67 Note that SYCL backends require building with oneAPI compilers as well:
144 ## Backends  section in libCEED: Efficient Extensible Discretization
146 There are multiple supported backends, which can be selected at runtime in the examples:
160 | `/cpu/self/memcheck/*`     | Memcheck backends, undefined value checks         | Yes             …
187 The `/cpu/self/*/serial` backends process one element at a time and are intended for meshes with a …
188 The `/cpu/self/*/blocked` backends process blocked batches of eight interlaced elements and are int…
190 The `/cpu/self/ref/*` backends are written in pure C and provide basic functionality.
192 The `/cpu/self/opt/*` backends are written in pure C and use partial e-vectors to improve performan…
194 The `/cpu/self/avx/*` backends rely upon AVX instructions to provide vectorized CPU performance.
196 The `/cpu/self/memcheck/*` backends rely upon the [Valgrind](https://valgrind.org/) Memcheck tool t…
197 To use, run your code with Valgrind and the Memcheck backends, e.g. `valgrind ./build/ex1 -ceed /cp…
201 The `/cpu/self/xsmm/*` backends rely upon the [LIBXSMM](https://github.com/libxsmm/libxsmm) package…
205 The `/gpu/cuda/*` backends provide GPU performance strictly using CUDA.
207 The `/gpu/hip/*` backends provide GPU performance strictly using HIP.
208 They are based on the `/gpu/cuda/*` backends.
211 The `/gpu/sycl/*` backends provide GPU performance strictly using SYCL.
212 They are based on the `/gpu/cuda/*` and `/gpu/hip/*` backends.
214 The `/gpu/*/magma/*` backends rely upon the [MAGMA](https://bitbucket.org/icl/magma) package.
215 To enable the MAGMA backends, the environment variable `MAGMA_DIR` must point to the top-level MAGM…
216 By default, `MAGMA_DIR` is set to `../magma`; to build the MAGMA backends with a MAGMA installation…
219 The corresponding set of libCEED backends (`/gpu/cuda/magma/*` or `/gpu/hip/magma/*`) will automati…
221 Users can specify a device for all CUDA, HIP, and MAGMA backends through adding `:device_id=#` afte…
227 However, some libCEED backends use non-deterministic operations, such as `atomicAdd` for increased …
228 The backends which are capable of generating reproducible results, with the proper compilation opti…
322 The above code assumes a GPU-capable machine with the CUDA backends enabled.
323 Depending on the available backends, other CEED resource specifiers can be provided with the `-ceed…
330 A sequence of benchmarks for all enabled backends can be run using: