Main merge release/25.12 5 #715
Conversation
## Summary by CodeRabbit ## Release Notes * **Performance** * Optimized branch-and-bound algorithm with improved search termination conditions * **Improvements** * Enhanced concurrency control mechanisms across solver components * Improved logger initialization and lifecycle management for better resource handling <sub>✏️ Tip: You can customize this high-level summary in your review settings.</sub> Authors: - Alice Boucher (https://github.com/aliceb-nv) Approvers: - Nicolas Blin (https://github.com/Kh4ster) URL: NVIDIA#691
rgsl888prabhu
requested review from
KyleFromNVIDIA,
chris-maes and
nguidotti
rgsl888prabhu
added
non-breaking
📝 WalkthroughWalkthroughThis pull request makes concurrency improvements by replacing volatile semantics with atomic operations across multiple solver components, refactors logger initialization with guard-based lifecycle management, adds a pre-commit hook for alpha spec verification, introduces an early-exit optimization in the branch-and-bound algorithm, and simplifies rapids-logger dependency constraints. Changes
Estimated code review effort🎯 4 (Complex) | ⏱️ ~50 minutes
Pre-merge checks and finishing touches❌ Failed checks (1 warning, 1 inconclusive)
✅ Passed checks (1 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (2)
cpp/src/dual_simplex/branch_and_bound.cpp (1)
1156-1157: Consider making the node limit configurable.The hardcoded limit of 1000 nodes per diving subtree is a reasonable heuristic for early exit, but consider exposing this as a configurable parameter (e.g.,
settings_.diving_node_limit) for performance tuning on different problem types.cpp/src/utilities/logger.cpp (1)
149-158: Consider logging when reusing existing configuration with different parameters.When an
init_logger_tis constructed while another is already active, it silently reuses the existing configuration even iflog_fileorlog_to_consolediffer. This may be intentional (first-wins semantics), but could cause confusion during debugging.Consider adding a debug log or assertion when the requested configuration differs from the active one:
auto existing_guard = g_active_guard.lock(); if (existing_guard) { // Reuse existing configuration, just hold a reference to keep it alive + // Note: log_file and log_to_console parameters are ignored when reusing guard_ = existing_guard; return; }
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (14)
.pre-commit-config.yaml(2 hunks)cpp/include/cuopt/linear_programming/pdlp/solver_settings.hpp(1 hunks)cpp/src/dual_simplex/branch_and_bound.cpp(3 hunks)cpp/src/dual_simplex/branch_and_bound.hpp(2 hunks)cpp/src/dual_simplex/simplex_solver_settings.hpp(1 hunks)cpp/src/linear_programming/solve.cu(1 hunks)cpp/src/mip/diversity/diversity_manager.cuh(1 hunks)cpp/src/mip/relaxed_lp/relaxed_lp.cuh(1 hunks)cpp/src/utilities/logger.cpp(2 hunks)cpp/src/utilities/logger.hpp(1 hunks)dependencies.yaml(1 hunks)python/cuopt/cuopt/linear_programming/pyproject.toml(3 hunks)python/cuopt/pyproject.toml(1 hunks)python/libcuopt/pyproject.toml(2 hunks)
🧰 Additional context used
📓 Path-based instructions (7)
**/*.{cu,cuh,cpp,hpp,h}
📄 CodeRabbit inference engine (.github/.coderabbit_review_guide.md)
**/*.{cu,cuh,cpp,hpp,h}: Track GPU device memory allocations and deallocations to prevent memory leaks; ensure cudaMalloc/cudaFree balance and cleanup of streams/events
Validate algorithm correctness in optimization logic: simplex pivots, branch-and-bound decisions, routing heuristics, and constraint/objective handling must produce correct results
Check numerical stability: prevent overflow/underflow, precision loss, division by zero/near-zero, and use epsilon comparisons for floating-point equality checks
Validate correct initialization of variable bounds, constraint coefficients, and algorithm state before solving; ensure reset when transitioning between algorithm phases (presolve, simplex, diving, crossover)
Ensure variables and constraints are accessed from the correct problem context (original vs presolve vs folded vs postsolve); verify index mapping consistency across problem transformations
For concurrent CUDA operations (barriers, async operations), explicitly create and manage dedicated streams instead of reusing the default stream; document stream lifecycle
Eliminate unnecessary host-device synchronization (cudaDeviceSynchronize) in hot paths that blocks GPU pipeline; use streams and events for async execution
Assess algorithmic complexity for large-scale problems (millions of variables/constraints); ensure O(n log n) or better complexity, not O(n²) or worse
Verify correct problem size checks before expensive GPU/CPU operations; prevent resource exhaustion on oversized problems
Identify assertions with overly strict numerical tolerances that fail on legitimate degenerate/edge cases (near-zero pivots, singular matrices, empty problems)
Ensure race conditions are absent in multi-GPU code and multi-threaded server implementations; verify proper synchronization of shared state
Refactor code duplication in solver components (3+ occurrences) into shared utilities; for GPU kernels, use templated device functions to avoid duplication
Check that hard-coded GPU de...
Files:
cpp/src/dual_simplex/simplex_solver_settings.hppcpp/include/cuopt/linear_programming/pdlp/solver_settings.hppcpp/src/linear_programming/solve.cucpp/src/mip/diversity/diversity_manager.cuhcpp/src/utilities/logger.hppcpp/src/dual_simplex/branch_and_bound.cppcpp/src/mip/relaxed_lp/relaxed_lp.cuhcpp/src/dual_simplex/branch_and_bound.hppcpp/src/utilities/logger.cpp
**/*.{h,hpp,py}
📄 CodeRabbit inference engine (.github/.coderabbit_review_guide.md)
Verify C API does not break ABI stability (no struct layout changes, field reordering); maintain backward compatibility in Python and server APIs with deprecation warnings
Files:
cpp/src/dual_simplex/simplex_solver_settings.hppcpp/include/cuopt/linear_programming/pdlp/solver_settings.hppcpp/src/utilities/logger.hppcpp/src/dual_simplex/branch_and_bound.hpp
**/*.{cpp,hpp,h}
📄 CodeRabbit inference engine (.github/.coderabbit_review_guide.md)
**/*.{cpp,hpp,h}: Check for unclosed file handles when reading MPS/QPS problem files; ensure RAII patterns or proper cleanup in exception paths
Validate input sanitization to prevent buffer overflows and resource exhaustion attacks; avoid unsafe deserialization of problem files
Prevent thread-unsafe use of global and static variables; use proper mutex/synchronization in server code accessing shared solver state
Files:
cpp/src/dual_simplex/simplex_solver_settings.hppcpp/include/cuopt/linear_programming/pdlp/solver_settings.hppcpp/src/utilities/logger.hppcpp/src/dual_simplex/branch_and_bound.cppcpp/src/dual_simplex/branch_and_bound.hppcpp/src/utilities/logger.cpp
**/*.{cu,cpp,hpp,h}
📄 CodeRabbit inference engine (.github/.coderabbit_review_guide.md)
Avoid inappropriate use of exceptions in performance-critical GPU operation paths; prefer error codes or CUDA error checking for latency-sensitive code
Files:
cpp/src/dual_simplex/simplex_solver_settings.hppcpp/include/cuopt/linear_programming/pdlp/solver_settings.hppcpp/src/linear_programming/solve.cucpp/src/utilities/logger.hppcpp/src/dual_simplex/branch_and_bound.cppcpp/src/dual_simplex/branch_and_bound.hppcpp/src/utilities/logger.cpp
cpp/include/cuopt/**/*
⚙️ CodeRabbit configuration file
cpp/include/cuopt/**/*: For public header files (C++ API):
- Check if new public functions/classes have documentation comments (Doxygen format)
- Flag API changes that may need corresponding docs/ updates
- Verify parameter descriptions match actual types/behavior
- Suggest documenting thread-safety, GPU requirements, and numerical behavior
- For breaking changes, recommend updating docs and migration guides
Files:
cpp/include/cuopt/linear_programming/pdlp/solver_settings.hpp
**/*.{cu,cuh}
📄 CodeRabbit inference engine (.github/.coderabbit_review_guide.md)
**/*.{cu,cuh}: Every CUDA kernel launch and memory operation must have error checking with CUDA_CHECK or equivalent verification
Avoid reinventing functionality already available in Thrust, CCCL, or RMM libraries; prefer standard library utilities over custom implementations
Files:
cpp/src/linear_programming/solve.cucpp/src/mip/diversity/diversity_manager.cuhcpp/src/mip/relaxed_lp/relaxed_lp.cuh
**/*.cu
📄 CodeRabbit inference engine (.github/.coderabbit_review_guide.md)
**/*.cu: Verify race conditions and correctness of GPU kernel shared memory, atomics, and warp-level operations
Detect inefficient GPU kernel launches with low occupancy or poor memory access patterns; optimize for coalesced memory access and minimize warp divergence in hot paths
Files:
cpp/src/linear_programming/solve.cu
🧠 Learnings (13)
📓 Common learnings
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Reduce tight coupling between solver components (presolve, simplex, basis, barrier); increase modularity and reusability of optimization algorithms
📚 Learning: 2025-11-25T10:20:49.822Z
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Applies to **/*.{cpp,hpp,h} : Prevent thread-unsafe use of global and static variables; use proper mutex/synchronization in server code accessing shared solver state
Applied to files:
cpp/src/dual_simplex/simplex_solver_settings.hppcpp/include/cuopt/linear_programming/pdlp/solver_settings.hppcpp/src/linear_programming/solve.cucpp/src/mip/diversity/diversity_manager.cuhcpp/src/dual_simplex/branch_and_bound.hpp
📚 Learning: 2025-12-04T20:09:09.264Z
Learnt from: chris-maes
Repo: NVIDIA/cuopt PR: 602
File: cpp/src/linear_programming/solve.cu:732-742
Timestamp: 2025-12-04T20:09:09.264Z
Learning: In cpp/src/linear_programming/solve.cu, the barrier solver does not currently return INFEASIBLE or UNBOUNDED status. It only returns OPTIMAL, TIME_LIMIT, NUMERICAL_ISSUES, or CONCURRENT_LIMIT.
Applied to files:
cpp/src/dual_simplex/simplex_solver_settings.hppcpp/src/linear_programming/solve.cucpp/src/mip/relaxed_lp/relaxed_lp.cuhcpp/src/dual_simplex/branch_and_bound.hpp
📚 Learning: 2025-11-25T10:20:49.822Z
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Reduce tight coupling between solver components (presolve, simplex, basis, barrier); increase modularity and reusability of optimization algorithms
Applied to files:
cpp/src/dual_simplex/simplex_solver_settings.hppcpp/src/dual_simplex/branch_and_bound.cpp
📚 Learning: 2025-11-25T10:20:49.822Z
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Applies to **/*.{cu,cuh,cpp,hpp,h} : Validate algorithm correctness in optimization logic: simplex pivots, branch-and-bound decisions, routing heuristics, and constraint/objective handling must produce correct results
Applied to files:
cpp/src/dual_simplex/simplex_solver_settings.hppcpp/src/linear_programming/solve.cucpp/src/dual_simplex/branch_and_bound.cppcpp/src/mip/relaxed_lp/relaxed_lp.cuhcpp/src/dual_simplex/branch_and_bound.hpp
📚 Learning: 2025-11-25T10:20:49.822Z
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Applies to **/*.{cu,cuh,cpp,hpp,h} : Validate correct initialization of variable bounds, constraint coefficients, and algorithm state before solving; ensure reset when transitioning between algorithm phases (presolve, simplex, diving, crossover)
Applied to files:
cpp/src/dual_simplex/simplex_solver_settings.hppcpp/src/linear_programming/solve.cucpp/src/dual_simplex/branch_and_bound.cppcpp/src/mip/relaxed_lp/relaxed_lp.cuhcpp/src/dual_simplex/branch_and_bound.hpp
📚 Learning: 2025-11-25T10:20:49.822Z
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Applies to **/*.{cu,cuh,cpp,hpp,h} : Refactor code duplication in solver components (3+ occurrences) into shared utilities; for GPU kernels, use templated device functions to avoid duplication
Applied to files:
cpp/src/dual_simplex/simplex_solver_settings.hppcpp/src/linear_programming/solve.cu
📚 Learning: 2025-11-25T10:20:49.822Z
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Applies to **/*.{cu,cuh,cpp,hpp,h} : Ensure race conditions are absent in multi-GPU code and multi-threaded server implementations; verify proper synchronization of shared state
Applied to files:
cpp/include/cuopt/linear_programming/pdlp/solver_settings.hppcpp/src/linear_programming/solve.cucpp/src/mip/diversity/diversity_manager.cuh
📚 Learning: 2025-11-25T10:20:49.822Z
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Applies to **/*.{cu,cuh,cpp,hpp,h} : Ensure variables and constraints are accessed from the correct problem context (original vs presolve vs folded vs postsolve); verify index mapping consistency across problem transformations
Applied to files:
cpp/include/cuopt/linear_programming/pdlp/solver_settings.hppcpp/src/linear_programming/solve.cucpp/src/dual_simplex/branch_and_bound.cppcpp/src/mip/relaxed_lp/relaxed_lp.cuhcpp/src/dual_simplex/branch_and_bound.hpp
📚 Learning: 2025-11-25T10:20:49.822Z
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Applies to **/*.cu : Verify race conditions and correctness of GPU kernel shared memory, atomics, and warp-level operations
Applied to files:
cpp/src/linear_programming/solve.cu
📚 Learning: 2025-11-25T10:20:49.822Z
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Applies to **/*.{cu,cuh,cpp,hpp,h} : Check that hard-coded GPU device IDs and resource limits are made configurable; abstract multi-backend support for different CUDA versions
Applied to files:
cpp/src/linear_programming/solve.cu
📚 Learning: 2025-11-25T10:20:49.822Z
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Applies to **/*.{cu,cuh,cpp,hpp,h} : Verify correct problem size checks before expensive GPU/CPU operations; prevent resource exhaustion on oversized problems
Applied to files:
cpp/src/linear_programming/solve.cu
📚 Learning: 2025-11-25T10:20:49.822Z
Learnt from: CR
Repo: NVIDIA/cuopt PR: 0
File: .github/.coderabbit_review_guide.md:0-0
Timestamp: 2025-11-25T10:20:49.822Z
Learning: Applies to **/*test*.{cpp,cu,py} : Add tests for algorithm phase transitions: verify correct initialization of bounds and state when transitioning from presolve to simplex to diving to crossover
Applied to files:
cpp/src/dual_simplex/branch_and_bound.cpp
🧬 Code graph analysis (1)
cpp/src/utilities/logger.cpp (1)
cpp/src/utilities/logger.hpp (2)
reset_default_logger(34-34)init_logger_t(42-42)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (9)
- GitHub Check: wheel-build-cuopt-mps-parser / 13.0.2, 3.11, arm64, rockylinux8
- GitHub Check: wheel-build-cuopt-mps-parser / 13.0.2, 3.13, arm64, rockylinux8
- GitHub Check: wheel-build-cuopt-mps-parser / 13.0.2, 3.11, amd64, rockylinux8
- GitHub Check: wheel-build-cuopt-mps-parser / 13.0.2, 3.13, amd64, rockylinux8
- GitHub Check: wheel-build-cuopt-mps-parser / 13.0.2, 3.12, amd64, rockylinux8
- GitHub Check: wheel-build-cuopt-mps-parser / 13.0.2, 3.12, arm64, rockylinux8
- GitHub Check: wheel-build-cuopt-mps-parser / 13.0.2, 3.10, arm64, rockylinux8
- GitHub Check: wheel-build-cuopt-mps-parser / 13.0.2, 3.10, amd64, rockylinux8
- GitHub Check: checks / check-style
🔇 Additional comments (20)
python/cuopt/pyproject.toml (1)
51-51: LGTM - Simplified version constraint.Removing the redundant
>=0.0.0a0prerelease lower-bound is appropriate since==0.2.*already constrains to the 0.2.x series.cpp/src/dual_simplex/branch_and_bound.cpp (2)
1138-1138: LGTM - Local counter for early-exit optimization.The
nodes_exploredcounter is correctly scoped to each diving subtree iteration, resetting when a newstart_nodeis processed.
1171-1172: LGTM - Correct increment placement.The counter is incremented after
solve_nodecompletes, accurately tracking the number of nodes actually explored.cpp/src/dual_simplex/simplex_solver_settings.hpp (1)
148-149: LGTM - Correct migration from volatile to std::atomic.Replacing
volatile int*withstd::atomic<int>*provides proper memory ordering guarantees for inter-thread communication. Thevolatilekeyword does not provide atomicity and is unsuitable for synchronization between threads.dependencies.yaml (1)
563-563: LGTM - Simplified version constraint.Removing the redundant
>=0.0.0a0prerelease lower-bound fromrapids-loggeris appropriate. The==0.2.*constraint already pins to the 0.2.x series effectively.cpp/src/linear_programming/solve.cu (1)
309-309: LGTM - Correct atomic initialization for concurrent termination signaling.The global
concurrent_haltis correctly changed tostd::atomic<int>with explicit zero-initialization. All accesses throughout the codebase—reads via*settings.concurrent_halt == 1and writes via*settings.concurrent_halt = 1;—properly invoke atomic operators to maintain sequentially consistent memory ordering. This replaces the unsafevolatilequalifier with proper synchronization semantics for the concurrent termination signal shared between PDLP and Dual Simplex/Barrier/Crossover threads.cpp/src/dual_simplex/branch_and_bound.hpp (2)
116-117: LGTM - Correct atomic usage for concurrent halt signaling.The migration from
volatile inttostd::atomic<int>is the right approach. The direct assignment on line 117 uses the defaultstd::memory_order_seq_cst, which provides strong ordering guarantees appropriate for halt signaling across threads.
173-173: Consistent with the codebase-wide atomic migration.The member variable correctly uses value initialization with
{0}, which is the proper way to initializestd::atomic<int>.cpp/src/mip/diversity/diversity_manager.cuh (1)
97-98: LGTM - Proper atomic signaling for PDLP concurrency control.The change from
volatile inttostd::atomic<int>correctly provides the necessary synchronization guarantees for signaling the PDLP solver to stop. The initialization with{0}is appropriate.python/libcuopt/pyproject.toml (2)
43-44: Dependency spec simplified for release mode.Removing the
>=0.0.0a0constraint simplifies the version spec and restricts to stable 0.2.x releases. This aligns with the pre-commit hook addition forverify-alpha-specin release mode.
88-89: Build dependency spec consistent with runtime dependency.The same simplification applied to build requirements ensures consistency.
.pre-commit-config.yaml (2)
59-63: New pre-commit hook for alpha spec verification.The
verify-alpha-spechook with--mode releasewill enforce that dependencies don't include alpha/pre-release version constraints in release builds. The--fixflag enables automatic remediation.
91-92: YAML formatting correction.The
default_language_versionis now properly formatted as a mapping rather than a list item.python/cuopt/cuopt/linear_programming/pyproject.toml (3)
21-24: Consistent dependency spec simplification.The rapids-logger constraint is simplified across all dependency sections in this file, aligning with the release mode configuration enforced by the new
verify-alpha-specpre-commit hook.
42-43: Test dependencies aligned.
86-87: Build requirements aligned.cpp/src/mip/relaxed_lp/relaxed_lp.cuh (1)
19-28: LGTM - Atomic pointer type for concurrent halt control.The change from
volatile int*tostd::atomic<int>*correctly aligns with the codebase-wide migration to atomic semantics. The pointer beingnullptrby default is appropriate since not all callers may need concurrent halt capability.cpp/src/utilities/logger.hpp (1)
36-43: Ref-counted logger initialization pattern.The use of
std::shared_ptr<void>provides type-erased ref-counting for the logger configuration guard. This enables multipleinit_logger_tinstances to share the same logger configuration, with automatic reset when the last instance is destroyed.cpp/src/utilities/logger.cpp (2)
140-147: Guard-based lifecycle management for logger configuration.The
logger_config_guardwith weak reference tracking enables proper ref-counting. The mutex ensures thread-safe access to the global guard state.
185-189: LGTM - Proper guard creation and weak reference storage.The guard is correctly created as a
shared_ptr, stored in both the weak reference (for future lookups) and the member variable (for ownership). When allinit_logger_tinstances are destroyed, the guard's destructor will reset the logger.
| bool inside_mip{false}; | ||
| // For concurrent termination | ||
| volatile int* concurrent_halt{nullptr}; | ||
| std::atomic<int>* concurrent_halt{nullptr}; |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
# Search for all usages of concurrent_halt in the repository
echo "=== Searching for concurrent_halt usages ==="
rg -n 'concurrent_halt' --type=cpp --type=h --type=hpp 2>/dev/null | head -50
echo ""
echo "=== Check context in solver_settings.hpp ==="
rg -B 5 -A 5 'concurrent_halt' cpp/include/cuopt/linear_programming/pdlp/solver_settings.hpp
echo ""
echo "=== Check if there are other volatile declarations ==="
rg 'volatile.*\*' cpp/include/cuopt/linear_programming/pdlp/solver_settings.hppRepository: NVIDIA/cuopt
Length of output: 570
🏁 Script executed:
#!/bin/bash
# Check the struct/class definition context
echo "=== Full struct context around concurrent_halt ==="
cat -n cpp/include/cuopt/linear_programming/pdlp/solver_settings.hpp | sed -n '200,230p'Repository: NVIDIA/cuopt
Length of output: 1312
🏁 Script executed:
#!/bin/bash
# Check for documentation/comments about this field
echo "=== Check for doxygen or comments about concurrent_halt ==="
rg -B 10 'concurrent_halt' cpp/include/cuopt/linear_programming/pdlp/solver_settings.hpp | grep -E '(///|/\*|--)'Repository: NVIDIA/cuopt
Length of output: 115
🏁 Script executed:
#!/bin/bash
# Search for actual usages in implementation files
echo "=== Check implementation usages ==="
rg 'concurrent_halt' cpp/src --type=cpp --type=h --type=hpp 2>/dev/null | head -20Repository: NVIDIA/cuopt
Length of output: 91
Add Doxygen documentation for public field concurrent_halt.
The change from volatile int* to std::atomic<int>* is correct for proper inter-thread synchronization. However, as a public API field in this settings struct, it requires documentation per the coding guidelines for public header files. Add a Doxygen comment explaining:
- Purpose: coordination for concurrent solver termination
- Expected usage: users provide a pointer to std::atomic for concurrent halt signaling, or leave as nullptr for single-threaded mode
- Thread-safety guarantees
No existing usages found in the codebase, so the API change poses minimal risk of breaking downstream users.
🤖 Prompt for AI Agents
In cpp/include/cuopt/linear_programming/pdlp/solver_settings.hpp around line
215, the public field std::atomic<int>* concurrent_halt lacks Doxygen
documentation; add a Doxygen comment immediately above the declaration that
states its purpose (coordination for concurrent solver termination), expected
usage (users may provide a pointer to a std::atomic<int> to signal halt across
threads or leave nullptr for single-threaded mode), and the thread-safety
guarantees (safe for concurrent reads/writes via std::atomic, pointer itself
must remain valid for the solver lifetime).
3 participants
For #693
Summary by CodeRabbit
New Features
Bug Fixes
Chores
✏️ Tip: You can customize this high-level summary in your review settings.