首页 文章

程序堆内存的初始大小

提问于
浏览 次
0

程序：

#include<stdio.h>
#include<sys/types.h>
#include<malloc.h>
main()
{
    int *i1, *i2;
    char *s;
    printf("sbrk(0) before malloc(4): %x\n", sbrk(0));
    i1 = (int *) malloc(sizeof(int));
    printf("sbrk(0) after `i1 = (int *) malloc(4)': %x\n", sbrk(0));
    i2 = (int *) malloc(sizeof(int));
    printf("sbrk(0) after `i2 = (int *) malloc(4)': %x\n", sbrk(0));
}

输出：

sbrk(0) before malloc(4): a027000
sbrk(0) after `i1 = (int *) malloc(4)': a048000
sbrk(0) after `i2 = (int *) malloc(4)': a048000

上述程序的堆内存的初始大小是多少 . 我认为，在程序的初始阶段，堆的起始和结束地址是相同的 . 有一次，我们调用malloc，它使用sbrk分配内存 .

上面的程序显示，首先sbrk（0）返回a027000作为程序中断 . 执行malloc语句后，程序中断更改为a048000 . 它表明，在调用malloc之前，堆没有足够的内存 . 因此，调用malloc后只更改程序中断 . 在初始状态下，如果堆具有足够的内存，则无需更改程序中断 . 所以首先，堆的大小为零 . 这样对吗？

c linux unix

2 回答

  • 1

    发生什么取决于你的O / S在调用_2560952之前的作用 .

    在Mac OS X上，在调用 main() 函数之前会发生大量的内存分配 . 我目前有一个18个左右的列表被抑制'memory abuses'，这些都是在系统提供的启动代码调用 main() 之前发生的事情 .

    运行一个简单的程序，每行打印一个参数，但不给它任何参数，因此没有输出，从而产生valgrind的使用报告，如下所示：

    ==59405== Memcheck, a memory error detector
==59405== Copyright (C) 2002-2013, and GNU GPL'd, by Julian Seward et al.
==59405== Using Valgrind-3.11.0.SVN and LibVEX; rerun with -h for copyright info
==59405== Command: /Users/jleffler/bin/al
==59405== 
--59405-- /Users/jleffler/bin/al:
--59405-- dSYM directory is missing; consider using --dsymutil=yes
--59405-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option
--59405-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 2 times)
--59405-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 4 times)
==59405== 
==59405== HEAP SUMMARY:
==59405==     in use at exit: 34,808 bytes in 417 blocks
==59405==   total heap usage: 517 allocs, 100 frees, 41,784 bytes allocated
==59405== 
==59405== LEAK SUMMARY:
==59405==    definitely lost: 16 bytes in 1 blocks
==59405==    indirectly lost: 0 bytes in 0 blocks
==59405==      possibly lost: 13,002 bytes in 109 blocks
==59405==    still reachable: 21,790 bytes in 307 blocks
==59405==         suppressed: 0 bytes in 0 blocks
==59405== Rerun with --leak-check=full to see details of leaked memory
==59405== 
==59405== For counts of detected and suppressed errors, rerun with: -v
==59405== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)

    该程序的代码？

    #include <stdio.h>
#include <stdlib.h>

int main(int argc, char **argv)  
{ 
    if (argc > 1)
    {
        while (*++argv) 
            puts(*argv);
    }
    return(EXIT_SUCCESS); 
}

    大量的内存活动，但在我写的代码中都没有 . 泄漏是我在抑制文件中记录的数字之一 .

    回复于
  • 0

    当安装glibc debuginfo时（例如在Fedora 22上： dnf debuginfo-install glibc-2.21-5.fc22.x86_64 ），可以打印出 main_arena 和 mp_ .

    使用空 int main(void) 调试程序：

    mp_.sbrk_base 是堆的基础 . 在我的Linux上调用 main 时，堆不存在 . 致电 malloc(1000) mp_.sbrk_base 后 0x602000 . main_arena.top 是堆顶部的空闲内存块 . 这是 0x6023f0

    (gdb) b main
Breakpoint 1 at 0x4004fa: file test.c, line 3.
(gdb) r 
Starting program: /home/m/a.out 

Breakpoint 1, main () at test.c:3
3       return 0;
(gdb) p mp_
$1 = {trim_threshold = 131072, top_pad = 131072, mmap_threshold = 131072, arena_test = 8, arena_max = 0, n_mmaps = 0, 
  n_mmaps_max = 65536, max_n_mmaps = 0, no_dyn_threshold = 0, mmapped_mem = 0, max_mmapped_mem = 0, max_total_mem = 0, 
  sbrk_base = 0x0}
(gdb) p main_arena 
$2 = {mutex = 0, flags = 0, fastbinsY = {0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0}, top = 0x0, 
  last_remainder = 0x0, bins = {0x0 <repeats 254 times>}, binmap = {0, 0, 0, 0}, next = 0x7ffff7dd6b20 <main_arena>, 
  next_free = 0x0, system_mem = 0, max_system_mem = 0}
(gdb) call malloc(1000)
$3 = (void *) 0x602010
(gdb) p mp_
$4 = {trim_threshold = 131072, top_pad = 131072, mmap_threshold = 131072, arena_test = 8, arena_max = 0, n_mmaps = 0, 
  n_mmaps_max = 65536, max_n_mmaps = 0, no_dyn_threshold = 0, mmapped_mem = 0, max_mmapped_mem = 0, max_total_mem = 0, 
  sbrk_base = 0x602000 ""}
    (gdb) p main_arena
$5 = {mutex = 0, flags = 1, fastbinsY = {0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0}, top = 0x6023f0, 
  last_remainder = 0x0, bins = {0x7ffff7dd6b78 <main_arena+88>, 0x7ffff7dd6b78 <main_arena+88>, [...]
0x7ffff7dd7198 <main_arena+1656>, 0x7ffff7dd71a8 <main_arena+1672>, 0x7ffff7dd71a8 <main_arena+1672>...}, binmap = {0, 
0, 0, 0}, next = 0x7ffff7dd6b20 <main_arena>, next_free = 0x0, system_mem = 135168, max_system_mem = 135168}

(gdb) call sbrk(0)
$6 = 6434816
(gdb) p (void*)$6
$7 = (void *) 0x623000
(gdb)
    回复于

相关问题