| commit | 8007d749733c5109cddc323eb8b52a5f2a0bad9f | [log] [tgz] |
|---|---|---|
| author | Ben Vanik <ben.vanik@gmail.com> | Mon Mar 06 10:59:50 2023 -0800 |
| committer | GitHub <noreply@github.com> | Mon Mar 06 10:59:50 2023 -0800 |
| tree | f17028e5f7130b115ccc57b5b7458ba699ef35b4 | |
| parent | e6425d6b193843e560ed9f3f669d3e3476d590bd [diff] | |
| parent | ff9bc6abbaa6c3b6245d54b5fece9eed36371cc1 [diff] |
Adding min/max VM ops and VM buffer allocation alignment. (#12522) Ops for i32/i64/f32 variants of min/max have been added to avoid emulated compare + select sequences and `vm.buffer.alloc` (and `vm.buffer.clone`) have gained an alignment parameter that's been on the backlog for ages: now we properly pass alignment all the way from memref/util.buffer down to the `iree_vm_buffer_*` allocation calls. This also fixes some missing `vm.round.f32` disassembly support and tests that weren't added when the op was. This breaks binary compatibility due to the buffer alloc encoding changes. Fixes #9165.
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.