mirror of
https://github.com/ROCm/jax.git
synced 2025-04-15 19:36:06 +00:00
f93c2a1aa5
On CPU and GPU, almost all of the primitives in lax.linalg are backed by custom calls that support simple semantics when batch dimensions are sharded. Before this change, all linalg operations on CPU and GPU will insert an `all-gather` before being executed when called on sharded inputs, even when that shouldn't be necessary. This change adds support for this type of partitioning, to cover a wide range of use cases. There are a few remaining GPU ops that don't support partitioning either because they are backed by HLO ops that don't partition properly (Cholesky factorization and triangular solves), or because they're still using descriptors with problem dimensions in kernel. I'm going to fix these in follow up changes. PiperOrigin-RevId: 731732301
jaxlib: support library for JAX
jaxlib is the support library for JAX. While JAX itself is a pure Python package, jaxlib contains the binary (C/C++) parts of the library, including Python bindings, the XLA compiler, the PJRT runtime, and a handful of handwritten kernels. For more information, including installation and build instructions, refer to main JAX README: https://github.com/jax-ml/jax/.