Zard-C

Seems it crashed in rclcpp, since you are using ` Iron Irwini compiled from source`, I think you could build a debug version of rclcpp package, reproduce this bug, and...

> > Btw, I noticed that you are using jemalloc > > Are there choices? I mean, I didn't know I can have other choices here... can you explain me...

> Seems that in `rcl_wait_set_resize` something goes wrong. I'm also rebuilding `rcl` to dig more... Hi, I believe the crash is related to memory deallocation. To address this, you can...

I've tried your code in `iron-desktop` docker, unfortunately, I can't reproduce: I modified your launch.py ```python # prefix="tmux split-window gdb -ex run --args", # run then with "tmux new-session ros2...

> 是否有手段测量到替换前后性能的变化。以便确认，替换后性能是提升的 1. 最简单的方案，在linux下使用`time`，使用相同的测试输入运行一定时间查看user/sys cpu 差别，使用mimalloc之后sys应该会下降

Hello @Chaochao0215, maybe we could build a debug type of this source code and use `gdb` to backtrace when it crashed

I tried your code and I can't reproduce the segment fault. However, in gdb backtrace it seems like a multi-thread problem, rebuild your program with ``` cmake set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=thread...

> 6\. send the message subscribed in the new shell terminal, such as 'ros2 topic pub /test_callback_group std_msgs/msg/String "data: hello" I missed this step.

Hi guys, I would like to try it with `[rclc]` demo nodes if we had a c version of typesupport, (I've checked https://github.com/ros2/rosidl_typesupport_fastrtps/pull/106 and https://github.com/ros2/rosidl/pull/758, but I didn't find out...

hi @SteveMacenski, glad to see this draft, I think that `sched_setscheduler` api is only for linux system, but for MacOS or Windows, they may have their own apis to set...