FHElium

fhelium.backend.implementation ​

Register and resolve operation implementations contributed by Backend components.

Implementation classes declare the operations they support and a stable name. The registry stores supplied implementation objects, rejects duplicate identities, and resolves operation-class and name selections.

FusionImplementation ​

class View source

python
FusionImplementation()

Bases: OperationImplementation, Protocol

Describe region support supplied by one joint-execution implementation.

Implementations may provide new_fusion_matcher() -> FusionMatcher to analyze successive candidates incrementally during one scheduling window. Other implementations use match_fusion for each complete candidate.

match_fusion ​

method

python
def match_fusion(operations: Sequence[Operation]) -> int | None: ...

Return a computational-op count for an eligible fusion candidate.

Zero denotes plumbing; None denotes an unsupported or ineligible candidate. Unknown facts may remain unresolved until execution binding. The count records computational operations.

FusionMatcher ​

class View source

python
FusionMatcher()

Bases: Protocol

Maintain candidate facts while a scheduler reads an unchanged SSA graph.

probe ​

method

python
def probe(operation: Operation) -> int | None: ...

Match an extension without committing it to the current region.

append ​

method

python
def append(operation: Operation) -> None: ...

Commit a selected extension, even if this implementation rejects it.

reset ​

method

python
def reset() -> None: ...

Discard region facts before considering a new region.

ImplementationRegistry ​

class View source

python
ImplementationRegistry(implementations: Sequence[ImplementationT]=(), *, category: str)

Store implementations by supported operation class and stable name.

Backend assembly supplies the implementation objects and their declarations. Resolution selects among those registered candidates.

operation_types ​

property

python
operation_types: tuple[type[Operation], ...]

Return operation classes in first declaration order.

implementations ​

property

python
implementations: tuple[ImplementationT, ...]

Return implementation contributions in construction order.

available ​

method

python
def available(operation_type: type[Operation]) -> tuple[str, ...]: ...

Return implementation names in declaration order.

supports ​

method

python
def supports(operation_type: type[Operation], *, name: str | None=None) -> bool: ...

Return whether at least one matching implementation is registered.

resolve ​

method

python
def resolve(operation_type: type[Operation], *, requested: str | None) -> ImplementationT: ...

Resolve one named implementation, refusing ambiguous defaults.

OperationImplementation ​

class View source

python
OperationImplementation()

Bases: Protocol

Execute one or more registered operations on Tensor payloads.

name ​

property

python
name: str

operation_types ​

property

python
operation_types: tuple[type[Operation], ...]

supports_in_place ​

property

python
supports_in_place: bool

resource_requirements ​

method

python
def resource_requirements(invocation: OperationInvocation) -> tuple[ResourceRequirement, ...]: ...

execute ​

method

python
def execute(invocation: OperationInvocation, inputs: tuple[torch.Tensor, ...], resources: tuple[BoundResource, ...], /, *, in_place: bool) -> tuple[torch.Tensor, ...]: ...

OperationImplementationRegistry ​

class View source

python
OperationImplementationRegistry(implementations: Sequence[OperationImplementation]=(), *, operation_specs: OperationSpecRegistry=DEFAULT_OPERATION_SPECS)

Resolve implementations by operation class and identity.

operation_types ​

property

python
operation_types: tuple[type[Operation], ...]

implementations ​

property

python
implementations: tuple[OperationImplementation, ...]

Return implementation contributions in construction order.

with_implementations ​

method

python
def with_implementations(implementations: Sequence[OperationImplementation]) -> OperationImplementationRegistry: ...

Return this registry extended by caller-supplied contributions.

available ​

method

python
def available(operation_type: type[Operation]) -> tuple[str, ...]: ...

supports ​

method

python
def supports(operation: Operation, *, implementation: str | None=None) -> bool: ...

resolve ​

method

python
def resolve(operation: Operation, *, requested: str | None, in_place: bool=False) -> OperationImplementation: ...

resolve_type ​

method

python
def resolve_type(operation_type: type[Operation], *, requested: str | None, in_place: bool=False) -> OperationImplementation: ...

effect ​

method

python
def effect(operation_type: type[Operation]) -> OperationEffect: ...

Return the registered effect of one operation class.

OperationInvocation ​

class View source

python
OperationInvocation(operation_type: type[Operation], operand_count: int, result_count: int, attributes: Mapping[str, object] = field(default_factory=dict), operand_prime_ids: tuple[tuple[int, ...] | None, ...] = (), operand_bases: tuple[str | None, ...] = (), operand_components: tuple[int | None, ...] = (), result_prime_ids: tuple[tuple[int, ...] | None, ...] = ())

Describe an operation class, operand/result counts, attributes, and represented value metadata.

Attributes ​

NameTypeDefault/value
operation_typetype[Operation]
operand_countint
result_countint
attributesMapping[str, object]field(default_factory=dict)
operand_prime_idstuple[tuple[int, ...] | None, ...]()
operand_basestuple[str | None, ...]()
operand_componentstuple[int | None, ...]()
result_prime_idstuple[tuple[int, ...] | None, ...]()

PreparingOperationImplementation ​

class View source

python
PreparingOperationImplementation()

Bases: OperationImplementation, Protocol

Prepare one represented operation for direct Tensor execution.

Code-generating implementations consume the operation and its regions once during Backend resolution. The returned implementation executes the complete operation without interpreting its nested operations at runtime.

prepare_operation ​

method

python
def prepare_operation(operation: Operation) -> OperationImplementation: ...

RegionCallable ​

data View source

python
RegionCallable = Callable[[tuple[torch.Tensor, ...]], tuple[torch.Tensor, ...]]

RegionOperationImplementation ​

class View source

python
RegionOperationImplementation()

Bases: OperationImplementation, Protocol

Execute an operation whose regions contain callable Tensor programs.

execute_regions ​

method

python
def execute_regions(invocation: OperationInvocation, inputs: tuple[torch.Tensor, ...], resources: tuple[BoundResource, ...], regions: tuple[RegionCallable, ...], /, *, in_place: bool) -> tuple[torch.Tensor, ...]: ...

operation_invocation ​

function View source

python
def operation_invocation(operation: Operation) -> OperationInvocation: ...

Convert one linked IR operation to its runtime invocation descriptor.

requested_implementation ​

function View source

python
def requested_implementation(operation: Operation, caller_selection: str | None=None) -> str | None: ...

Return the implementation recorded in IR or selected by the caller.

An assignment written by a Compile pass is a hard constraint. A caller may select an unassigned operation, but it may not contradict the Program.