optimization_config

`OptimizationConfig` `dataclass`

Configuration for running the full-program optimization

`gpu = field(default_factory=GPU)` `class-attribute` `instance-attribute`

GPU-only optimization options

`hint = OptimizationHint.AUTO` `class-attribute` `instance-attribute`

Hint for all optimizations passes

`serial_cpu_vectorization = OptimizationOption.AUTO` `class-attribute` `instance-attribute`

Series of transformed aimed at helping the generic compiler do auto-vectorization for AVX instructions.

`stree = field(default_factory=Tree)` `class-attribute` `instance-attribute`

Schedule Tree optimization options

`GPU` `dataclass`

Optimization dedicated for GPU

`common_gpu_xforms = False` `class-attribute` `instance-attribute`

DaCe common xforms bundled in apply_gpu_transformations

`Tree` `dataclass`

Optimization using the Schedule Tree IR

`enabled = os.getenv('NDSL_STREE_OPT', 'False').lower() == 'true'` `class-attribute` `instance-attribute`

Enable Schedule Tree transformations.

`inline_K_loops_size_one = False` `class-attribute` `instance-attribute`

"Remove serial for loops of size one in the K-axis.

`kernelize = OptimizationOption.AUTO` `class-attribute` `instance-attribute`

Enable maximizing 3-axis kernelization by duplicating maps.

`merger = field(default_factory=Merger)` `class-attribute` `instance-attribute`

Configuration object for cartesian axis merging.

`refine_transients = True` `class-attribute` `instance-attribute`

Reduce dimensionality of transient arrays based on their usage.

`Merger` `dataclass`

`align_lhs_on_center_horizontal = False` `class-attribute` `instance-attribute`

Attempt on aligning the LHS for I and J axis on center so we can further merging. EXPERIMENTAL, therefore defaulting to False.

`align_lhs_on_center_vertical = False` `class-attribute` `instance-attribute`

Attempt on aligning the LHS for K axis on center so we can further merging. EXPERIMENTAL, therefore defaulting to False.

`enabled = True` `class-attribute` `instance-attribute`

Enable cartesian axis merging.

`order = 'default'` `class-attribute` `instance-attribute`

Allows to manually override the merging order (e.g. KJI will merge K, then J, then I). The default follows loop order of the backend given to CartesianMerge.

`overcompute_horizontal = False` `class-attribute` `instance-attribute`

When merging allow horizontal maps (I and J) of different sizes to merge by inserting an if guard.

`overcompute_vertical = True` `class-attribute` `instance-attribute`

When merging allow vertical maps (K) of different sizes to merge by inserting an if guard.

`concretize(backend)`

Concretize the remaining AUTO argument based on various metric.

`is_concretized()`

Return whether no nested configuration contains an automatic hint.

`OptimizationHint`

Bases: Enum

Hint for the configuration system that will drive the OptimizationOption.AUTO value

`AUTO = enum.auto()` `class-attribute` `instance-attribute`

Automatic strategy resolved by querying software and hardware capabilities

`PARALLEL = enum.auto()` `class-attribute` `instance-attribute`

Suitable for GPU or many-thread (parallelization) CPU strategy.

`SERIAL = enum.auto()` `class-attribute` `instance-attribute`

Suitable for many-cores (serial) CPU strategy.

`OptimizationOption`

Bases: Enum

Options for configuration element. AUTO will rely on the best guess default

`APPLY = 1` `class-attribute` `instance-attribute`

Pass will always be applied

`AUTO = 2` `class-attribute` `instance-attribute`

Best guess relying on the OptimizationHint

`DO_NOT_APPLY = 0` `class-attribute` `instance-attribute`

Pass will never be applied

optimization_config