Uh oh!
There was an error while loading. Please reload this page.
Add debug logging for threading and parallelization dec… - #3
Draft
eginez wants to merge 3 commits into
Draft
Conversation
…isions Adds debug logging across Rust and Python components to help diagnose threading behavior and parallelization decisions. Includes logging for: - Rust threading decisions in getenv_use_multiple_threads() - PassManager parallel execution choices and process counts - SABRE layout trial count configuration - parallel_map execution paths and results Co-Authored with the help of: Claude
There was a problem hiding this comment.
Pull Request Overview
Adds debug logging across Python and Rust components to help diagnose threading and parallelization decisions.
- Python: Injected
logger.debugstatements inshould_run_in_parallel,parallel_map, SABRE trial count, and PassManager execution paths. - Rust: Emits threading decision to stderr when
QISKIT_DEBUG_THREADINGis set.
Reviewed Changes
Copilot reviewed 4 out of 4 changed files in this pull request and generated 4 comments.
| File | Description |
|---|---|
| qiskit/utils/parallel.py | Added logging import, logger setup, and debug statements around parallel decisions and execution. |
| qiskit/transpiler/preset_passmanagers/builtin_plugins.py | Added debug logging in _get_trial_count for SABRE threading trials. |
| qiskit/passmanager/passmanager.py | Added debug logs in callback_func for serial vs parallel PassManager runs. |
| crates/accelerate/src/lib.rs | Added Rust debug printing under QISKIT_DEBUG_THREADING environment variable. |
Comments suppressed due to low confidence (1)
qiskit/utils/parallel.py:197
- The environment variable QISKIT_IN_PARALLEL is only reset in the KeyboardInterrupt branch, so other exceptions leave it set to TRUE. Consider moving the reset into a finally block to ensure it's always restored.
os.environ["QISKIT_IN_PARALLEL"] = "FALSE"
Uh oh!
There was an error while loading. Please reload this page.
| del callback | ||
| del kwargs | ||
| logger.debug("PassManager running %d programs in parallel with %d processes", |
There was a problem hiding this comment.
The serial execution branch assigns to out but lacks a return, so execution always falls through to the parallel path. Add a return statement inside the if-block to avoid unintended parallel runs.
Uh oh!
There was an error while loading. Please reload this page.
| // Log threading decision if debug logging is enabled | ||
| if env::var("QISKIT_DEBUG_THREADING").is_ok() { | ||
| eprintln!( |
There was a problem hiding this comment.
[nitpick] Using eprintln! for debug output is inconsistent with typical Rust logging practices. Consider using the log crate to emit debug-level messages and integrate with existing logging infrastructure.
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
chookity-pokk pushed a commit
that referenced
this pull request
Apr 29, 2026
Qiskit#15943) * fix(transpiler): TemplateOptimization drops circuit global_phase on substitution circuit_to_dagdependency did not copy global_phase when converting a template QuantumCircuit to DAGDependency, so template phases were lost before matching. TemplateSubstitution.run_dag_opt also constructed a new DAGDependency for the optimised output without inheriting the original circuit's global_phase, resetting it to zero whenever at least one substitution was applied. Fixed by copying global_phase in circuit_to_dagdependency and seeding dag_dep_opt.global_phase from the circuit DAG at construction time. Also added the per-match phase subtraction (Fix 3) for when templates with nonzero global_phase become accepted via the identity check fix in Qiskit#14538. FixesQiskit#14537. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(transpiler): TemplateOptimization drops circuit global_phase on substitution circuit_to_dagdependency did not copy global_phase when converting a template QuantumCircuit to DAGDependency, so template phases were lost before matching. TemplateSubstitution.run_dag_opt also constructed a new DAGDependency for the optimised output without inheriting the original circuit's global_phase, resetting it to zero whenever at least one substitution was applied. Additionally, when a template carries a nonzero global_phase phi_T (gate content implements e^{-i*phi_T} * I while the full operator is I), the per-match phase contribution was not being subtracted from the output circuit. FixesQiskit#14537. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * style: black reformat test_template_matching.py Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * docs: use RST hyperlink for issue reference in release note The :issue: role is not available in this project's Sphinx config. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * test(transpiler): add combined circuit+template global_phase regression test Covers the case where both the circuit and the template carry a nonzero global_phase simultaneously, confirming that fixes#2 (circuit phase preserved across substitution) and #3 (per-match template phase subtracted) compose correctly. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * test(transpiler): use assertEqual for Operator comparisons in template tests Replace assertTrue(Operator(x) == Operator(y)) with assertEqual(Operator(x), Operator(y)) so that assertion failures show the actual vs expected values rather than just "False is not True". Also shorten multi-line test docstrings to concise one-liners. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * Address review feedback on template optimization global phase fix - Simplify comment on global phase update in template_substitution.py - Split release note into per-bug files; add fix-circuit-to-dagdependency-global-phase-14537.yaml - Update tests: use issue reproducer (4-gate circuit) in test_template_nonzero_global_phase_applied_to_circuit, merge single/multiple match tests, drop modulo from assertAlmostEqual Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * Fix test assertions and remove duplicate test - Fix test_circuit_and_template_both_have_nonzero_global_phase expected value from np.pi/4 to 7*np.pi/12 (pi/3 + pi/4) - Remove duplicate test_circuit_global_phase_preserved_with_multiple_template_matches - Update test_template_nonzero_global_phase_applied_to_circuit to use Operator equivalence instead of phase/count assertions since partial template match does occur Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * Update releasenotes/notes/fix-circuit-to-dagdependency-global-phase-14537.yaml Co-authored-by: Julien Gacon <gaconju@gmail.com> * Update releasenotes/notes/fix-template-optimization-global-phase-14537.yaml Co-authored-by: Julien Gacon <gaconju@gmail.com> * Update test/python/transpiler/test_template_matching.py Co-authored-by: Julien Gacon <gaconju@gmail.com> * Update test/python/transpiler/test_template_matching.py Actually I removed all reference to "Regression test" in the test descriptions based on this comment Qiskit#15943 (comment) Co-authored-by: Julien Gacon <gaconju@gmail.com> * Remove 'Regression test' from docstring per review feedback Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com> Co-authored-by: Julien Gacon <gaconju@gmail.com>
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 freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
…isions
Adds debug logging across Rust and Python components to help diagnose threading behavior and parallelization decisions. Includes logging for:
Co-Authored with the help of: Claude
Summary
Details and comments