commit | 4d91ffb09a8e48844b6460edce58ccba245bced7 | [log] [tgz] |
---|---|---|
author | Oleksandr "Alex" Zinenko <git@ozinenko.com> | Mon Sep 01 13:52:27 2025 +0200 |
committer | GitHub <noreply@github.com> | Mon Sep 01 11:52:27 2025 +0000 |
tree | 6d734064488bc9477a5474d8c127279392feee38 | |
parent | c37c680ae6e71f715bd7c540909155061bc44491 [diff] |
[codegen] more consumer fusion (#21521) Enable consumer fusion to fuse more trailing operations, in particular to fuse all operations used in layer normalization. The particular enablement is to allow fusion of consumers of fused operations that are not the "main" operation into which fusion happens. This required three modifications: 1. Generate replacement values for any fused operation whose results are used by operations dominated by the main operation (that will turn into a loop). This is needed because the consumer fusion finds candidates based on `insert_slice` operations in the loop that are not produced unless the replacement values are generated. In the case of layer normalizaiton, we also need actual replacement values so we can return them. Spurious values are being cleaned up by canonicalization patterns in the same pass. 2. Avoid generating replacement values for values used as DPS inits as those cannot (currently) be fused. 3. Sort consumers to fuse by dominance and first most-dominating consumer fuse to avoid triangular-use situation that prevents fusion. This generally makes fusion more aggressive on the consumer side. Signed-off-by: Alex Zinenko <git@ozinenko.com>
IREE (Intermediate Representation Execution Environment, 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.