mirror of
https://github.com/mutouyun/cpp-ipc.git
synced 2025-12-06 16:56:45 +08:00
Problem: The previous fix unconditionally called pthread_mutex_unlock() at the beginning of close(), which could interfere with other threads/processes that still had valid references to the mutex. This caused test failures on FreeBSD when running tests multiple times (ShmTest.RemoveByName would fail on the second run). Root cause: Calling unlock() too early could affect the mutex state for other references that are still using it, leading to unexpected behavior. Solution: Move pthread_mutex_unlock() to only be called when we're about to destroy the mutex (i.e., when we're the last reference: shm_->ref() <= 1 && self_ref <= 1). This ensures: 1. We don't interfere with other threads/processes using the mutex 2. We still unlock before destroying to avoid FreeBSD robust list issues 3. The unlock happens at the correct time - right before pthread_mutex_destroy() This is the correct approach because: - Only the last reference holder should clean up the mutex - Unlocking should be paired with destroying for the final cleanup - Other references should not be affected by one reference closing Fixes the second-run test failure on FreeBSD while maintaining the segfault fix. |
||
|---|---|---|
| .github/workflows | ||
| 3rdparty | ||
| demo | ||
| include/libipc | ||
| src | ||
| test | ||
| .gitignore | ||
| CMakeLists.txt | ||
| LICENSE | ||
| performance.xlsx | ||
| README.md | ||
cpp-ipc (libipc) - C++ IPC Library
A high-performance inter-process communication library using shared memory on Linux/Windows/FreeBSD.
- Compilers with C++17 support are recommended (msvc-2017/gcc-7/clang-4)
- No other dependencies except STL.
- Only lock-free or lightweight spin-lock is used.
- Circular array is used as the underline data structure.
ipc::routesupports single write and multiple read.ipc::channelsupports multiple read and write. (Note: currently, a channel supports up to 32 receivers, but there is no such a limit for the sender.)- Broadcasting is used by default, but user can choose any read/ write combinations.
- No long time blind wait. (Semaphore will be used after a certain number of retries.)
- Vcpkg way of installation is supported. E.g.
vcpkg install cpp-ipc
Usage
See: Wiki
Performance
| Environment | Value |
|---|---|
| Device | Lenovo ThinkPad T450 |
| CPU | Intel® Core™ i5-4300U @ 2.5 GHz |
| RAM | 16 GB |
| OS | Windows 7 Ultimate x64 |
| Compiler | MSVC 2017 15.9.4 |
Unit & benchmark tests: test
Performance data: performance.xlsx
Reference
- Lock-Free Data Structures | Dr Dobb's
- Yet another implementation of a lock-free circular array queue | CodeProject
- Lock-Free 编程 | 匠心十年 - 博客园
- 无锁队列的实现 | 酷 壳 - CoolShell
- Implementing Condition Variables with Semaphores
使用共享内存的跨平台(Linux/Windows/FreeBSD,x86/x64/ARM)高性能IPC通讯库
- 推荐支持C++17的编译器(msvc-2017/gcc-7/clang-4)
- 除STL外,无其他依赖
- 无锁(lock-free)或轻量级spin-lock
- 底层数据结构为循环数组(circular array)
ipc::route支持单写多读,ipc::channel支持多写多读【注意:目前同一条通道最多支持32个receiver,sender无限制】- 默认采用广播模式收发数据,支持用户任意选择读写方案
- 不会长时间忙等(重试一定次数后会使用信号量进行等待),支持超时
- 支持Vcpkg方式安装,如
vcpkg install cpp-ipc
使用方法
详见:Wiki
性能
| 环境 | 值 |
|---|---|
| 设备 | 联想 ThinkPad T450 |
| CPU | 英特尔® Core™ i5-4300U @ 2.5 GHz |
| 内存 | 16 GB |
| 操作系统 | Windows 7 Ultimate x64 |
| 编译器 | MSVC 2017 15.9.4 |
单元测试和Benchmark测试: test
性能数据: performance.xlsx