-
Notifications
You must be signed in to change notification settings - Fork 418
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add lifecycle subscriber to lifecycle nodes #2254
Open
carmiac
wants to merge
2
commits into
ros2:rolling
Choose a base branch
from
carmiac:rolling
base: rolling
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
Changes from all commits
Commits
Show all changes
2 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
83 changes: 83 additions & 0 deletions
83
rclcpp_lifecycle/include/rclcpp_lifecycle/lifecycle_subscriber.hpp
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,83 @@ | ||
| // Copyright 2023 Elroy Air, Inc. | ||
| // | ||
| // Licensed under the Apache License, Version 2.0 (the "License"); | ||
| // you may not use this file except in compliance with the License. | ||
| // You may obtain a copy of the License at | ||
| // | ||
| // http://www.apache.org/licenses/LICENSE-2.0 | ||
| // | ||
| // Unless required by applicable law or agreed to in writing, software | ||
| // distributed under the License is distributed on an "AS IS" BASIS, | ||
| // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. | ||
| // See the License for the specific language governing permissions and | ||
| // limitations under the License. | ||
|
|
||
| #ifndef RCLCPP_LIFECYCLE__LIFECYCLE_SUBSCRIBER_HPP_ | ||
| #define RCLCPP_LIFECYCLE__LIFECYCLE_SUBSCRIBER_HPP_ | ||
|
|
||
| #include <memory> | ||
| #include <string> | ||
|
|
||
| #include "rclcpp/subscription.hpp" | ||
|
|
||
| namespace rclcpp_lifecycle | ||
| { | ||
|
|
||
| /// @brief Child class of rclcpp::Subscription that adds lifecycle functionality. | ||
| /** | ||
| * This class is a child class of rclcpp::Subscription that adds a lifecycle | ||
| * check to the callback function. If the node is in an inactive state, the | ||
| * callback will not be called. | ||
| */ | ||
| template< | ||
| typename MessageT, | ||
| typename AllocatorT = std::allocator<void>, | ||
| typename SubscribedT = typename rclcpp::TypeAdapter<MessageT>::custom_type, | ||
| typename ROSMessageT = typename rclcpp::TypeAdapter<MessageT>::ros_message_type, | ||
| typename MessageMemoryStrategyT = rclcpp::message_memory_strategy::MessageMemoryStrategy< | ||
| ROSMessageT, | ||
| AllocatorT | ||
| >> | ||
| class LifecycleSubscription : public SimpleManagedEntity, | ||
| public rclcpp::Subscription<MessageT, AllocatorT> | ||
| { | ||
| private: | ||
| using SubscriptionTopicStatisticsSharedPtr = | ||
| std::shared_ptr<rclcpp::topic_statistics::SubscriptionTopicStatistics<ROSMessageT>>; | ||
|
|
||
| public: | ||
| LifecycleSubscription( | ||
| rclcpp::node_interfaces::NodeBaseInterface * node_base, | ||
| const rosidl_message_type_support_t & type_support_handle, | ||
| const std::string & topic_name, | ||
| const rclcpp::QoS & qos, | ||
| rclcpp::AnySubscriptionCallback<MessageT, AllocatorT> callback, | ||
| const rclcpp::SubscriptionOptionsWithAllocator<AllocatorT> & options, | ||
| typename MessageMemoryStrategyT::SharedPtr message_memory_strategy, | ||
| SubscriptionTopicStatisticsSharedPtr subscription_topic_statistics = nullptr) | ||
| : rclcpp::Subscription<MessageT>( | ||
| node_base, | ||
| type_support_handle, | ||
| topic_name, | ||
| qos, | ||
| callback, | ||
| options, | ||
| message_memory_strategy, | ||
| subscription_topic_statistics) | ||
| { | ||
| } | ||
|
|
||
| /// Check if we need to handle the message, and execute the callback if we do. | ||
| void handle_message( | ||
| std::shared_ptr<void> & message, const rclcpp::MessageInfo & message_info) override | ||
| { | ||
| if (!this->is_activated()) { | ||
| return; | ||
| } | ||
| rclcpp::Subscription<MessageT>::handle_message(message, message_info); | ||
| } | ||
| }; | ||
|
|
||
| } // namespace rclcpp_lifecycle | ||
|
|
||
| #endif // RCLCPP_LIFECYCLE__LIFECYCLE_SUBSCRIBER_HPP_ | ||
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
with current implementation, if the node is
deactivated, all messages during that state will be ignored and dropped on the subscription because the message is already taken. i am not sure if this is the expected behavior, since application would expect that the subscription can keep the message up to the specified history depth?CC: @ros2/team
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, this is an extremely good point; at this level, we are fetching it out of the RMW but then dropping it on the floor and not calling the user callback.
This gets into a bit of the philosophy behind the lifecycle transitions. I would expect that when a
LifecycleSubscriptionis deactivated, then it is actually not part of the graph at all anymore, i.e.handle_messagewould never be called. That releases the most resources in the system, though it does have a possible side-effect of a "thundering herd" problem if a bunch of them are activated at once.All of that said, I wasn't really involved with any of the initial concepts with the lifecycle code, so I don't fully know what the expectation is. Maybe @wjwwood could chime in here with some ideas?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We discussed this in the weekly meeting, and @wjwwood reminded me that there is a difference between the
on_configureand theon_activatecallbacks. Specifically, the former is when subscriptions should be created and do discovery, and the latter is when you should start receiving data. So from that perspective, this PR is actually doing the right thing by doing discovery, but not calling the callback untilon_activateis called.There is another thing to think about here, though, which is what happens with data that comes in from before a subscriber is active. Conceptually there are two ways to go here:
The design document says:
"In the inactive state, any data that arrives on managed topics will not be read and or processed. Data retention will be subject to the configured QoS policy for the topic."
That suggests to me case 1, but this PR actually implements case 2. It seems like there is a case for both of them, so personally I'd be OK with going forward with this implementation and a
TODOthat says implementing case 1 is future work.@fujitatomoya @carmiac Does this all make sense to you? Thoughts?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think that this approach provides very little improvement in terms of performance.
By the time you are dropping the message, you already wasted CPU cycles to go through all the ROS and network stack and potentially copying the message multiple times.
I'm not opposed to the change, as long as it's just a starting point from which we can later re-implement lifecycle subscriptions/services to effectively disconnect the entity.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
for me, this design makes sense. that is the responsibility for QoS but LifecycleNode state. user can configure the behavior with QoS for LifecycleSubscription.
agree.
IMO, we need to add a new state for subscription. (the one is
created and discovery, and the other isstart taking the data from RMW queue(or wait_set is added))i am good to go for now, with follow-up issue for this TODO.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I agree.
We are going to soon publish a proposal for updated lifecycle management and this is a topic we discussed while creating it.
Note that a similar concept applies also to non-lifecycle entities: a subscription may be visible but not added to any executor (or the executor may be not spinning), effectively making it "inactive"