Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[request] suitesparse/5.10.1 #5763

Open
SpaceIm opened this issue Jun 4, 2021 · 3 comments
Open

[request] suitesparse/5.10.1 #5763

SpaceIm opened this issue Jun 4, 2021 · 3 comments

Comments

@SpaceIm
Copy link
Contributor

SpaceIm commented Jun 4, 2021

Package Details

Description Of The Library / Tool

SuiteSparse is a suite of sparse matrix algorithms.

@ericriff
Copy link
Contributor

ericriff commented Jun 15, 2021

Related:

cmake.definitions["SUITESPARSE"] = False #Optional. Not supported right now because SuiteSparse is not part of conan-index

@sxsong1207
Copy link

Yeah, that would be helpful. Without SUITESPARSE, the covariance computation of ceres-solver is extremely slow.

ceres-solver/ceres-solver#766 (comment)

@valgur
Copy link
Contributor

valgur commented Apr 15, 2024

SuiteSparse (v7.7.0) is ready for packaging, now that we have a working OpenBLAS with BLAS and LAPACK support and an almost fully-functional OpenMP support (#22353).

As the name suggests, SuiteSparse consists of a suite of mostly independent libraries, where each of them:

  • has a distinct versioning scheme,
  • varying license terms (e.g. some are GPL, while others are more permissive),
  • installs a separate CMake config file (i.e. you would do find_package(AMD) to use suitesparse-amd),
  • can be implemented in pure C, while some use C++ as well,
  • possibly has library-specific options, e.g. for CUDA support.

So it makes more sense to package each of the sub-components separately, despite it requiring the creation of a considerable number of new packages and related PRs. On the upside, this also allows only specific components to be used as needed, with binaries available and without having to build everything (such as suitesparse-graphblas, which can take up to an hour to build with pre-built kernels enabled).

Here's the list of PRs for each of the sub-package:

I did not create a suitesparse meta-package currently, but it could be added later.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants