commit | a8c6561c0da87cd056e639fdccea685866868693 | [log] [tgz] |
---|---|---|
author | Stella Laurenzo <laurenzo@google.com> | Tue Mar 21 23:01:00 2023 -0700 |
committer | GitHub <noreply@github.com> | Tue Mar 21 23:01:00 2023 -0700 |
tree | f090e15d95dbc9fab641814f9175608708ee626c | |
parent | 475971c68c56f159333fc672c093af304a7a7d18 [diff] |
Add compiler CAPI for output to memory and mapping. (#12714) * Adds entrypoints `ireeCompilerOutputOpenMembuffer` and `ireeCompilerOutputMapMemory`. * On Linux, these outputs are backed by memfd_create and mmap. * On others (Windows/MacOS), they just use std::string streams to accumulate output. This replaces a bunch of platform specific code in the PJRT plugin that was being used to directly map executable outputs from the compiler and makes it available to everyone. It will be used to rework iree-run-mlir to a pure CAPI using JIT.
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.
IREE is still in its early phase. We have settled down on the overarching infrastructure and are actively improving various software components as well as project logistics. It is still quite far from ready for everyday use and is made available without any support at the moment. With that said, we welcome any kind of feedback on any communication channels!
See our website for more information.
IREE is licensed under the terms of the Apache 2.0 License with LLVM Exceptions. See LICENSE for more information.