ORNL

ORNL/ReSolve

Library of GPU-resident linear solvers

9 good first / help-wanted issues · C++ · last activity Jul 8, 2026

79 stars 9 forks 79 watchers C++ Other

9 Open Issues Need Help Last updated: Jul 8, 2026

Open Issues Need Help

View All on GitHub

Bug: rocm/7.1.1 4 months ago

bug help wanted hip

79

Library of GPU-resident linear solvers

C++

SystemSolver still uses string to select preconditioning side 4 months ago

enhancement good first issue

79

Library of GPU-resident linear solvers

C++

Consider reviewing naming scheme for `VectorHandler` methods 8 months ago

help wanted question

79

Library of GPU-resident linear solvers

C++

Bug: Refactorization fails when used after 0th matrix in the ACTIVs series 9 months ago

bug help wanted question

79

Library of GPU-resident linear solvers

C++

Interface for CFD solver 10 months ago

good first issue iterative solver

79

Library of GPU-resident linear solvers

C++

Create sysGmres example 10 months ago

enhancement good first issue examples iterative solver

79

Library of GPU-resident linear solvers

C++

Bug: MatrixHandler::addConst return type doesn't match documentation 12 months ago

good first issue

79

Library of GPU-resident linear solvers

C++

Incorporate initial guesses into residual for iterative solvers. about 1 year ago

good first issue

79

Library of GPU-resident linear solvers

C++

Improve documentation on how memory spaces are synchronized over 1 year ago

AI Summary: Enhance ReSolve's documentation to clarify the memory synchronization model. This involves creating a comprehensive guide explaining which memory spaces are synchronized automatically and which require explicit synchronization, similar to NVIDIA's CUDA documentation. The goal is to improve user understanding and reduce the need for experimentation.

Complexity: 4/5

documentation help wanted hip cuda

79

Library of GPU-resident linear solvers

C++