xla
xla copied to clipboard
openxla
A machine learning compiler for GPUs, CPUs, and ML accelerators
This PR introduces a `.pylintrc` file to the XLA project. XLA, which is used in TensorFlow, relies on TensorFlow's CI builds to automatically test its code, including Python components like...
Algebraic simplifier: simplify exp(0) to 1.
![copybara-service[bot] avatar](https://avatars.githubusercontent.com/in/44061?v=4 "copybara-service[bot] avatar"){kind=avatar}
Simplify STREAM_EXECUTOR_REGISTER_MODULE_INITIALIZER support to be the same in all environments. Also removed some unneeded related macros.
PR #74090: Removing distutils leftover Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/74090 This aims to finish #58073. The patch is untested but follow https://setuptools.pypa.io/en/latest/deprecated/distutils-legacy.html. I did not touch the install scripts. Thanks,...
collective_send_recv_combiner prototype implementation: wrap send/recv into async-start calls
[XLA_UNSTACKER] Fix a bug in xla unstacker which would triggered when there are multiple uses of unstackable operand at the loop root.
Use RunAndCheckHloRewrite in collective_permute_cycle_decomposer_test.cc
PR #76831: aarch64: set compute library openmp runtime to off by default Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/76831 since the official wheel binary is using Eigen thread pool runtime, there is...
Hello everyone ! I encounter a weird issue trying to compile OpenXLA on Mac M1 for CPU (Sonoma 14.7). By running the following command, after having `python configure.py --config=CPU` :...
