| commit | f8a0c44e19c6fd1acc6ef9f9a0370fb16fcd1c2e | [log] [tgz] |
|---|---|---|
| author | Oleksandr "Alex" Zinenko <zinenko@google.com> | Fri May 06 13:50:49 2022 +0200 |
| committer | GitHub <noreply@github.com> | Fri May 06 13:50:49 2022 +0200 |
| tree | c152cb1d922b605b89f824484b61cdda12917882 | |
| parent | 2ab9cb3011cc1039a6981efc4fc1352b51e50f79 [diff] |
Use upstream MLIR Transform dialect (#9057)
* Cherry-pick LLVM to add support for Tile Transform Op
3c2a74a3ae02d16e899e280953c055f92aa6cdaa
* Port transform ops to use upstream interfaces
MLIR upstream now includes the Transform dialect along with op
interfaces and the new extension mechanism. Port most transform ops to
use the upstream interface. The LinalgTransform dialect still contains
the "util" ops for scoped transformation, which arguably should have
been placed in a separate dialect to start with. These will be cleaned
up separately as a decision is required whether to keep them at all.
Notable changes in the naming scheme:
- `iree_linalg_transform.` goes away, general `transform.` should
be used instead as the ops are now injected into the transform
dialect;
- `*.iree_*` transform ops are now `transform.iree.*` to align on the
naming scheme of the upstream transform dialect;
- transform ops related to the structured ops philosophy are prefixed
with `structured.`, e.g., `transform.structured.pad`;
- due to a bug in the upstream parser, three-piece op names cannot
omit the dialect prefix inside transform ops, i.e.
`transform.structured.pad` must be spelled out completely instead of
using the shorter `structured.pad` form.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!
| CI System | Build System | Platform | Architecture | Configuration / Component | Status |
|---|---|---|---|---|---|
| Kokoro | Bazel | Linux | x86-64 |  | |
| Kokoro | CMake & Bazel | Linux | x86-64 (swiftshader) | Integrations |  |
| Kokoro | CMake & Bazel | Linux | x86-64 (turing) | Integrations |  |
| Kokoro | CMake | Linux | x86-64 (swiftshader) |  | |
| Kokoro | CMake | Linux | x86-64 (swiftshader) | asan |  |
| Kokoro | CMake | Linux | x86-64 (turing) |  | |
| Kokoro | CMake | Android | arm64-v8a | Runtime (build only) |  |
| Kokoro | CMake | Bare Metal | risc-v-32 | Runtime |  |
| Kokoro | CMake | Linux | risc-v-64 | Runtime |  |
| Buildkite | CMake | Android | arm64-v8a | Runtime |  |
| BuildKite | CMake | Android | arm64-v8a | Runtime Benchmarks |  |
| BuildKite | CMake | Linux | x86-64 | Tracing + Standalone Runtime |  |
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.