Skip to content

Sparse matrices

Sparse matrices #3687

Triggered via pull request January 15, 2026 22:48
Status Failure
Total duration 8m 35s
Artifacts

ci.yml

on: pull_request
Matrix: downstream_tests
Fit to window
Zoom out
Zoom in

Annotations

31 errors and 52 warnings
Conda Examples
Process completed with exit code 1.
Documentation
Process completed with exit code 2.
basedpyright: pytato/target/loopy/codegen.py#L526
Argument of type "int64" cannot be assigned to parameter "dtype" of type "dtype[Any]" in function "add_store"   Type "int64" is not assignable to type "dtype[Any]" (reportArgumentType)
basedpyright: pytato/target/loopy/codegen.py#L522
Argument of type "Expression" cannot be assigned to parameter "expr" of type "ScalarExpression" in function "__init__"   Type "Expression" is not assignable to type "ScalarExpression"     Type "tuple[Expression, ...]" is not assignable to type "ScalarExpression"       "tuple[Expression, ...]" is not assignable to "int"       "tuple[Expression, ...]" is not assignable to "integer[Any]"       "tuple[Expression, ...]" is not assignable to "float"       "tuple[Expression, ...]" is not assignable to "complex"       "tuple[Expression, ...]" is not assignable to "inexact[Any, float | complex]"       "tuple[Expression, ...]" is not assignable to "ExpressionNode" ... (reportArgumentType)
basedpyright: pytato/array.py#L2285
Cannot access attribute "array" for class "SparseMatmul*"   Attribute "array" is unknown (reportAttributeAccessIssue)
basedpyright: pytato/array.py#L2285
Cannot access attribute "matrix" for class "SparseMatmul*"   Attribute "matrix" is unknown (reportAttributeAccessIssue)
basedpyright: pytato/array.py#L2284
"dtype" incorrectly overrides property of same name in class "Array" (reportIncompatibleMethodOverride)
basedpyright: pytato/array.py#L2280
Cannot access attribute "array" for class "SparseMatmul*"   Attribute "array" is unknown (reportAttributeAccessIssue)
basedpyright: pytato/array.py#L2280
Cannot access attribute "matrix" for class "SparseMatmul*"   Attribute "matrix" is unknown (reportAttributeAccessIssue)
basedpyright: pytato/array.py#L2279
"shape" incorrectly overrides property of same name in class "Array" (reportIncompatibleMethodOverride)
basedpyright: pytato/array.py#L2250
Base classes for class "SparseMatmul" define variable "non_equality_tags" in incompatible way (reportIncompatibleVariableOverride)
basedpyright: pytato/array.py#L2250
Base classes for class "SparseMatmul" define variable "axes" in incompatible way (reportIncompatibleVariableOverride)
Tests for downstream project arraycontext
Process completed with exit code 1.
Tests for downstream project arraycontext: arraycontext/test/test/test_arraycontext.py#L282
test_array_context_np_workalike[<_PytatoPyOpenCLArrayContextForTests for <pyopencl.Device 'cpu-haswell-AMD EPYC 7763 64-Core Processor' on 'Portable Computing Language'>>-maximum-2-float64] NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Tests for downstream project arraycontext: arraycontext/test/test/test_arraycontext.py#L282
test_array_context_np_workalike[<_PytatoPyOpenCLArrayContextForTests for <pyopencl.Device 'cpu-haswell-AMD EPYC 7763 64-Core Processor' on 'Portable Computing Language'>>-minimum-2-float64] NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Tests for downstream project grudge
The strategy configuration was canceled because "downstream_tests.arraycontext" failed
Tests for downstream project grudge
The operation was canceled.
Tests for downstream project grudge: grudge/test/test/test_dt_utils.py#L86
test_geometric_factors_regular_refinement[<PytatoPyOpenCLArrayContext for <pyopencl.Device 'cpu-haswell-AMD EPYC 7763 64-Core Processor' on 'Portable Computing Language'>>-box3d] NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Tests for downstream project grudge: grudge/test/test/test_dt_utils.py#L86
test_geometric_factors_regular_refinement[<PytatoPyOpenCLArrayContext for <pyopencl.Device 'cpu-haswell-AMD EPYC 7763 64-Core Processor' on 'Portable Computing Language'>>-box2d] NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Tests for downstream project grudge: grudge/test/test/test_dt_utils.py#L86
test_geometric_factors_regular_refinement[<PytatoPyOpenCLArrayContext for <pyopencl.Device 'cpu-haswell-AMD EPYC 7763 64-Core Processor' on 'Portable Computing Language'>>-interval] NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Tests for downstream project meshmode
The strategy configuration was canceled because "downstream_tests.arraycontext" failed
Tests for downstream project meshmode
The operation was canceled.
Tests for downstream project meshmode: meshmode/test/test/test_array.py#L150
test_dof_array_pickling[<PytatoPyOpenCLArrayContext for <pyopencl.Device 'cpu-haswell-AMD EPYC 7763 64-Core Processor' on 'Portable Computing Language'>>] NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Tests for downstream project meshmode: meshmode/test/test/test_array.py#L126
test_container_norm[<PytatoPyOpenCLArrayContext for <pyopencl.Device 'cpu-haswell-AMD EPYC 7763 64-Core Processor' on 'Portable Computing Language'>>-inf] NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Conda Pytest
Process completed with exit code 1.
Conda Pytest: test/test/test_codegen.py#L1473
test_random_dag_against_numpy[<context factory for <pyopencl.Device 'cpu-haswell-AMD EPYC 7763 64-Core Processor' on 'Portable Computing Language'>>] NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Conda Pytest: test/test/test_distributed.py#L248
test_distributed_execution_random_dag subprocess.CalledProcessError: Command '['mpiexec', '-np', '2', '--oversubscribe', '-x', 'RUN_WITHIN_MPI=1', '-x', 'INVOCATION_INFO=gAWVRQAAAAAAAACMEHRlc3RfZGlzdHJpYnV0ZWSUjClfZG9fdGVzdF9kaXN0cmlidXRlZF9leGVjdXRpb25fcmFuZG9tX2RhZ5STlCmGlC4=', '/home/runner/work/pytato/pytato/.conda-root/envs/testing/bin/python3', '-m', 'mpi4py', '/home/runner/work/pytato/pytato/test/test_distributed.py']' returned non-zero exit status 1.
Conda Pytest: test/test/test_codegen.py#L1689
test_deterministic_codegen NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Conda Pytest: test/test/test_codegen.py#L1625
test_regression_reduction_in_conditional[<context factory for <pyopencl.Device 'cpu-haswell-AMD EPYC 7763 64-Core Processor' on 'Portable Computing Language'>>] NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Conda Pytest: test/test/test_codegen.py#L746
test_maximum_minimum[<context factory for <pyopencl.Device 'cpu-haswell-AMD EPYC 7763 64-Core Processor' on 'Portable Computing Language'>>-minimum] NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Conda Pytest: test/test/test_codegen.py#L746
test_maximum_minimum[<context factory for <pyopencl.Device 'cpu-haswell-AMD EPYC 7763 64-Core Processor' on 'Portable Computing Language'>>-maximum] NotImplementedError: ReductionBoundsCollector cannot handle <class 'pymbolic.primitives.NaN'>
Tests for downstream project mirgecom
No point in testing mirgecom at the moment, see https://github.com/illinois-ceesd/mirgecom/pull/898. Test not performed.
Ruff
The `python-version` input is not set. The version of Python currently in `PATH` will be used.
basedpyright: pytato/array.py#L2285
Type of "dtype" is unknown (reportUnknownMemberType)
basedpyright: pytato/array.py#L2285
Type of "array" is unknown (reportUnknownMemberType)
basedpyright: pytato/array.py#L2285
Argument type is unknown   Argument corresponds to parameter "arrays_and_dtypes" in function "result_type" (reportUnknownArgumentType)
basedpyright: pytato/array.py#L2285
Type of "dtype" is unknown (reportUnknownMemberType)
basedpyright: pytato/array.py#L2285
Type of "matrix" is unknown (reportUnknownMemberType)
basedpyright: pytato/array.py#L2280
Type of "shape" is unknown (reportUnknownMemberType)
basedpyright: pytato/array.py#L2280
Type of "array" is unknown (reportUnknownMemberType)
basedpyright: pytato/array.py#L2280
Type of "shape" is unknown (reportUnknownMemberType)
basedpyright: pytato/array.py#L2280
Type of "matrix" is unknown (reportUnknownMemberType)
basedpyright: pytato/array.py#L2280
Return type, "tuple[Unknown, *tuple[Unknown, ...]]", is partially unknown (reportUnknownVariableType)
Tests for downstream project arraycontext: arraycontext/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project arraycontext: arraycontext/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project arraycontext: arraycontext/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project arraycontext: arraycontext/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project arraycontext: arraycontext/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project arraycontext: arraycontext/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project arraycontext: arraycontext/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project arraycontext: arraycontext/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project arraycontext: arraycontext/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project arraycontext: arraycontext/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project grudge: grudge/.conda-root/envs/testing/lib/python3.14/site-packages/meshmode/dof_array.py#L79
Broadcasting array context array types across <class 'meshmode.dof_array.DOFArray'> has been implicitly enabled. As of 2026, this will no longer work. Use arraycontext.Bcast* object wrappers for roughly equivalent functionality. See the discussion in https://github.com/inducer/arraycontext/pull/190. To opt out now (and avoid this warning), pass _bcast_actx_array_type=False.
Tests for downstream project grudge: grudge/.conda-root/envs/testing/lib/python3.14/site-packages/meshmode/dof_array.py#L79
'bcast_numpy_array=True' is deprecated and will be unsupported from 2025.
Tests for downstream project grudge: grudge/test/test_reductions.py#L172
Broadcasting array context array types across <class 'test_reductions.MyContainer'> has been implicitly enabled. As of 2026, this will no longer work. Use arraycontext.Bcast* object wrappers for roughly equivalent functionality. See the discussion in https://github.com/inducer/arraycontext/pull/190. To opt out now (and avoid this warning), pass _bcast_actx_array_type=False.
Tests for downstream project grudge: grudge/.conda-root/envs/testing/lib/python3.14/site-packages/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project grudge: grudge/.conda-root/envs/testing/lib/python3.14/site-packages/meshmode/dof_array.py#L79
Broadcasting array context array types across <class 'meshmode.dof_array.DOFArray'> has been implicitly enabled. As of 2026, this will no longer work. Use arraycontext.Bcast* object wrappers for roughly equivalent functionality. See the discussion in https://github.com/inducer/arraycontext/pull/190. To opt out now (and avoid this warning), pass _bcast_actx_array_type=False.
Tests for downstream project grudge: grudge/.conda-root/envs/testing/lib/python3.14/site-packages/meshmode/dof_array.py#L79
'bcast_numpy_array=True' is deprecated and will be unsupported from 2025.
Tests for downstream project grudge: grudge/test/test_reductions.py#L172
Broadcasting array context array types across <class 'test_reductions.MyContainer'> has been implicitly enabled. As of 2026, this will no longer work. Use arraycontext.Bcast* object wrappers for roughly equivalent functionality. See the discussion in https://github.com/inducer/arraycontext/pull/190. To opt out now (and avoid this warning), pass _bcast_actx_array_type=False.
Tests for downstream project grudge: grudge/.conda-root/envs/testing/lib/python3.14/site-packages/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project grudge: grudge/.conda-root/envs/testing/lib/python3.14/site-packages/meshmode/dof_array.py#L79
Broadcasting array context array types across <class 'meshmode.dof_array.DOFArray'> has been implicitly enabled. As of 2026, this will no longer work. Use arraycontext.Bcast* object wrappers for roughly equivalent functionality. See the discussion in https://github.com/inducer/arraycontext/pull/190. To opt out now (and avoid this warning), pass _bcast_actx_array_type=False.
Tests for downstream project grudge: grudge/.conda-root/envs/testing/lib/python3.14/site-packages/meshmode/dof_array.py#L79
'bcast_numpy_array=True' is deprecated and will be unsupported from 2025.
Tests for downstream project meshmode: meshmode/test/test_array.py#L59
'bcast_obj_array' is deprecated and will be unsupported from 2025. Use 'bcasts_across_obj_array', with equivalent meaning.
Tests for downstream project meshmode: meshmode/test/test_array.py#L59
'bcast_obj_array' is deprecated and will be unsupported from 2025. Use 'bcasts_across_obj_array', with equivalent meaning.
Tests for downstream project meshmode: meshmode/meshmode/dof_array.py#L79
Broadcasting array context array types across <class 'meshmode.dof_array.DOFArray'> has been implicitly enabled. As of 2026, this will no longer work. Use arraycontext.Bcast* object wrappers for roughly equivalent functionality. See the discussion in https://github.com/inducer/arraycontext/pull/190. To opt out now (and avoid this warning), pass _bcast_actx_array_type=False.
Tests for downstream project meshmode: meshmode/meshmode/dof_array.py#L79
Broadcasting array context array types across <class 'meshmode.dof_array.DOFArray'> has been implicitly enabled. As of 2026, this will no longer work. Use arraycontext.Bcast* object wrappers for roughly equivalent functionality. See the discussion in https://github.com/inducer/arraycontext/pull/190. To opt out now (and avoid this warning), pass _bcast_actx_array_type=False.
Tests for downstream project meshmode: meshmode/meshmode/dof_array.py#L79
'bcast_numpy_array=True' is deprecated and will be unsupported from 2025.
Tests for downstream project meshmode: meshmode/meshmode/dof_array.py#L79
'bcast_numpy_array=True' is deprecated and will be unsupported from 2025.
Tests for downstream project meshmode: meshmode/.conda-root/envs/testing/lib/python3.14/site-packages/arraycontext/container/dataclass.py#L82
Encountered 'numpy.ndarray' in a dataclass_array_container. This is deprecated and will stop working in 2026. If you meant an object array, use pytools.obj_array.ObjectArray. For other uses, file an issue to discuss.
Tests for downstream project meshmode: meshmode/test/test_array.py#L59
'bcast_obj_array' is deprecated and will be unsupported from 2025. Use 'bcasts_across_obj_array', with equivalent meaning.
Tests for downstream project meshmode: meshmode/meshmode/dof_array.py#L79
Broadcasting array context array types across <class 'meshmode.dof_array.DOFArray'> has been implicitly enabled. As of 2026, this will no longer work. Use arraycontext.Bcast* object wrappers for roughly equivalent functionality. See the discussion in https://github.com/inducer/arraycontext/pull/190. To opt out now (and avoid this warning), pass _bcast_actx_array_type=False.
Tests for downstream project meshmode: meshmode/meshmode/dof_array.py#L79
'bcast_numpy_array=True' is deprecated and will be unsupported from 2025.
Conda Pytest: pytato/target/loopy/__init__.py#L251
TranslationUnit.__call__ is uncached as of 2025, meaning it will incur possibly substantial compilation cost with every invocation. Use TranslationUnit.executor to obtain an object that holds longer-lived caches.
Conda Pytest: pytato/target/loopy/__init__.py#L251
TranslationUnit.__call__ is uncached as of 2025, meaning it will incur possibly substantial compilation cost with every invocation. Use TranslationUnit.executor to obtain an object that holds longer-lived caches.
Conda Pytest: pytato/target/loopy/__init__.py#L251
TranslationUnit.__call__ is uncached as of 2025, meaning it will incur possibly substantial compilation cost with every invocation. Use TranslationUnit.executor to obtain an object that holds longer-lived caches.
Conda Pytest: pytato/target/loopy/__init__.py#L251
TranslationUnit.__call__ is uncached as of 2025, meaning it will incur possibly substantial compilation cost with every invocation. Use TranslationUnit.executor to obtain an object that holds longer-lived caches.
Conda Pytest: pytato/target/loopy/__init__.py#L251
TranslationUnit.__call__ is uncached as of 2025, meaning it will incur possibly substantial compilation cost with every invocation. Use TranslationUnit.executor to obtain an object that holds longer-lived caches.
Conda Pytest: pytato/target/loopy/__init__.py#L251
TranslationUnit.__call__ is uncached as of 2025, meaning it will incur possibly substantial compilation cost with every invocation. Use TranslationUnit.executor to obtain an object that holds longer-lived caches.
Conda Pytest: pytato/target/loopy/__init__.py#L251
TranslationUnit.__call__ is uncached as of 2025, meaning it will incur possibly substantial compilation cost with every invocation. Use TranslationUnit.executor to obtain an object that holds longer-lived caches.
Conda Pytest: pytato/target/loopy/__init__.py#L251
TranslationUnit.__call__ is uncached as of 2025, meaning it will incur possibly substantial compilation cost with every invocation. Use TranslationUnit.executor to obtain an object that holds longer-lived caches.
Conda Pytest: pytato/target/loopy/__init__.py#L251
TranslationUnit.__call__ is uncached as of 2025, meaning it will incur possibly substantial compilation cost with every invocation. Use TranslationUnit.executor to obtain an object that holds longer-lived caches.
Conda Pytest: pytato/target/loopy/__init__.py#L251
TranslationUnit.__call__ is uncached as of 2025, meaning it will incur possibly substantial compilation cost with every invocation. Use TranslationUnit.executor to obtain an object that holds longer-lived caches.