GithubHelp home page GithubHelp logo

ecrc / h2opus Goto Github PK

View Code? Open in Web Editor NEW
11.0 11.0 2.0 7.92 MB

H2Opus: a performance-oriented library for hierarchical matrices

License: BSD 3-Clause "New" or "Revised" License

Makefile 0.78% C 5.99% Cuda 13.19% C++ 79.56% Python 0.47%

h2opus's People

Contributors

stefanozampini avatar wajihboukaram avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar

h2opus's Issues

HMatrix directly created and filled on the GPU ?

Hello,

Currently I would like to use H2Opus to accelerate my home made 3D electrostatic BEM solver (Boundary Element Method) which runs on the GPU. The N cells data used to build an NxN non symmetric dense matrix are stored in the GPU memory and the matrix elements are thus computed on the GPU. I would like to know if H2Opus has a method to construct a non symmetric HMatrix directly on the GPU.

Thanks a lot for any answer/clue and for this very promising library.

Example compilation missing functions

Hi,

Up to now I managed to compile H2Opus on a laptop running Ubuntu 22.04 LTS with Cuda 11.5 and with a Quadro RTX 5000 (16 GB). However, when trying to compile the examples, there are some missing functions or other bugs.

For eig:
/opt/nvidia/hpc_sdk/Linux_x86_64/22.5/cuda//bin/nvcc -I/home/quemener/local/h2opus/include -I/opt/intel/oneapi/mkl/include -I/include -I/home/quemener/local/kblas-gpu/include -I/home/quemener/local/magma-2.6.0/include -I//include -I/include -I/home/quemener/install/include -I/home/quemener/PETSc/petsc//include -I/home/quemener/PETSc/petsc/include -x cu -m64 -O3 -std=c++14 --expt-relaxed-constexpr -ccbin g++ -gencode arch=compute_75,code=sm_75 -Xcompiler -Wall -Xcompiler -fPIC -Xcompiler -fopenmp -Xcompiler -O3 -Xcompiler -std=c++14 -Xcompiler -pthread -Xcompiler -DH2OPUS_PROFILING_ENABLED -o obj/test_eig_slepc.o -c test_eig_slepc.cpp

test_eig_slepc.cpp(64): error: identifier "MatCreateH2OpusFromKernel" is undefined
test_eig_slepc.cpp(72): error: identifier "MatH2OpusSetNativeMult" is undefined

For hara:
/opt/nvidia/hpc_sdk/Linux_x86_64/22.5/cuda//bin/nvcc -I/home/quemener/local/h2opus/include -I/opt/intel/oneapi/mkl/include -I/include -I/home/quemener/local/kblas-gpu/include -I/home/quemener/local/magma-2.6.0/include -x cu -m64 -O3 -std=c++14 --expt-relaxed-constexpr -ccbin g++ -gencode arch=compute_75,code=sm_75 -Xcompiler -Wall -Xcompiler -fPIC -Xcompiler -fopenmp -Xcompiler -O3 -Xcompiler -std=c++14 -Xcompiler -pthread -Xcompiler -DH2OPUS_PROFILING_ENABLED -o obj/test_hara.o -c test_hara.cpp
test_hara.cpp(189): error: identifier "ReconstructSampler" is undefined

test_hara.cpp(189): error: identifier "gpu_reconstruct_sampler" is undefined

For spatialstatistics:
spatial_statistics.cpp(73): error: identifier "MatCreateH2OpusFromKernel" is undefined
spatial_statistics.cpp(95): error: identifier "MatH2OpusCompress" is undefined
spatial_statistics.cpp(125): error: identifier "MatH2OpusSetNativeMult" is undefined

For tlr:
$ test_tlr
Tile Size = 512. ARA Block size = 32
Construction threshold = 1.000000e-06. Cholesky threshold = 1.000000e-06. Diagonal shift s = 0.000000e+00
Erreur de segmentation (core dumped)

Where could I get the missing functions which are not included in the repository files ?

Thanks a lot for any help,

Gilles

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.