Backend Planning for IR Graphs #
Adapter from TorchLean's op-tagged IR to backend capsules.
This is intentionally a planning layer, not an evaluator. The graph remains the semantic object; the planner selects which backend contract should implement each node or node family.
Backend operation tag for an IR op.
These tags are the same vocabulary used by backend capsules and runtime reports. If an op maps to a tag that has no capsule for the selected profile, planning fails at the backend boundary instead of silently using a broader bucket.
Instances For
Backend operation requested by a graph node, if the node needs runtime work.
Instances For
Backend choice for one concrete IR node.
- nodeId : ℕ
- kind : IR.OpKind
- op : BackendOp
- capsule : KernelCapsule
Instances For
Instances For
Graph-aware execution plan that preserves the IR node identity for every backend choice.
- kernels : List PlannedNodeKernel
Instances For
Instances For
Node ids covered by backend kernels, in graph order.
Instances For
Selected capsule names, in graph order.
Instances For
Plan a single IR node when it corresponds to runtime work.
Instances For
Plan every runtime-relevant node in graph order.
Instances For
Check graph well-formedness, then plan every runtime-relevant node.