commit | 0e71e72479f05cfc5635b5bd9765d2f5472e930c | [log] [tgz] |
---|---|---|
author | Andrew Woloszyn <andrew.woloszyn@gmail.com> | Wed Dec 11 08:29:49 2024 -0800 |
committer | GitHub <noreply@github.com> | Wed Dec 11 11:29:49 2024 -0500 |
tree | b9282b5bd21cf22c96c1b1be557986086f28bad4 | |
parent | c31583338a1d8cd099211d8980af7e2dd5791ce5 [diff] |
[hip] Update the hip runtime so that multiple physical devices can back a logical device. (#18790) This gives us an interface for creating a logical device from a set of physical hip devices. Each physical device in the new logical device is represented as a queue in the logical device and therefore can be accessed by specifying a queue affinity for operations and buffers. This is a pretty substantial change to how hip devices are handled. In general: 1. Synchronization is updated so that we no longer have to do a CPU round trip for GPU<->GPU synchronization. 2. The deferred work queue is no longer in use in HIP at all. 3. We have 2 much simpler helper threads per device. They handle keeping high-cost operations off the main thread. Specifically, async memory operations, and queue executions which are emulated in HIP, and have a relatively high cost. Signed-off-by: Andrew Woloszyn <andrew.woloszyn@gmail.com>
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
Package | Release status |
---|---|
GitHub release (stable) | |
GitHub release (nightly) | |
Python iree-base-compiler | |
Python iree-base-runtime |
Host platform | Build status |
---|---|
Linux | |
macOS | |
Windows |
For the full list of workflows see https://iree.dev/developers/general/github-actions/.
See our website for more information.
Community meeting recordings: IREE YouTube channel
IREE is licensed under the terms of the Apache 2.0 License with LLVM Exceptions. See LICENSE for more information.