| commit | f1adc7897c620fd835da718c68ce0c4156d45688 | [log] [tgz] |
|---|---|---|
| author | Scott Todd <scott.todd0@gmail.com> | Tue Feb 06 11:24:51 2024 -0800 |
| committer | GitHub <noreply@github.com> | Tue Feb 06 11:24:51 2024 -0800 |
| tree | 99bea6fd2af8a0441e1e95679384f17f6d90e3c0 | |
| parent | 93193cef60478ffcac315042c73b8d428def4278 [diff] |
Fold test_all into build_all and run iree-dialects tests. (#16318) Progress on https://github.com/openxla/iree/issues/16203. The eventual goal with these jobs is to have a build job run _unit tests_ with the full build directory already on disk and then test jobs run _integration tests_ using packages. This is one incremental step that together with a few other PRs will let us drop the `build-dir` artifact generation and passing. Another PR will split "integration tests" out and add back a "test_cpu" job, and I'll eventually adjust the job names. ~~Adding coverage for iree-dialects tests here will also let us consider switching the ASan job to be opt-in / postsubmit only.~~ The new steps add about 1m30s to the build_all job. We can get 30s of that back by turning off the `iree-test-deps` build (for "integration tests"), when those are moved back to another job.
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.