Add WebGPU dependencies and refactor *_TO_BUILD CMake options. (#7905) Part of https://github.com/google/iree/issues/7840 `IREE_HAL_DRIVERS_TO_BUILD` and `IREE_TARGET_BACKENDS_TO_BUILD` now use "default" instead of "all". The new WebGPU compiler target is included in the list of all targets but is _not_ enabled by default, and the Metal+CUDA targets similarly disable themselves on incompatible platforms. Developers must explicitly opt in to building the WebGPU compiler target or the Metal/CUDA targets in unsupported configurations. I'm using `-DIREE_TARGET_BACKENDS_TO_BUILD=CUDA;Dylib-LLVM-AOT;WASM-LLVM-AOT;Vulkan-SPIRV;VMVX;WebGPU`, for example. Thanks to being conditionally enabled, we can also use CMake's [FetchContent](https://cmake.org/cmake/help/latest/module/FetchContent.html) instead of git submodules for some dependencies, which will _hopefully_ keep the rest of IREE's CMake and git infrastructure simple for most users while only adding costs for those us of working on or using these optional components.
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.