Several bug fixes

[nguidotti]

• Fixed early termination in CMS750_4 due to an incorrect root objective from the PDLP. It provides an objective in user space and B&B expects the objective in the solver space.
• Fixed setting the number of threads to a fixed value when running cuopt_cli. Now the number of threads in the probing cache can be controlled by the bounds_update::settings_t.
• Inverted loop order in probing cache to avoid creating and deleting the OpenMP thread pool each iteration.
• PDLP/Barrier no longer set the root objective and solution when the B&B is already running.
• Improved the logs when solving the root relaxation using the concurrent mode.

Checklist

• I am familiar with the Contributing Guidelines.
• Testing
  • New or existing tests cover these changes
  • Added tests
  • Created an issue to follow-up
  • NA
• Documentation
  • The documentation is up to date with these changes
  • Added new documentation
  • NA

Summary by CodeRabbit

• New Features

  • Per-instance threading for presolve with sensible defaults and clamping.
  • Bidirectional objective conversion so callbacks receive both solver-space and user-space objective values.

• Bug Fixes

  • Root relaxation solution handling made idempotent to avoid overwrites.
  • Reduced verbose logging inside MIP solves and refined root/crossover status messages.

• Style

  • Updated header/license years and minor logging formatting improvements.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

📝 Walkthrough

Walkthrough

Reworks objective handling to pass both user-space and solver-space objectives, adds bidirectional objective conversion, makes root-solution setting idempotent, introduces configurable presolve threading, and adds conditional logging to suppress verbose LP messages during MIP solves.

Changes

Cohort / File(s)	Summary
Objective conversion & B&B callback cpp/src/mip/diversity/diversity_manager.cu, cpp/src/mip/problem/problem.cuh, cpp/src/mip/problem/problem.cu	Add get_solver_obj_from_user_obj(f_t) and use it to derive a solver-space objective from the PDLP/user objective. Update callback invocations to pass solver_obj then user_obj. Adjust related comments and data flow.
Root solution idempotency & logging flow cpp/src/dual_simplex/branch_and_bound.hpp, cpp/src/dual_simplex/branch_and_bound.cpp	Add is_root_solution_set flag and guard assignments in set_root_relaxation_solution/root handling to avoid overwriting cached root crossover data; propagate solver name, iterations and objective from crossover when present; minor logging/formatting adjustments.
Presolve threading & probing cache cpp/src/mip/presolve/bounds_presolve.cuh, cpp/src/mip/presolve/probing_cache.cu	Add settings.num_threads (default -1), include OpenMP header, compute/clamp num_threads (fallback 0.2 * omp_get_max_threads()), resize per-thread pools to num_threads, and run parallel regions with num_threads.
Conditional LP logging & macros cpp/src/linear_programming/solve.cu, cpp/src/dual_simplex/phase2.cpp	Introduce CUOPT_LOG_CONDITIONAL_INFO(condition, ...) and replace direct INFO logs with conditional macro to suppress logs when inside MIP; remove redundant root-relaxation logging in phase2; update copyright year.
LP status helper cpp/src/dual_simplex/solve.hpp	Add internal helper lp_status_to_string(lp_status_t) mapping enum values to strings.
Build/manifest files manifest_file, CMakeLists.txt	Minor manifest/CMake adjustments listed in diff (lines changed small).

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~50 minutes

🚥 Pre-merge checks | ✅ 1 | ❌ 2

❌ Failed checks (1 warning, 1 inconclusive)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 9.09% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.
Title check	❓ Inconclusive	The title 'Several bug fixes' is vague and generic, using non-descriptive language that does not convey meaningful information about the specific changes in the changeset.	Replace with a more specific title that highlights the primary change, such as 'Fix root objective handling in Branch & Bound PDLP integration' or 'Fix PDLP objective space mapping in root relaxation'.

✅ Passed checks (1 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

---

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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 0

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

cpp/src/mip/presolve/probing_cache.cu (1)

859-910: Add defensive validation for num_threads before pool sizing and parallel launch

The code assumes settings.num_threads is always positive, but since settings_t is a public struct, external code can set num_threads to 0 (or modify it between initialization and use). If 0 is assigned:

• modification_vector_pool and substitution_vector_pool would have size 0
• #pragma omp parallel num_threads(0) produces undefined behavior (OpenMP spec)
• If threads spawn despite 0, thread_idx from omp_get_thread_num() will exceed pool bounds, causing out-of-bounds access

Add a defensive clamp before sizing the pools:

Proposed fix

-  const size_t num_threads = bound_presolve.settings.num_threads;
+  const i_t requested_threads = bound_presolve.settings.num_threads;
+  size_t num_threads =
+    requested_threads > 0 ? static_cast<size_t>(requested_threads)
+                          : static_cast<size_t>(omp_get_max_threads());
+  if (num_threads == 0) { num_threads = 1; }

Aligns with multithreading safety guidelines to prevent resource exhaustion and thread safety violations.

[akifcorduk]

Can't we undef the previous macro and still use CUOPT_LOG_INFO here?

[nguidotti]

I think it is better to add another macro, since there are some methods where we do not pass the setting as argument, hence they need to use the old version.

[nguidotti]

Ideally, we should add logger_t to the settings rather than CUOPT_LOG_INFO or rapids::logger_t.

[chris-maes]

This is where we should indicate that PDLP/Barrier won. Can you use the is_pdlp_solution to determine which of them won and print it here?

[nguidotti]

I did not find a way to get the winner solver from the optimization_problem_solution_t unless we change the API (#787). This object is user-facing, so it needs to be reflect on the Python side as well.

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@cpp/src/dual_simplex/branch_and_bound.hpp`:
- Around line 87-97: The plain bool is_root_solution_set is accessed
concurrently (read in set_root_relaxation_solution and written in
solve_root_relaxation), causing a data race; change the declaration of
is_root_solution_set to std::atomic<bool> and update all reads/writes in
functions such as set_root_relaxation_solution and solve_root_relaxation to use
.load()/.store() with appropriate memory orders (e.g., memory_order_acquire for
reads and memory_order_release for writes) to match the existing
root_crossover_solution_set_ usage, ensuring thread-safe checks and updates.

[coderabbitai]

Actionable comments posted: 2

🤖 Fix all issues with AI agents

In `@cpp/src/dual_simplex/branch_and_bound.cpp`:
- Around line 1372-1386: The plain bool is_root_solution_set is written in
branch_and_bound.cpp (at the shown block) and concurrently read by
set_root_relaxation_solution(), causing a data race; change its declaration to
std::atomic<bool> (matching root_crossover_solution_set_) or guard all accesses
with mutex_upper_ to synchronize with upper-bound updates; update all
reads/writes of is_root_solution_set to use atomic operations (or lock
mutex_upper_) and ensure consistency with existing root_crossover_solution_set_
usage patterns.
- Around line 1351-1358: The composite iteration count from the
barrier+crossover solution (root_crossover_soln_) is being copied into
root_relax_soln_.iterations and later used via exploration_stats_.total_lp_iters
to compute diving limits, which is incorrect; update the swap logic around
root_relax_soln_, root_crossover_soln_, root_vstatus_, root_status,
user_objective and solver_name so that you do not propagate the composite
barrier+crossover iteration total into the dual-simplex/diving accounting—either
reset iter to 0 (or to a separate dual-simplex pivot counter) immediately after
assigning root_relax_soln_ = root_crossover_soln_ or preserve and copy a stored
dual-simplex iteration field instead; ensure exploration_stats_.total_lp_iters
and lp_settings.iteration_limit continue to use the dual-simplex pivot count (or
a fresh counter) rather than the barrier+crossover composite value so diving
workers aren’t terminated prematurely.

[nguidotti]

/merge