docs.rs failed to build rlx-mlx-sys-0.2.18
Please check the build logs for more information.
See Builds for ideas on how to fix a failed build, or Metadata for how to configure docs.rs builds.
If you believe this is docs.rs' fault, open an issue.

rlx-mlx-sys

Vendored MLX C++ (git submodule at vendor/mlx) plus the rlx_mlx_shim C ABI built via CMake + cc in build.rs. Consumed by rlx-mlx; not meant for direct use outside RLX unless you accept the shim API stability policy (none yet).

After clone:

git submodule update --init rlx-mlx-sys/vendor/mlx

The submodule is marked shallow = true in .gitmodules (depth-1 clone of the pinned MLX commit) so local .git stays small. A full MLX history is not required to build.

Linux compile times

Backend	First build (approx.)	Opt-in
CPU only (default)	~5–10 min	always
CUDA (nvcc kernels)	~45–90 min	`RLX_MLX_CUDA=1` or `--features cuda`

CPU-only is the default even when CUDA/cuDNN are installed. Auto-building the CUDA backend added an hour+ to every clean cargo build on WSL rigs.

Faster Linux builds

Install system LAPACK headers (avoids bootstrapping OpenBLAS into OUT_DIR):
```
sudo apt-get install libblas-dev liblapack-dev liblapacke-dev
```
Use debug cargo profile for iteration — build.rs maps cargo build → CMake Debug (much faster nvcc when CUDA is enabled).
Install ccache (MLX enables it automatically when found):
```
sudo apt-get install ccache
```

Pin parallel nvcc jobs if WSL OOMs or thrashes:

RLX_MLX_JOBS=4 cargo build -p rlx-mlx-sys --features cuda

Pin GPU arch when cross-compiling or CI has no GPU:

RLX_MLX_CUDA_ARCH=89 RLX_MLX_CUDA=1 cargo build -p rlx-mlx --features cuda

Runtime device selection (after build): RLX_MLX_DEVICE=cpu|gpu.

Which CUDA toolkit

With CUDA on, build.rs picks the toolkit itself and names it (and any it skipped) in a build warning:

RLX_MLX_CUDA_ROOT (a directory with bin/nvcc) chooses one outright. Otherwise: CUDA_HOME, CUDA_PATH, /usr/local/cuda, /usr/local/cuda-* (newest first), /opt/cuda, then Debian/Ubuntu's nvidia-cuda-toolkit.
Versions MLX refuses are skipped — read from MLX's own CMakeLists.txt (currently 13.1), so the list follows the submodule.
Ubuntu's toolkit splits itself across /usr/lib/nvidia-cuda-toolkit/bin, /usr/include and the multiarch lib dir; the build gives CMake a shim root in OUT_DIR for it.
The CUDA host compiler is the toolkit's own bin/g++ wrapper when it has one, else the newest g++-N within the toolkit's GCC limit when the default c++ is too new. CUDAHOSTCXX overrides.
cuDNN comes from CUDNN_ROOT / CUDNN_PATH, the toolkit, or the system dirs.
The chosen toolkit is also the JIT's default header root at run time (CUDA_HOME / CUDA_PATH still override it), and the JIT uses the CCCL headers MLX installed into OUT_DIR (patch 0003).
When CMake finds NCCL, MLX builds its NCCL distributed backend and the build links the same libnccl.

If CUDA is requested but nothing fits, the build says why per toolkit and builds the CPU-only MLX. Binaries need the chosen toolkit's libraries on the loader path (LD_LIBRARY_PATH for a /usr/local/cuda-* install).

docker/cuda_matrix.py checks all of this against NVIDIA's CUDA images (12.8, 13.0, 13.1 refused, 13.2, two toolkits side by side), on the host's architecture and without a GPU: toolkit selection, then the MLX CUDA compile and the link of every rlx-mlx test binary. --list shows the scenarios.

Full guide: docs/benchmarks/mlx-linux.md.

macOS / Windows

Requires macOS + Xcode (xcrun metal) for Metal kernels. CPU MLX builds on Linux and Windows; on Linux, lapacke.h must be present (liblapacke-dev) or build.rs bootstraps OpenBLAS into OUT_DIR.

License