zhenshenglee/ros2_shm_msgs

ROS2 shm_msg based zero-copy for image and pointcloud.

ros2_shm_msgs's Issues

[FEA]Support docker env.

docker env for humble with cuda.

[FEA] Add gpu processing demo with type-adaption.

the info of type-adaption is in #14

for transport

• intra-process cuda(with cv-bridge)
• intra-process-zero-copy opencv-cuda(GpuMat)
• intra-type-adaption cuda

for gpu implementation

• cuda-c (for jetson utils and other raw processing algorithms)
  - [ ] thrust (for thrust based algorithms like cupoch)
• opencv-cuda(for opencv algorithms like opencv)

[FEA] add support to compressed-image msg type.

[BUG] root is not in the sudoers file when starting container.

[FEA] create demo with message filter time sync or N:1 pattern.

[OPT] Apply Optimized ROS->PCL conversion

https://discourse.ros.org/t/optimized-ros-pcl-conversion/25833

open3d not support in docker rolling

Can we eliminate the dependence on open3d? I want to reproduce it in docker, but it's stuck.

零拷贝发送与接受是一个地址吗

为啥我测试着不是一个地址呢

[QST] zero-init of loaned_msg?

track this with MatthiasKillat/ros2_shm_vision_demo#12

[BUG] qt apps starting failed when in docker containers.

[QST] shm_msg based zero-copy, intra-process based zero-copy, and type adaption.

@ZhenshengLee thank you so much for your information and quick helps. May I ask you some questions related to ros2_shm_msgs?

• In folder "intra", i found that you added intra type adaption within cpu. Currently, ros2_shm_msgs only support type adaption (within CPU) for image only, but not for pointcloud?
• Is it correct if "zero-copy" in ros2_shm_msgs called "minimal copy" when we publish topics?
• In your opinion, whether the performance (regarding to speed and CPU consumption) much better if we use type adaption?

Even I did not use https://github.com/IntelRealSense/realsense-ros#efficient-intra-process-communication, the performance of ROS2 Realsense Wrapper (regarding to speed and CPU consumption) is almost same with ros2_shm-msgs. I am not sure the hints inside since i am quite new in this topic. Maybe, my tested point-cloud data is small (camera resolution 640x480, with decimation filter).

[FEA] add cpu metrics.

[BUG] use std::array size rather than DATA_MAX_SIZE macro.

https://stackoverflow.com/questions/69299925/std-array-difference-between-size-and-max-size

[CLEAN] use memcpy to replace std::copy to get a better performance.

Is it reasonable?

https://stackoverflow.com/questions/4707012/is-it-better-to-use-stdmemcpy-or-stdcopy-in-terms-to-performance

how about std::move

is there something like memmove?

请问 opencv_conversions.hpp 定义的 toCVshare 相对于 cvbridge 的 toCVshare 有什么区别么

抱歉在issue 里提出这种问题，请见谅

opencv_conversions.hpp 定义的 toCVshare 相对于 cvbridge 的 toCVshare 是之进行了多次重载么？
我目前面临的问题是可能需要 使用 python 版的 opencv_conversions.hpp 实现的 toCVshare 方法，不知道是否 该项目 有 python 的实现，亦或能指点一下，谢谢

Linker errors while compiling with Open3D

Hi
I am getting linker errors with Open3D. I tried adding rclcpp in cmakelists but it doesn't help
Following is my build environment:

• Ubuntu 22.04
• Ros Humble
• Opencv 4.6.0
  err.txt

Please help

[FEA] add type adaptation feature in Humble.

https://github.com/ros-infrastructure/rep/blob/master/rep-2007.rst

have some issues when i compile the project

err.txt

The DATA_MAX_SIZE parameter is set too large， causing memory allocation errors

Hi,

thanks for the awesome examples.

i tried to create a PointCloud240m.msg,and change uint32 DATA_MAX_SIZE=251658240.
when i run ./pcl240m_talker
error i have got:
'''

terminate called after throwing an instance of 'std::bad_alloc'
what(): std::bad_alloc
Aborted (core dumped)

'''
it seems DATA_MAX_SIZE is set too large

Any idea how to publish the pcd file with 240M ?

[FEA] Add intra-process demo with image.

ordinary node

• image-intra-talker
• image-intra-listener
• compose demo

component

[FEA]Set open3d as optional component.

motivation

• As PCL is used usually.
• Open3d is not available in the ros docker image #1

consideration

using CMake to check if open3d is installed.

[BUG] adding another user failed when starting docker container.

using this sw means agreeing to the license agreement above
[INFO] Start docker container based on local image : ga_team/ga_ros:22.04
[INFO] Starting docker container "ga_22.04_neousys" ...
++ local_volumes
++ set +x
+ docker run --gpus all -it -d --privileged --name ga_22.04_neousys -e DOCKER_USER=neousys -e USER=neousys -e DOCKER_USER_ID=1000 -e DOCKER_GRP=neousys -e DOCKER_GRP_ID=1000 -e DOCKER_IMG=ga_team/ga_ros:22.04 -e USE_GPU=1 -e NVIDIA_VISIBLE_DEVICES=all -e NVIDIA_DRIVER_CAPABILITIES=compute,video,utility -v /home/neousys/zs_ws/ga_ros:/ga_ros -v /home/neousys/.cache:/home/neousys/.cache -v /dev:/dev -v /media:/media -v /tmp/.X11-unix:/tmp/.X11-unix:rw -v /etc/localtime:/etc/localtime:ro -v /usr/src:/usr/src -v /lib/mgaules:/lib/mgaules --net host -w /ga_ros --add-host in_dev_docker:127.0.0.1 --add-host neousys-Nuvo-8208GC-Series:127.0.0.1 --hostname in_dev_docker --shm-size 2G --pid=host -v /dev/null:/dev/raw1394 ga_team/ga_ros:22.04 /bin/bash
07bc0a9a3856c41fcc2906562d3c0b935f038c97adfd04859af78560d2addb36
+ '[' 0 -ne 0 ']'
+ set +x
addgroup: The GID `1000' is already in use.
usermod: user 'neousys' does not exist
root is not in the sudoers file.  This incident will be reported.
root is not in the sudoers file.  This incident will be reported.
root is not in the sudoers file.  This incident will be reported.
root is not in the sudoers file.  This incident will be reported.
chown: invalid user: 'neousys:neousys'
chown: invalid user: 'neousys:neousys'
[ OK ] Finished setting up ga_team/ga ros environment. Now you can enter with: 
bash docker/scripts/ga_into.sh 
[ OK ] Enjoy!

a workaround is to change ga_into.sh to be like the follow

docker exec \
    -u zs \
    -e HISTFILE=/ga_ros/.dev_bash_hist \
    -it ga_${GA_DIST}_$USER \
    /bin/bash

[CLEAN] using function templates for multiple shm msg types.

• https://github.com/ZhenshengLee/ros2_shm_msgs/blob/master/include/shm_msgs/open3d_conversions.hpp
• https://github.com/ZhenshengLee/ros2_shm_msgs/blob/master/include/shm_msgs/opencv_conversions.hpp
• https://github.com/ZhenshengLee/ros2_shm_msgs/blob/master/include/shm_msgs/point_cloud2_iterator.hpp

GpuMat invalid argument

Hey,

thanks a lot for the awesome examples.
I tried your cuda GpuMat examples and it is working fine.
Doing stuff with the GpuMat at the listener side is somehow impossible.
Functions like download, clone(), etc. yield an invalid argument error.
The gpu_mat is not empty.

Any idea how to access the data of the GpuMat at listener side?