Skip to content

mesh-node-not-leaf

Status: stub. The full-length analysis is queued for a v1.0.x patch release per ADR 0018, section 5, criterion #6. The companion rule page at docs/rules/mesh-node-not-leaf.md contains the canonical detection logic + GPU reasoning.

TL;DR

Mesh nodes integrate the mesh-shader pipeline into work graphs: a node receives input records from upstream nodes and emits vertices / primitives to the rasterizer rather than records to a downstream node. On NVIDIA Ada Lovelace and AMD RDNA 3 (the IHVs initially supporting the preview API), the runtime schedules mesh nodes against the rasterizer's input queue exactly as DispatchMesh does, but with the work-graph dispatch arguments standing in for the application's DispatchMesh arguments.

What the rule fires on

A work-graph mesh node (a node with [NodeLaunch("mesh")]) that has any outgoing NodeOutput<...> declarations. The Mesh Nodes in Work Graphs preview specification requires mesh nodes to be leaf nodes — they consume input records but produce only rasterized output. Slang reflection identifies the node-launch kind and the output declarations; the rule fires when both coexist.

See the What it detects section of the rule page for the full pattern definition.

Why it matters

The full GPU-mechanism analysis lives in the Why it matters on a GPU section of the companion rule page.

Examples

The bad / good code snippets are kept canonical on the rule page; see mesh-node-not-leaf.md -> Examples.

See also


This is a v1.0-ship stub. Full analysis pending; track issue link TBD.

© 2026 NelCit — Apache-2.0 (code), CC-BY-4.0 (docs).