Fixing future::wait_until (and wait_for) to return once future was made ready#7367
Open
hkaiser wants to merge 2 commits into
Open
Fixing future::wait_until (and wait_for) to return once future was made ready#7367hkaiser wants to merge 2 commits into
hkaiser wants to merge 2 commits into
Conversation
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Repository UI Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (8)
🚧 Files skipped from review as they are similar to previous changes (7)
📝 WalkthroughSummary by CodeRabbit
WalkthroughTimed suspension APIs now accept move-only predicates and return ChangesPredicate-aware timed suspension
Scoped lock guards
Estimated code review effort: 4 (Complex) | ~60 minutes Sequence Diagram(s)sequenceDiagram
participant FutureData
participant ConditionVariable
participant ExecutionAgent
participant WaitPredicate
FutureData->>ConditionVariable: wait_until with readiness predicate
ConditionVariable->>ExecutionAgent: sleep_until with wait_cond
ExecutionAgent->>WaitPredicate: evaluate condition
WaitPredicate-->>ExecutionAgent: true or false
ExecutionAgent-->>ConditionVariable: signaled or timeout
ConditionVariable-->>FutureData: wait result
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
Up to standards ✅🟢 Issues
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@libs/core/execution_base/tests/unit/execution_context.cpp`:
- Around line 56-64: Return an explicit threads::thread_restart_state::timeout
value from both sleep_for and sleep_until overrides, matching default_agent
behavior and preventing non-void functions from reaching the end without
returning.
In `@libs/core/futures/src/future_data.cpp`:
- Around line 393-400: Update the lambda assigned to wait_cond in
future_data_base::wait_until by adding the explicit empty parameter list,
changing the capture form from [this] to [this](), so it remains compatible with
C++20.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro Plus
Run ID: e668afd1-f0dc-405a-9f88-f8edd7c77bc2
📒 Files selected for processing (10)
libs/core/execution_base/include/hpx/execution_base/agent_base.hpplibs/core/execution_base/include/hpx/execution_base/agent_ref.hpplibs/core/execution_base/src/agent_ref.cpplibs/core/execution_base/src/this_thread.cpplibs/core/execution_base/tests/unit/execution_context.cpplibs/core/futures/src/future_data.cpplibs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpplibs/core/synchronization/src/detail/condition_variable.cpplibs/core/threading_base/include/hpx/threading_base/execution_agent.hpplibs/core/threading_base/src/execution_agent.cpp
…de ready Signed-off-by: Hartmut Kaiser <hartmut.kaiser@gmail.com>
hkaiser
force-pushed
the
future_wait_until
branch
from
131f15d to
42f4bd0
Compare
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@libs/core/execution_base/include/hpx/execution_base/this_thread.hpp`:
- Around line 78-85: Update the default desc value in this_thread::sleep_until
to identify sleep_until rather than sleep_for, preserving the existing
agent().sleep_until call and diagnostics behavior.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro Plus
Run ID: fdc8c829-1508-4707-95a7-d40b2ea453ad
📒 Files selected for processing (4)
libs/core/execution_base/include/hpx/execution_base/this_thread.hpplibs/core/execution_base/tests/unit/execution_context.cpplibs/core/futures/tests/unit/future.cpplibs/core/synchronization/tests/unit/condition_variable.cpp
hkaiser
force-pushed
the
future_wait_until
branch
from
c1ad7b2 to
8201c2b
Compare
There was a problem hiding this comment.
🧹 Nitpick comments (1)
libs/core/synchronization/src/detail/condition_variable.cpp (1)
250-272: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winNon-predicate
wait_untilduplicates the predicate overload; delegate instead.This body is identical to the predicate overload except for the always-false predicate. Delegating keeps the two in sync and is behavior-preserving: with an always-false predicate
sissignaledon notify (f.ctx_cleared) andtimeouton timer expiry (f.ctx_still set), yielding the same result as the currentf.ctx_ ? timeout : signaled.♻️ Proposed delegation
threads::thread_restart_state condition_variable::wait_until( std::unique_lock<mutex_type>& lock, hpx::chrono::steady_time_point const& abs_time, - char const* /* description */, error_code& /* ec */) - { - HPX_ASSERT_OWNS_LOCK(lock); - - // enqueue the request and block this thread - auto const this_ctx = hpx::execution_base::this_thread::agent(); - queue_entry f(this_ctx, &queue_); - queue_.push_back(f); - - reset_queue_entry r(f); - - { - // suspend this thread - unlock_guard<std::unique_lock<mutex_type>> ul(lock); - this_ctx.sleep_until(abs_time.value(), [] { return false; }); - } - - return f.ctx_ ? threads::thread_restart_state::timeout : - threads::thread_restart_state::signaled; - } + char const* description, error_code& ec) + { + return wait_until(lock, abs_time, + hpx::move_only_function<bool()>([] { return false; }), description, + ec); + }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@libs/core/synchronization/src/detail/condition_variable.cpp` around lines 250 - 272, Refactor the non-predicate condition_variable::wait_until overload to delegate to the predicate-based wait_until overload using an always-false predicate, preserving its existing timeout/signaled behavior. Remove the duplicated queueing and sleep logic from this overload while forwarding the lock, abs_time, description, and error_code arguments.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@libs/core/synchronization/src/detail/condition_variable.cpp`:
- Around line 250-272: Refactor the non-predicate condition_variable::wait_until
overload to delegate to the predicate-based wait_until overload using an
always-false predicate, preserving its existing timeout/signaled behavior.
Remove the duplicated queueing and sleep logic from this overload while
forwarding the lock, abs_time, description, and error_code arguments.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 94716dff-06a9-43d0-ade5-fa0ebd5703c0
📒 Files selected for processing (7)
libs/core/execution_base/include/hpx/execution_base/this_thread.hpplibs/core/execution_base/tests/unit/execution_context.cpplibs/core/futures/tests/unit/future.cpplibs/core/synchronization/include/hpx/synchronization/condition_variable.hpplibs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpplibs/core/synchronization/src/detail/condition_variable.cpplibs/core/synchronization/tests/unit/condition_variable.cpp
🚧 Files skipped from review as they are similar to previous changes (4)
- libs/core/execution_base/include/hpx/execution_base/this_thread.hpp
- libs/core/futures/tests/unit/future.cpp
- libs/core/execution_base/tests/unit/execution_context.cpp
- libs/core/synchronization/tests/unit/condition_variable.cpp
hkaiser
force-pushed
the
future_wait_until
branch
from
8201c2b to
7a913ea
Compare
Collaborator
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
hkaiser
force-pushed
the
future_wait_until
branch
from
7a913ea to
fa8e4e3
Compare
Collaborator
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
hkaiser
force-pushed
the
future_wait_until
branch
from
fa8e4e3 to
418bcae
Compare
There was a problem hiding this comment.
Actionable comments posted: 3
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
libs/core/synchronization/src/detail/condition_variable.cpp (1)
276-300: 🎯 Functional Correctness | 🟠 Major | ⚡ Quick winReturn
signaledwhen the wait predicate ends the suspension.When
wait_condbecomes true,sleep_untilreportssignaled, butf.ctx_remains set because no notification removed the entry. Line 298 therefore incorrectly converts the predicate wake-up totimeout.Proposed restart-state mapping
- return f.ctx_ || s == threads::thread_restart_state::timeout ? + return s == threads::thread_restart_state::timeout ? threads::thread_restart_state::timeout : threads::thread_restart_state::signaled;🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@libs/core/synchronization/src/detail/condition_variable.cpp` around lines 276 - 300, Update condition_variable::wait_until so a sleep_until result of signaled is returned as signaled even when f.ctx_ remains set. Preserve timeout handling for an actual timeout, and keep notification-driven wake-ups mapped to the existing signaled behavior.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In
`@libs/core/synchronization/include/hpx/synchronization/condition_variable.hpp`:
- Around line 358-368: Update both predicate-based condition-variable wait
paths, including the overload around wait_until and the corresponding path near
the second occurrence, so user pred is evaluated while the caller’s lock remains
held. Do not forward pred through the lower predicate-aware
wait_until/sleep_until path after constructing unlock_guard; instead use the
non-predicate wait and recheck pred under the lock, while preserving the
existing inner-lock ordering and timeout behavior.
- Around line 436-440: Update the timed stop-token wait path around
wait_until_pred to forward the original predicate instead of passing the lvalue
copy. Ensure wait_until_pred and the subsequent wait layer preserve the same
predicate object, supporting move-only predicates and retaining state changes
from stateful predicates.
In `@libs/core/synchronization/src/detail/condition_variable.cpp`:
- Around line 250-274: The condition_variable::wait_until predicate must stop
polling the non-atomic queue_entry field f.ctx_ while the lock is released. Add
or reuse an atomic notification marker in queue_entry, set it before
notify_one/notify_all clear ctx_, and have the sleep_until predicate poll that
marker; update the final timeout/signaled decision to use the marker
consistently.
---
Outside diff comments:
In `@libs/core/synchronization/src/detail/condition_variable.cpp`:
- Around line 276-300: Update condition_variable::wait_until so a sleep_until
result of signaled is returned as signaled even when f.ctx_ remains set.
Preserve timeout handling for an actual timeout, and keep notification-driven
wake-ups mapped to the existing signaled behavior.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 53282baf-6814-4f64-a633-a814d2113840
📒 Files selected for processing (7)
libs/core/execution_base/include/hpx/execution_base/this_thread.hpplibs/core/execution_base/tests/unit/execution_context.cpplibs/core/futures/tests/unit/future.cpplibs/core/synchronization/include/hpx/synchronization/condition_variable.hpplibs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpplibs/core/synchronization/src/detail/condition_variable.cpplibs/core/synchronization/tests/unit/condition_variable.cpp
🚧 Files skipped from review as they are similar to previous changes (5)
- libs/core/execution_base/include/hpx/execution_base/this_thread.hpp
- libs/core/execution_base/tests/unit/execution_context.cpp
- libs/core/futures/tests/unit/future.cpp
- libs/core/synchronization/tests/unit/condition_variable.cpp
- libs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpp
Collaborator
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
hkaiser
force-pushed
the
future_wait_until
branch
from
418bcae to
17397ac
Compare
There was a problem hiding this comment.
🧹 Nitpick comments (2)
libs/core/thread_support/include/hpx/thread_support/unlock_guard.hpp (1)
23-24: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low valueForward declaration's template parameter name diverges from the definition.
The forward declaration uses
Lockablewhile the actual definition at line 69 usesLock. Not a functional issue since parameter names in declarations are independent, but keeping them consistent aids readability.✏️ Suggested consistency fix
- HPX_CXX_CORE_EXPORT template <typename Lockable> + HPX_CXX_CORE_EXPORT template <typename Lock> class relock_guard;Also applies to: 69-70
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@libs/core/thread_support/include/hpx/thread_support/unlock_guard.hpp` around lines 23 - 24, Align the template parameter name in the relock_guard forward declaration with its definition by changing Lockable to Lock, while leaving the class behavior and declaration structure unchanged.libs/core/synchronization/src/detail/condition_variable.cpp (1)
259-286: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low valueOptional: the no-predicate overload largely duplicates the predicate overload.
Both overloads differ only in the
sleep_untilpredicate and are otherwise identical (enqueue,reset_queue_entry, restart-state mapping). The no-predicate version is equivalent to delegating to the predicate overload with an always-false predicate. Consider consolidating to reduce drift between the two paths (weigh against the smallmove_only_functionallocation on this hot path).🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@libs/core/synchronization/src/detail/condition_variable.cpp` around lines 259 - 286, Consolidate the no-predicate condition_variable::wait_until implementation by delegating to the existing predicate overload with an always-false predicate, preserving the enqueue, reset, and restart-state behavior. Confirm this does not introduce an unacceptable move_only_function allocation on this hot path; if it does, retain the separate implementation.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@libs/core/synchronization/src/detail/condition_variable.cpp`:
- Around line 259-286: Consolidate the no-predicate
condition_variable::wait_until implementation by delegating to the existing
predicate overload with an always-false predicate, preserving the enqueue,
reset, and restart-state behavior. Confirm this does not introduce an
unacceptable move_only_function allocation on this hot path; if it does, retain
the separate implementation.
In `@libs/core/thread_support/include/hpx/thread_support/unlock_guard.hpp`:
- Around line 23-24: Align the template parameter name in the relock_guard
forward declaration with its definition by changing Lockable to Lock, while
leaving the class behavior and declaration structure unchanged.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 7586e4a7-d6f9-401d-ba12-16c41dac8d8a
📒 Files selected for processing (8)
libs/core/execution_base/include/hpx/execution_base/this_thread.hpplibs/core/execution_base/tests/unit/execution_context.cpplibs/core/futures/tests/unit/future.cpplibs/core/synchronization/include/hpx/synchronization/condition_variable.hpplibs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpplibs/core/synchronization/src/detail/condition_variable.cpplibs/core/synchronization/tests/unit/condition_variable.cpplibs/core/thread_support/include/hpx/thread_support/unlock_guard.hpp
🚧 Files skipped from review as they are similar to previous changes (6)
- libs/core/execution_base/include/hpx/execution_base/this_thread.hpp
- libs/core/synchronization/include/hpx/synchronization/condition_variable.hpp
- libs/core/synchronization/tests/unit/condition_variable.cpp
- libs/core/futures/tests/unit/future.cpp
- libs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpp
- libs/core/execution_base/tests/unit/execution_context.cpp
Collaborator
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
hkaiser
force-pushed
the
future_wait_until
branch
2 times, most recently
from
d1e4604 to
10c7adb
Compare
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@libs/core/futures/tests/unit/future.cpp`:
- Around line 478-527: In test_wait_for_returns_as_soon_as_value_is_set and
test_wait_until_returns_as_soon_as_value_is_set, tighten the elapsed-time
assertions to use a short timeout and CI-tolerant bound appropriate for the
100-millisecond fulfillment, rather than allowing nearly five seconds. Apply the
same adjustment to the additional readiness timing test covering the indicated
range, while preserving the existing status and value assertions.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 2683699c-79a1-4b6a-b3d4-d2fa3348bdc2
📒 Files selected for processing (8)
libs/core/execution_base/include/hpx/execution_base/this_thread.hpplibs/core/execution_base/tests/unit/execution_context.cpplibs/core/futures/tests/unit/future.cpplibs/core/synchronization/include/hpx/synchronization/condition_variable.hpplibs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpplibs/core/synchronization/src/detail/condition_variable.cpplibs/core/synchronization/tests/unit/condition_variable.cpplibs/core/thread_support/include/hpx/thread_support/unlock_guard.hpp
🚧 Files skipped from review as they are similar to previous changes (7)
- libs/core/execution_base/include/hpx/execution_base/this_thread.hpp
- libs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpp
- libs/core/synchronization/include/hpx/synchronization/condition_variable.hpp
- libs/core/thread_support/include/hpx/thread_support/unlock_guard.hpp
- libs/core/execution_base/tests/unit/execution_context.cpp
- libs/core/synchronization/tests/unit/condition_variable.cpp
- libs/core/synchronization/src/detail/condition_variable.cpp
hkaiser
force-pushed
the
future_wait_until
branch
from
10c7adb to
945f667
Compare
There was a problem hiding this comment.
🧹 Nitpick comments (1)
libs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpp (1)
118-123: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low valueConsider a default-description convenience overload for symmetry.
The plain
wait_untilgets a convenience overload with a default description (Lines 124-131), but this new predicate-taking overload has no equivalent — callers must always supplydescriptionexplicitly. Minor API asymmetry; worth a short overload if predicate-based callers elsewhere in the stack would benefit from a default description.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@libs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpp` around lines 118 - 123, Add a convenience overload alongside the predicate-taking wait_until declaration that omits description and forwards to the existing overload with the established default description, preserving all other parameters and error_code behavior. Keep the current overload unchanged and mirror the nearby plain wait_until overload’s API pattern.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In
`@libs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpp`:
- Around line 118-123: Add a convenience overload alongside the predicate-taking
wait_until declaration that omits description and forwards to the existing
overload with the established default description, preserving all other
parameters and error_code behavior. Keep the current overload unchanged and
mirror the nearby plain wait_until overload’s API pattern.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro Plus
Run ID: bbc9df83-87a0-4ed8-8622-9cf9e98018aa
📒 Files selected for processing (8)
libs/core/execution_base/include/hpx/execution_base/this_thread.hpplibs/core/execution_base/tests/unit/execution_context.cpplibs/core/futures/tests/unit/future.cpplibs/core/synchronization/include/hpx/synchronization/condition_variable.hpplibs/core/synchronization/include/hpx/synchronization/detail/condition_variable.hpplibs/core/synchronization/src/detail/condition_variable.cpplibs/core/synchronization/tests/unit/condition_variable.cpplibs/core/thread_support/include/hpx/thread_support/unlock_guard.hpp
🚧 Files skipped from review as they are similar to previous changes (7)
- libs/core/execution_base/include/hpx/execution_base/this_thread.hpp
- libs/core/execution_base/tests/unit/execution_context.cpp
- libs/core/thread_support/include/hpx/thread_support/unlock_guard.hpp
- libs/core/synchronization/include/hpx/synchronization/condition_variable.hpp
- libs/core/synchronization/src/detail/condition_variable.cpp
- libs/core/futures/tests/unit/future.cpp
- libs/core/synchronization/tests/unit/condition_variable.cpp
hkaiser
force-pushed
the
future_wait_until
branch
from
945f667 to
de2cf58
Compare
Implemented from a Change Stack AI coding task. Co-authored-by: CodeRabbit <noreply@coderabbit.ai> Signed-off-by: Hartmut Kaiser <hartmut.kaiser@gmail.com>
hkaiser
force-pushed
the
future_wait_until
branch
from
de2cf58 to
d478f99
Compare
Collaborator
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Future waits now stop promptly once the associated shared state transitions to ready.