[英]exception in boost:interprocess, shared memory object deletion
there is a simple piece of code which I moved to main function of sample app to avoid the affecting of any other application parts: 我将一小段代码移到示例应用程序的主要功能,以避免影响其他任何应用程序部分:
namespace bi = boost:interprocess;
bi::shared_memory_object::remove("shm");
bi::named_mutex::remove("mtx");
bi::named_condition::remove("cnd");
just first line raise an exception in internal boost interprocess code: in windows_intermodule_singleton.hpp, 118 (v 1.60.0) bolded line is where it's happening. 仅第一行在内部boost进程间代码中引发了异常:在windows_intermodule_singleton.hpp中,118(v 1.60.0)粗体行是发生这种情况的地方。 Win64, VC v19, Windows 10
Win64,VC v19,Windows 10
...
caster.addr = m;
BOOST_ASSERT((caster.addr_uint64 & boost::uint64_t(3)) == 0);
max_count = boost::uint32_t(caster.addr_uint64 >> 32);
**initial_count = boost::uint32_t(caster.addr_uint64);**
initial_count = initial_count/4;
//Make sure top two bits are zero
BOOST_ASSERT((max_count & boost::uint32_t(0xC0000000)) == 0);
//Set quasi-top bit
max_count |= boost::uint32_t(0x40000000);
...
ANy idea why this happens? 知道为什么会这样吗? can this be related with x64 memory model?
这可以与x64内存模型相关吗?
这不是一个例外,我不注意,使用所有/ RTC编译器设置进行编译,并且该位置看起来像潜在的数据丢失(?)8字节至4。windbg报告:_RTC_Check_8_to_4
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.