[VectorDistribute] Refactor VectorLayoutAnalysis into 2-phase forward/backward design (#23611) Restructure the layout analysis from an interleaved forward+backward worklist into a clean two-phase design: Phase 1 (forward): Multi-candidate propagation from ToLayoutOp anchors through uses. No IR mutation. Resolve: Pick first candidate per value. This is a placeholder cost model for now which matches the old analysis. Eventually, we will consider coalescing, compute ops, mma layout, etc. Phase 2 (backward fixup): Walk operations in reverse program order via recursive fixupRegion/fixupOp. For each op, derive operand layouts from resolved result layouts. Assign missing layouts, clone cheap ops (constants, create_mask, step), or insert to_layout conversions on conflict. This naturally handles conflicts better in a predictable manner. // The forward analysis is the main driver of the analysis. The reason for this is that for a program to be well-formed for vector distribution, there must be some way for the final store/return to get a layout. Otherwise, there is not enough information in the program to determine how distribution should be done. The forward analysis ensures that the final return/store gets a layout in a well-formed program. The rest of the program can get their layouts from backward propagation, everything in the program must eventually reach the store/return. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
IREE (Intermediate Representation Execution Eenvironment, pronounced as “eerie”) is an MLIR-based end-to-end compiler and runtime that lowers Machine Learning (ML) models to a unified IR that scales up to meet the needs of the datacenter and down to satisfy the constraints and special considerations of mobile and edge deployments.
See our website for project details, user guides, and instructions on building from source.
Releases notes are published on GitHub releases.
| Package | Release status |
|---|---|
| GitHub release (stable) | |
| GitHub release (nightly) | |
iree-base-compiler | |
iree-base-runtime |
For more details on the release process, see https://iree.dev/developers/general/release-management/.
| Operating system | Build status |
|---|---|
| Linux | |
| macOS | |
| macOS |
For the full list of workflows see https://iree.dev/developers/general/github-actions/.
See our website for more information.
Community meeting recordings: IREE YouTube channel
| Date | Title | Recording | Slides |
|---|---|---|---|
| 2025-06-10 | Data-Tiling in IREE: Achieving High Performance Through Compiler Design (AsiaLLVM) | recording | slides |
| 2025-05-17 | Introduction to GPU architecture and IREE's GPU CodeGen Pipeline | recording | slides |
| 2025-02-12 | The Long Tail of AI: SPIR-V in IREE and MLIR (Vulkanised) | recording | slides |
| 2024-10-01 | Unveiling the Inner Workings of IREE: An MLIR-Based Compiler for Diverse Hardware | recording | |
| 2021-06-09 | IREE Runtime Design Tech Talk | recording | slides |
| 2020-08-20 | IREE CodeGen (MLIR Open Design Meeting) | recording | slides |
| 2020-03-18 | Interactive HAL IR Walkthrough | recording | |
| 2020-01-31 | End-to-end MLIR Workflow in IREE (MLIR Open Design Meeting) | recording | slides |
IREE is licensed under the terms of the Apache 2.0 License with LLVM Exceptions. See LICENSE for more information.