1# libCEED: the CEED API Library 2 3[](https://travis-ci.org/CEED/libCEED) 4[](https://codecov.io/gh/CEED/libCEED/) 5[](https://opensource.org/licenses/BSD-2-Clause) 6[](https://codedocs.xyz/CEED/libCEED/) 7 8## Code for Efficient Extensible Discretization 9 10This repository contains an initial low-level API library for the efficient 11high-order discretization methods developed by the ECP co-design [Center for 12Efficient Exascale Discretizations (CEED)](http://ceed.exascaleproject.org). 13While our focus is on high-order finite elements, the approach is mostly 14algebraic and thus applicable to other discretizations in factored form, as 15explained in the API documentation portion of the [Doxygen documentation](https://codedocs.xyz/CEED/libCEED/md_doc_libCEEDapi.html). 16 17One of the challenges with high-order methods is that a global sparse matrix is 18no longer a good representation of a high-order linear operator, both with 19respect to the FLOPs needed for its evaluation, as well as the memory transfer 20needed for a matvec. Thus, high-order methods require a new "format" that still 21represents a linear (or more generally non-linear) operator, but not through a 22sparse matrix. 23 24The goal of libCEED is to propose such a format, as well as supporting 25implementations and data structures, that enable efficient operator evaluation 26on a variety of computational device types (CPUs, GPUs, etc.). This new operator 27description is based on algebraically [factored form](https://codedocs.xyz/CEED/libCEED/md_doc_libCEEDapi.html), 28which is easy to incorporate in a wide variety of applications, without significant 29refactoring of their own discretization infrastructure. 30 31The repository is part of the [CEED software suite][ceed-soft], a collection of 32software benchmarks, miniapps, libraries and APIs for efficient exascale 33discretizations based on high-order finite element and spectral element methods. 34See http://github.com/ceed for more information and source code availability. 35 36The CEED research is supported by the [Exascale Computing Project][ecp] 37(17-SC-20-SC), a collaborative effort of two U.S. Department of Energy 38organizations (Office of Science and the National Nuclear Security 39Administration) responsible for the planning and preparation of a [capable 40exascale ecosystem](https://exascaleproject.org/what-is-exascale), including 41software, applications, hardware, advanced system engineering and early testbed 42platforms, in support of the nation’s exascale computing imperative. 43 44For more details on the CEED API see http://ceed.exascaleproject.org/ceed-code/. 45 46## Building 47 48The CEED library, `libceed`, is a C99 library with no external dependencies. It 49can be built using 50 51 make 52 53or, with optimization flags 54 55 make OPT='-O3 -march=skylake-avx512 -ffp-contract=fast' 56 57These optimization flags are used by all languages (C, C++, Fortran) and this 58makefile variable can also be set for testing and examples (below). 59 60The library attempts to automatically detect support for the AVX 61instruction set using gcc-style compiler options for the host. 62Support may need to be manually specified via 63 64 make AVX=1 65 66or 67 68 make AVX=0 69 70if your compiler does not support gcc-style options, if you are cross 71compiling, etc. 72 73## Testing 74 75The test suite produces [TAP](https://testanything.org) output and is run by: 76 77 make test 78 79or, using the `prove` tool distributed with Perl (recommended) 80 81 make prove 82 83## Backends 84 85There are multiple supported backends, which can be selected at runtime in the examples: 86 87| CEED resource | Backend | 88| :----------------------- | :------------------------------------------------ | 89| `/cpu/self/ref/serial` | Serial reference implementation | 90| `/cpu/self/ref/blocked` | Blocked refrence implementation | 91| `/cpu/self/tmpl` | Backend template, delegates to `/cpu/self/ref/blocked` | 92| `/cpu/self/ref/memcheck` | Memcheck backend, undefined value checks | 93| `/cpu/self/avx/serial` | Serial AVX implementation | 94| `/cpu/self/avx/blocked` | Blocked AVX implementation | 95| `/cpu/self/xsmm/serial` | Serial LIBXSMM implementation | 96| `/cpu/self/xsmm/blocked` | Blocked LIBXSMM implementation | 97| `/cpu/occa` | Serial OCCA kernels | 98| `/gpu/occa` | CUDA OCCA kernels | 99| `/omp/occa` | OpenMP OCCA kernels | 100| `/ocl/occa` | OpenCL OCCA kernels | 101| `/gpu/cuda/ref` | Reference pure CUDA kernels | 102| `/gpu/cuda/reg` | Pure CUDA kernels using one thread per element | 103| `/gpu/magma` | CUDA MAGMA kernels | 104 105 106The `/cpu/self/*/serial` backends process one element at a time and are intended for meshes 107with a smaller number of high order elements. The `/cpu/self/*/blocked` backends process 108blocked batches of eight interlaced elements and are intended for meshes with higher numbers 109of elements. 110 111The `/cpu/self/ref/*` backends are written in pure C and provide basic functionality. 112 113The `/cpu/self/avx/*` backends rely upon AVX instructions to provide vectorized CPU performance. 114 115The `/cpu/self/xsmm/*` backends rely upon the [LIBXSMM](http://github.com/hfp/libxsmm) package 116to provide vectorized CPU performance. The LIBXSMM backend does not use BLAS or MKL; however, 117if LIBXSMM was linked to MKL, this can be specified with the compilation flag `MKL=1`. 118 119The `/cpu/self/ref/memcheck` backend relies upon the [Valgrind](http://valgrind.org/) Memcheck tool 120to help verify that user QFunctions have no undefined values. To use, run your code with 121Valgrind and the Memcheck backend, e.g. `valgrind ./build/ex1 -ceed /cpu/self/ref/memcheck`. 122 123The `/*/occa` backends rely upon the [OCCA](http://github.com/libocca/occa) package to provide 124cross platform performance. 125 126The `/gpu/cuda/*` backends provide GPU performance strictly using CUDA. 127 128The `/gpu/magma` backend relies upon the [MAGMA](https://bitbucket.org/icl/magma) package. 129 130## Examples 131 132libCEED comes with several examples of its usage, ranging from standalone C 133codes in the `/examples/ceed` directory to examples based on external packages, 134such as MFEM, PETSc, and Nek5000. Nek5000 v18.0 or greater is required. 135 136To build the examples, set the `MFEM_DIR`, `PETSC_DIR` and `NEK5K_DIR` variables 137and run: 138 139```console 140# libCEED examples on CPU and GPU 141cd examples/ceed 142make 143./ex1 -ceed /cpu/self 144./ex1 -ceed /gpu/occa 145cd ../.. 146 147# MFEM+libCEED examples on CPU and GPU 148cd examples/mfem 149make 150./bp1 -ceed /cpu/self -no-vis 151./bp1 -ceed /gpu/occa -no-vis 152cd ../.. 153 154# PETSc+libCEED examples on CPU and GPU 155cd examples/petsc 156make 157./bp1 -ceed /cpu/self 158./bp1 -ceed /gpu/occa 159cd ../.. 160 161cd navier-stokes 162make 163./navierstokes -ceed /cpu/self 164./navierstokes -ceed /gpu/occa 165cd ../.. 166 167# Nek+libCEED examples on CPU and GPU 168cd examples/nek5000 169./make-nek-examples.sh 170./run-nek-example.sh -ceed /cpu/self -b 3 171./run-nek-example.sh -ceed /gpu/occa -b 3 172cd ../.. 173``` 174 175The above code assumes a GPU-capable machine with the OCCA backend 176enabled. Depending on the available backends, other Ceed resource specifiers can 177be provided with the `-ceed` option. 178 179## Benchmarks 180 181A sequence of benchmarks for all enabled backends can be run using 182 183```console 184make benchmarks 185``` 186 187The results from the benchmarks are stored inside the `benchmarks/` directory 188and they can be viewed using the commands (requires python with matplotlib): 189 190```console 191cd benchmarks 192python postprocess-plot.py petsc-bp1-*-output.txt 193python postprocess-plot.py petsc-bp3-*-output.txt 194``` 195 196Using the `benchmarks` target runs a comprehensive set of benchmarks which may 197take some time to run. Subsets of the benchmarks can be run using targets such 198as `make bench-petsc-bp1`, or `make bench-petsc-bp3`. 199 200For more details about the benchmarks, see 201[`benchmarks/README.md`](benchmarks/README.md) 202 203 204## Install 205 206To install libCEED, run 207 208 make install prefix=/usr/local 209 210or (e.g., if creating packages), 211 212 make install prefix=/usr DESTDIR=/packaging/path 213 214Note that along with the library, libCEED installs kernel sources, e.g. OCCA 215kernels are installed in `$prefix/lib/okl`. This allows the OCCA backend to 216build specialized kernels at run-time. In a normal setting, the kernel sources 217will be found automatically (relative to the library file `libceed.so`). 218However, if that fails (e.g. if `libceed.so` is moved), one can copy (cache) the 219kernel sources inside the user OCCA directory, `~/.occa` using 220 221 $(OCCA_DIR)/bin/occa cache ceed $(CEED_DIR)/lib/okl/*.okl 222 223This will allow OCCA to find the sources regardless of the location of the CEED 224library. One may occasionally need to clear the OCCA cache, which can be accomplished 225by removing the `~/.occa` directory or by calling `$(OCCA_DIR)/bin/occa clear -a`. 226 227### pkg-config 228 229In addition to library and header, libCEED provides a [pkg-config][pkg-config1] 230file that can be used to easily compile and link. [For example][pkg-config2], if 231`$prefix` is a standard location or you set the environment variable 232`PKG_CONFIG_PATH`, 233 234 cc `pkg-config --cflags --libs ceed` -o myapp myapp.c 235 236will build `myapp` with libCEED. This can be used with the source or 237installed directories. Most build systems have support for pkg-config. 238 239## Contact 240 241You can reach the libCEED team by emailing [ceed-users@llnl.gov](mailto:ceed-users@llnl.gov) 242or by leaving a comment in the [issue tracker](https://github.com/CEED/libCEED/issues). 243 244## Copyright 245 246The following copyright applies to each file in the CEED software suite, unless 247otherwise stated in the file: 248 249> Copyright (c) 2017, Lawrence Livermore National Security, LLC. Produced at the 250> Lawrence Livermore National Laboratory. LLNL-CODE-734707. All Rights reserved. 251 252See files LICENSE and NOTICE for details. 253 254[ceed-soft]: http://ceed.exascaleproject.org/software/ 255[ecp]: https://exascaleproject.org/exascale-computing-project 256[pkg-config1]: https://en.wikipedia.org/wiki/Pkg-config 257[pkg-config2]: https://people.freedesktop.org/~dbn/pkg-config-guide.html#faq 258
){kind=link}
){kind=link}
){kind=link}
){kind=link}