stl 有內(nèi)存泄漏嗎？（高人指點(diǎn)一下）

dori

void testerr()
{
   std::vector<std::string>; vcon
   for(int i=0;i<1000000;++i)
     vcon.push_back("asdfasdfsd";
}

void testok()
{
  std::string ka="asdfasdfsd";
  std::vector<std::string>; vcon
  for(int i=0;i<1000000;++i)
    vcon.push_back(ka);
}
int main(int argc,char)
{
        testerr();//or testok
        getchar();
        return 123;
}

linux: redhat 8.0

can not see program release the memory.

but testok can release the memory.

note:use top tool to see the memory

hatname

應(yīng)該都沒有問題吧,按照C++的風(fēng)格,它應(yīng)該在析構(gòu)函數(shù)部分完成內(nèi)存釋放工作.
如果你不放心加一個 vcon.clear()

dori

有沒有高人啊

gooderfeng

問一下，怎么看出來release或者沒有release的。

renstone921

linux下可以用內(nèi)存泄漏用 valgrind 檢查。
你從TOP根本就看不出是否會產(chǎn)生內(nèi)存泄漏。

另外，上面的語句不會產(chǎn)生內(nèi)存泄漏，除非你中獎使用以前的某個版本的stl。

dori

不懂啊，為何 testok 可以看到內(nèi)存使用下降，而testerr卻看不到呢？

renstone921

系統(tǒng)有可能把已分配的內(nèi)存在進(jìn)程退出后作為系統(tǒng)緩存區(qū)使用。

依賽特小子

stl是C++的標(biāo)準(zhǔn),只要是標(biāo)準(zhǔn)就應(yīng)該沒有問題!但按標(biāo)準(zhǔn)實(shí)現(xiàn)的程序就不一定了!!!

dori

沒有令人滿意的回答

renstone921

不知道你的產(chǎn)生內(nèi)存泄漏的結(jié)論是如何的出來的.能不能給提供個數(shù)據(jù)看一下.


下面使用valgrind檢測的結(jié)果:

1. 
   
2. [root@localhost prog]# valgrind -v memoryLeak
   
3. ==2776== Memcheck, a memory error detector.
   
4. ==2776== Copyright (C) 2002-2005, and GNU GPL'd, by Julian Seward et al.
   
5. ==2776== Using LibVEX rev 1313, a library for dynamic binary translation.
   
6. ==2776== Copyright (C) 2004-2005, and GNU GPL'd, by OpenWorks LLP.
   
7. ==2776== Using valgrind-3.0.0, a dynamic binary instrumentation framework.
   
8. ==2776== Copyright (C) 2000-2005, and GNU GPL'd, by Julian Seward et al.
   
9. --2776-- Valgrind library directory: /usr/lib/valgrind
   
10. --2776-- Command line
    
11. --2776--    memoryLeak
    
12. --2776-- Startup, with flags:
    
13. --2776--    -v
    
14. --2776-- Contents of /proc/version:
    
15. --2776--   Linux version 2.6.13-1.1535_FC5 (bhcompile@bugs.build.redhat.com) (gcc version 4.0.1 20050831 (Red Hat 4.0.1-12)) #1 Sun Sep 4 02:20:28 EDT 2005
    
16. --2776-- Reading syms from /root/prog/memoryLeak (0x8048000)
    
17. --2776-- Reading syms from /lib/ld-2.3.90.so (0x1B8E4000)
    
18. --2776-- Reading syms from /usr/lib/valgrind/stage2 (0xB0000000)
    
19. --2776-- Reading suppressions file: /usr/lib/valgrind/default.supp
    
20. ==2776==
    
21. --2776-- Reading syms from /usr/lib/valgrind/vg_preload_core.so (0x1B8FF000)
    
22. --2776-- Reading syms from /usr/lib/valgrind/vgpreload_memcheck.so (0x1B902000)
    
23. --2776-- REDIR: 0x1B8F8DB0 (index) redirected to 0x1B905084 (index)
    
24. --2776-- Reading syms from /lib/libsafe.so.2.0.16c (0x1B908000)
    
25. --2776--    object doesn't have a symbol table
    
26. --2776-- Reading syms from /usr/lib/libstdc++.so.6.0.6 (0x1B934000)
    
27. --2776--    object doesn't have a symbol table
    
28. --2776-- Reading syms from /lib/libm-2.3.90.so (0x1BA1D000)
    
29. --2776-- Reading syms from /lib/libgcc_s-4.0.1-20050831.so.1 (0x1BA42000)
    
30. --2776--    object doesn't have a symbol table
    
31. --2776-- Reading syms from /lib/libc-2.3.90.so (0x1BA4C000)
    
32. --2776-- Reading syms from /lib/libdl-2.3.90.so (0x1BB78000)
    
33. --2776-- REDIR: 0x1BAB6D60 (rindex) redirected to 0x1B904F94 (rindex)
    
34. --2776-- REDIR: 0x1B8E47F0 (_dl_sysinfo_int80) redirected to 0xB0022B03 (???)
    
35. --2776-- REDIR: 0x1BAB1BFE (malloc) redirected to 0x1B903793 (malloc)
    
36. --2776-- REDIR: 0x1BAAFDC8 (free) redirected to 0x1B9042B2 (free)
    
37. --2776-- REDIR: 0x1BAB7BD0 (memset) redirected to 0x1B9053D8 (memset)
    
38. --2776-- REDIR: 0x1B9E8890 (operator new(unsigned)) redirected to 0x1B903B2E (operator new(unsigned))
    
39. --2776-- REDIR: 0x1BAB8030 (memcpy) redirected to 0x1B905490 (memcpy)
    
40. --2776-- REDIR: 0x1B9E7308 (operator delete(void*)) redirected to 0x1B904546 (operator delete(void*))
    
41. ==2776==
    
42. ==2776== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 26 from 2)
    
43. --2776--
    
44. --2776-- supp:    3 index-not-intercepted-early-enough-HACK-1
    
45. --2776-- supp:   23 dl_relocate_object
    
46. ==2776== malloc/free: in use at exit: 40417236 bytes in 9920 blocks.
    
47. ==2776== malloc/free: 9951 allocs, 31 frees, 57194292 bytes allocated.
    
48. ==2776==
    
49. ==2776== searching for pointers to 9920 not-freed blocks.
    
50. ==2776== checked 40536016 bytes.
    
51. ==2776==
    
52. ==2776== LEAK SUMMARY:
    
53. [color=red]==2776==    definitely lost: 0 bytes in 0 blocks.
    
54. ==2776==      possibly lost: 0 bytes in 0 blocks.
    
55. ==2776==    still reachable: 40417236 bytes in 9920 blocks.
    
56. ==2776==         suppressed: 0 bytes in 0 blocks.
    
57. ==2776== Reachable blocks (those to which a pointer was found) are not shown.[/color]
    
58. ==2776== To see them, rerun with: --show-reachable=yes
    
59. --2776--  memcheck: sanity checks: 5229 cheap, 210 expensive
    
60. --2776--  memcheck: auxmaps: 0 auxmap entries (0k, 0M) in use
    
61. --2776--  memcheck: auxmaps: 0 searches, 0 comparisons
    
62. --2776--  memcheck: secondaries: 765 issued (48960k, 47M)
    
63. --2776--  memcheck: secondaries: 33 accessible and distinguished (2112k, 2M)
    
64. --2776--     tt/tc: 5649 tt lookups requiring 5788 probes
    
65. --2776--     tt/tc: 5649 fast-cache updates, 3 flushes
    
66. --2776-- translate: new        2706 (60927 ->; 991661; ratio 162:10) [0 scs]
    
67. --2776-- translate: dumped     0 (0 ->; ??)
    
68. --2776-- translate: discarded  2 (95 ->; ??)
    
69. --2776-- scheduler: 261466397 jumps (bb entries).
    
70. --2776-- scheduler: 5229/18233 major/minor sched events.
    
71. --2776--    sanity: 5230 cheap, 210 expensive checks.
    
72. --2776--    exectx: 4999 lists, 26 contexts (avg 0 per list)
    
73. --2776--    exectx: 10008 searches, 9982 full compares (997 per 1000)
    
74. --2776--    exectx: 0 cmp2, 81 cmp4, 0 cmpAll
    

復(fù)制代碼