[ahcorde]

Description

After a backward time jump, Rate::sleep() correctly schedules the next wakeup at now + period, but then advances last_interval_ from the old timeline (last_interval_ += period_). Every following call sees now < last_interval_ again and sleeps a full period after the work finishes, so the loop runs at period + work time until the clock catches up (forever with a looping bag).

• Set last_interval_ = next_interval. Without a time jump this is the same value as before.
• Add a test using ROS time override that jumps back and checks the next wakeup is one period after the previous one.

Did you use Generative AI?

Claude Opus 5.5

[github-actions]

ABI Compliance Check

❌ Verdict: incompatible

Library	Verdict	Summary
libcomponent_manager.so	✅ compatible	No ABI changes detected.
librclcpp.so	❌ incompatible	ABI changed: 0 removed, 1 changed, 0 added declaration(s); removed or changed declarations break backward compatibility.
librclcpp_action.so	✅ compatible	No ABI changes detected.
librclcpp_lifecycle.so	✅ compatible	No ABI changes detected.

✅ libcomponent_manager.so — full abidiff report

Compared:

• Base: lib-base/libcomponent_manager.so
• Head: lib-pr/libcomponent_manager.so @ e909a2b

(empty report — no differences printed by abidiff)

❌ librclcpp.so — full abidiff report

Compared:

• Base: lib-base/librclcpp.so
• Head: lib-pr/librclcpp.so @ e909a2b

Functions changes summary: 0 Removed (45 filtered out), 1 Changed (1722 filtered out), 0 Added (20 filtered out) functions
Variables changes summary: 0 Removed, 0 Changed, 0 Added variable

1 function with some indirect sub-type change:

  [C] 'method rclcpp::node_interfaces::NodeParameters::NodeParameters(rclcpp::node_interfaces::NodeBaseInterface::SharedPtr, rclcpp::node_interfaces::NodeLoggingInterface::SharedPtr, rclcpp::node_interfaces::NodeTopicsInterface::SharedPtr, rclcpp::node_interfaces::NodeServicesInterface::SharedPtr, rclcpp::node_interfaces::NodeClockInterface::SharedPtr, const std::vector<rclcpp::Parameter, std::allocator<rclcpp::Parameter> >&, bool, bool, const rclcpp::QoS&, const rclcpp::PublisherOptionsBase&, bool, bool)' at node_parameters.cpp:67:1 has some indirect sub-type changes:
    implicit parameter 0 of type 'rclcpp::node_interfaces::NodeParameters*' has sub-type changes:
      in pointed to type 'class rclcpp::node_interfaces::NodeParameters' at node_parameters.hpp:83:1:
        type size changed from 2496 to 2624 (in bits)
        no member function changes (24 filtered);
        1 data member insertion:
          'rclcpp::node_interfaces::NodeLoggingInterface::SharedPtr node_logging_', at offset 2368 (in bits) at node_parameters.hpp:260:1
        1 data member changes (1 filtered):
          'rclcpp::node_interfaces::NodeClockInterface::SharedPtr node_clock_' offset changed from 2368 to 2496 (in bits) (by +128 bits)

✅ librclcpp_action.so — full abidiff report

Compared:

• Base: lib-base/librclcpp_action.so
• Head: lib-pr/librclcpp_action.so @ e909a2b

(empty report — no differences printed by abidiff)

✅ librclcpp_lifecycle.so — full abidiff report

Compared:

• Base: lib-base/librclcpp_lifecycle.so
• Head: lib-pr/librclcpp_lifecycle.so @ e909a2b

(empty report — no differences printed by abidiff)

_{Updated for commit e909a2b · check run · suppressions: /home/runner/work/_temp/ros2-abi-suppressions.txt}

[fujitatomoya]

i think this is racy condition. should we hold the mutex clock->get_clock_mutex() to avoid thta?