对c ++来说很新，但是我在白天和我一起玩过valgrind。除了使用外部库（xqilla）的部件之外，我的代码清理得很好。我可以看到有内存泄漏和错误。 这是否意味着我应该查看一个不同的库，或者图书馆是否存在错误和小漏洞，我不应该关心它？
==8779== Syscall param sendmsg(mmsg[0].msg_hdr) points to uninitialised byte(s)
==8779== at 0x6065829: sendmmsg (sendmmsg.c:32)
==8779== by 0x767C8FD: __libc_res_nsend (res_send.c:1140)
==8779== by 0x7679D48: __libc_res_nquery (res_query.c:226)
==8779== by 0x767A6F8: __libc_res_nsearch (res_query.c:582)
==8779== by 0x746CB57: _nss_dns_gethostbyname4_r (dns-host.c:314)
==8779== by 0x6035ADF: gaih_inet (getaddrinfo.c:849)
==8779== by 0x6039913: getaddrinfo (getaddrinfo.c:2473)
==8779== by 0x50B3F06: xercesc_3_1::UnixHTTPURLInputStream::UnixHTTPURLInputStream(xercesc_3_1::XMLURL const&, xercesc_3_1::XMLNetHTTPInfo const*) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x50B3A44: xercesc_3_1::SocketNetAccessor::makeNew(xercesc_3_1::XMLURL const&, xercesc_3_1::XMLNetHTTPInfo const*) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x4F8813A: xercesc_3_1::XMLURL::makeNewStream() const (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x4FEC737: xercesc_3_1::ReaderMgr::createReader(xercesc_3_1::InputSource const&, bool, xercesc_3_1::XMLReader::RefFrom, xercesc_3_1::XMLReader::Types, xercesc_3_1::XMLReader::Sources, bool, unsigned long) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x4FE6758: xercesc_3_1::IGXMLScanner::scanReset(xercesc_3_1::InputSource const&) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== Address 0x7feffdad0 is on thread 1's stack
==8779==
Hello
Good
Bye
==8779==
==8779== HEAP SUMMARY:
==8779== in use at exit: 11 bytes in 1 blocks
==8779== total heap usage: 8,144 allocs, 8,143 frees, 1,957,496 bytes allocated
==8779==
==8779== 11 bytes in 1 blocks are definitely lost in loss record 1 of 1
==8779== at 0x4C2A879: operator new(unsigned long) (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==8779== by 0x4FEC508: xercesc_3_1::MemoryManagerImpl::allocate(unsigned long) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x50B87D5: xercesc_3_1::IconvGNULCPTranscoder::transcode(unsigned short const*, xercesc_3_1::MemoryManager*) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x50B3C6D: xercesc_3_1::UnixHTTPURLInputStream::UnixHTTPURLInputStream(xercesc_3_1::XMLURL const&, xercesc_3_1::XMLNetHTTPInfo const*) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x50B3A44: xercesc_3_1::SocketNetAccessor::makeNew(xercesc_3_1::XMLURL const&, xercesc_3_1::XMLNetHTTPInfo const*) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x4F8813A: xercesc_3_1::XMLURL::makeNewStream() const (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x4FEC737: xercesc_3_1::ReaderMgr::createReader(xercesc_3_1::InputSource const&, bool, xercesc_3_1::XMLReader::RefFrom, xercesc_3_1::XMLReader::Types, xercesc_3_1::XMLReader::Sources, bool, unsigned long) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x4FE6758: xercesc_3_1::IGXMLScanner::scanReset(xercesc_3_1::InputSource const&) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x4FE0D03: xercesc_3_1::IGXMLScanner::scanDocument(xercesc_3_1::InputSource const&) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x50053B8: xercesc_3_1::XMLScanner::scanDocument(unsigned short const*) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x5008931: xercesc_3_1::XMLScanner::scanDocument(char const*) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779== by 0x5027474: xercesc_3_1::AbstractDOMParser::parse(char const*) (in /usr/lib/x86_64-linux-gnu/libxerces-c-3.1.so)
==8779==
==8779== LEAK SUMMARY:
==8779== definitely lost: 11 bytes in 1 blocks
==8779== indirectly lost: 0 bytes in 0 blocks
==8779== possibly lost: 0 bytes in 0 blocks
==8779== still reachable: 0 bytes in 0 blocks
==8779== suppressed: 0 bytes in 0 blocks
==8779==
==8779== For counts of detected and suppressed errors, rerun with: -v
==8779== Use --track-origins=yes to see where uninitialised values come from
==8779== ERROR SUMMARY: 5 errors from 2 contexts (suppressed: 2 from 2)
答案 0 :(得分：1)
如果库确实泄漏了内存，那么解决它是明智的，因为泄漏最终会影响应用程序。内存泄漏的典型症状包括挂起进程或突然终止的进程（例如Linux内存杀手凶手开始工作时）。
寻找另一个图书馆可能是可行的。如果可以联系到图书馆维护者，最好引起他们的注意。并且，如果它是开源的，那么跟踪它并提交修复将会更加棒极了。
虽然有一件事需要考虑。 Valgrind将标记任何最终未作为泄漏发布的内存。图书馆可能正在进行一次性分配，例如创建单身人士。如果是这样，那么这确实是一个无问题。
所以，要尝试的事情：
指向未初始化字节的指针的错误并不一定意味着任何错误，只是库已经分配了指针而没有将分配的内存设置为任何值。如果程序遇到分段错误，那么这些错误可能有助于跟踪它们。否则，这可能是完全正常的。例如，库可能是预分配缓冲区以供以后使用。
同样，我会考虑向维护者提及它们，但这些错误本身并不太令人担忧。