c++ - BOOST_LOG_SEV 在创建线程后破坏 async_read_some

Question

所以我在练习 boost 库时遇到了一些问题。我的应用程序使用 asio::serial_port，我决定添加一些日志功能。问题来了。

当我为 asio::io_service 创建线程时

boost::thread t(boost::bind(&boost::asio::io_service::run, &io_service_));

然后尝试用

记录一些东西

BOOST_LOG_SEV(MBLog::get(), logging::trivial::trace) << "something to trace;"

调用之前

port_->async_read_some(
        boost::asio::buffer(read_buf_raw_, SERIAL_PORT_READ_BUF_SIZE),
        boost::bind(
            &SerialPort::on_receive_,
            this, boost::asio::placeholders::error,
            boost::asio::placeholders::bytes_transferred));

然后处理程序 on_receive_永远不会被调用(端口仍然可以用 write_some thow 写一些东西)。当记录宏被注释掉或在执行 async_read_some 后完成时它神奇地工作。我想知道这背后的魔力是什么，或者从哪里开始寻找答案？也许你们中的一些人已经知道这个问题然后请分享。其他有用信息:

我正在使用 MSVC2017。

提升日志初始化

Logger.h

#define _HAS_AUTO_PTR_ETC 1
#include <boost/log/trivial.hpp>
#include <boost/log/sources/global_logger_storage.hpp>

namespace logging = boost::log;
namespace src = boost::log::sources;
namespace sinks = boost::log::sinks;
namespace keywords = boost::log::keywords;
namespace attrs = boost::log::attributes;
namespace expr = boost::log::expressions;

BOOST_LOG_GLOBAL_LOGGER(MBLog, src::severity_logger_mt<logging::trivial::severity_level>)

Logger.cpp

#include "Logger.h"

#include <boost/log/support/date_time.hpp>
#include <boost/log/expressions.hpp>
#include <boost/log/utility/setup/file.hpp>
#include <boost/log/utility/setup/console.hpp>
#include <boost/log/utility/setup/common_attributes.hpp>

void full_detail_formatter(logging::record_view const& rec, logging::formatting_ostream& strm);

void normal_formatter(logging::record_view const& rec, logging::formatting_ostream& strm);

BOOST_LOG_GLOBAL_LOGGER_INIT(MBLog, src::severity_logger_mt<logging::trivial::severity_level>)
{
    src::severity_logger_mt<logging::trivial::severity_level> lg;

    logging::add_common_attributes();

    auto file_sink = logging::add_file_log(
        keywords::file_name = "./logs/L%Y-%m-%d_%3N.log", // one file per date and execution
        keywords::rotation_size = 5 * 1024 * 1024,      // rotate after 5MB
        keywords::target = "./logs",
        keywords::min_free_space = 30 * 1024 * 1024,
        keywords::auto_flush = true,
        keywords::scan_method = boost::log::sinks::file::scan_matching
    );

    file_sink->set_formatter(&full_detail_formatter);
    file_sink->set_filter(
        logging::trivial::severity >= logging::trivial::trace
    );

    auto console_sink = logging::add_console_log(std::clog);
    console_sink->set_formatter(&full_detail_formatter);
    console_sink->set_filter(
        logging::trivial::severity >= logging::trivial::trace
    );

    logging::core::get()->set_filter(
        logging::trivial::severity >= logging::trivial::trace
    );

    return lg;
}

编辑

似乎有人已经解决了这个问题here通过为 asio::io_service::run 创建线程在执行 async_read_some 之后

Answer 1

Seems like someone already solved this problem here by creating thread for asio::io_service::run after execution of async_read_some

是的。否则 run() 可能会在任何工作被注册之前完成。

它的文档在这里:http://www.boost.org/doc/libs/1_65_0/doc/html/boost_asio/reference/io_service/run/overload1.html

The run() function blocks until all work has finished and there are no more handlers to be dispatched, or until the io_service has been stopped.

可以找到更多解释:

• Confused when boost::asio::io_service run method blocks/unblocks

考虑使用 io_service::work 保留服务:boost asio post not working , io_service::run exits right after post