepping your arrays, you allocate place for nbinsdoubles but write to nbins+1locations, and use nbinslocations for bin_countsbut have only allocated nbins-1.

您超出了数组,为nbins双打分配了位置但写入了nbins+1位置,并使用了nbins位置bin_counts但仅分配了nbins-1.