exanauts/SparseDirectSolver.jl

View on GitHub

Portable sparse direct linear solver for GPUs in Julia (KernelAbstractions.jl), mirroring the CUDSS.jl feature set

Is exanauts/SparseDirectSolver.jl beginner-friendly?

The maintainers of exanauts/SparseDirectSolver.jl usually reply to a pull request within 1 day. No beginner-friendly issue is open right now; the list below has the other issues we hold for it.

Stars: 3
Forks: 0
Open beginner issues: 0
Indexed issues: 31
Avg merge: 3h 2m
Merged PRs (30d): 65

Dominant language: Julia
License: MIT
Last GitHub push: Oct 9, 2026
Latest indexed: Oct 4, 2026
Contributing guide: No contributing guide
Code of conduct: No code of conduct
Beginner labels: No beginner labels indexed

How to contribute to exanauts/SparseDirectSolver.jl

There is no contributing guide, so start from the README and a few recently merged pull requests to see how changes are proposed.
Your contributions will be published under the project's MIT license.
Follow the repository to hear when a new beginner-friendly issue opens, or look through the other issues below.

Beginner friendly Everything

Open All

Issues somebody may already be working on are listed last. List everything by date

T31 — oneAPI and Metal extensions, allocation-free KA dense fallbacks (issues #112, #53)
Open
- Is something wrong?
task

Difficulty 5/5 Over a week Newbie friendliness 12/100

exanauts/SparseDirectSolver.jl#118 · Opened Oct 9, 2026

Maintainers usually reply within 1 day
T30 — ND partition-tree export/import and ordering cache
Open
- Is something wrong?
task

Difficulty 4/5 3-5 days Newbie friendliness 20/100

exanauts/SparseDirectSolver.jl#117 · Opened Oct 9, 2026

Maintainers usually reply within 1 day
T29 — Non-uniform batch (`BatchedDirectSolver`)
Open
- Is something wrong?
task

Difficulty 5/5 Over a week Newbie friendliness 8/100

exanauts/SparseDirectSolver.jl#116 · Opened Oct 9, 2026

Maintainers usually reply within 1 day
T28 — Robustness extras
Open
- Is something wrong?
task

Difficulty 5/5 Over a week Newbie friendliness 8/100

exanauts/SparseDirectSolver.jl#115 · Opened Oct 9, 2026

Maintainers usually reply within 1 day
T23 remainder: oneAPI and Metal extensions, allocation-free KA dense fallbacks (#53)
Open
- Is something wrong?
found-by-agent triaged

Difficulty 5/5 Over a week Newbie friendliness 12/100

exanauts/SparseDirectSolver.jl#112 · 1 comment · Opened Oct 9, 2026

Maintainers usually reply within 1 day
Regime-C Cholesky reportedly host-synchronizes per front through its info check (blocks stream overlap of wide fronts); verify
Open
- Is something wrong?
found-by-agent performance triaged

Difficulty 4/5 3-5 days Newbie friendliness 22/100

exanauts/SparseDirectSolver.jl#110 · 1 comment · Opened Oct 9, 2026

Maintainers usually reply within 1 day
Ordering chooser picks AMD on large KKT systems: cost model weighs column-etree depth, not supernodal schedule depth (2× schedule depth vs ND)
Open
- Is something wrong?
found-by-agent performance triaged

Difficulty 4/5 3-5 days Newbie friendliness 15/100

exanauts/SparseDirectSolver.jl#108 · 1 comment · Opened Oct 9, 2026

Maintainers usually reply within 1 day
Update-stack placement assumes level-synchronous execution: out-of-level-order kernels (experiment 5) need private contribution blocks or DAG lifetimes
Open
- Is something wrong?
found-by-agent performance triaged

Difficulty 5/5 Over a week Newbie friendliness 8/100

exanauts/SparseDirectSolver.jl#109 · 1 comment · Opened Oct 9, 2026

Maintainers usually reply within 1 day
Matching-based 2×2 pairs (algo5) cost ~2.1× nnz(L) on MadNLP K2 systems
Open
- Is something wrong?
found-by-agent triaged

Difficulty 4/5 3-5 days Newbie friendliness 45/100

exanauts/SparseDirectSolver.jl#96 · 2 comments · Opened Oct 7, 2026

Maintainers usually reply within 1 day
Compile time: fewer kernel specialisations per (T, INT)
Open
- Is something wrong?
performance triaged

Difficulty 5/5 Over a week Newbie friendliness 35/100

exanauts/SparseDirectSolver.jl#91 · 1 comment · Opened Oct 6, 2026

Maintainers usually reply within 1 day
Device LDLᵀ differs from ref_ldlt! on the K2 dumps (pivot sequence, nperturbed; inertia equal)
Open
- Is something wrong?
found-by-agent triaged

Difficulty 4/5 3-5 days Newbie friendliness 25/100

exanauts/SparseDirectSolver.jl#86 · 1 comment · Opened Oct 5, 2026

Maintainers usually reply within 1 day
KKT refactorization and solve are level-bound after the regime-A split: 54–59 launches per refactorization, 62–85 per solve, regime-B fronts on one or two blocks
Open
- Is something wrong?
found-by-agent performance triaged

Difficulty 5/5 Over a week Newbie friendliness 35/100

exanauts/SparseDirectSolver.jl#82 · Opened Oct 5, 2026

Maintainers usually reply within 1 day
Regime-A follow-ups: owed CUDA timings and per-backend local-memory caps
Open
- Is something wrong?
found-by-agent performance triaged

Difficulty 4/5 3-5 days Newbie friendliness 38/100

exanauts/SparseDirectSolver.jl#60 · Opened Oct 2, 2026

Maintainers usually reply within 1 day
Dense :ka/:generic paths and the KA CPU backend allocate in the numeric phase (Julia 1.10, coverage)
Open
- Is something wrong?
found-by-agent triaged

Difficulty 4/5 3-5 days Newbie friendliness 52/100

exanauts/SparseDirectSolver.jl#53 · 2 comments · Opened Oct 2, 2026

Maintainers usually reply within 1 day
External — MadNLPGPU integration (in the MadNLP repository)
Open
- Is something wrong?
external task

Difficulty 4/5 3-5 days Newbie friendliness 35/100

exanauts/SparseDirectSolver.jl#28 · Opened Oct 1, 2026

Maintainers usually reply within 1 day
T27 — Hybrid memory / hybrid execute
Open
- Is something wrong?
task

Difficulty 5/5 Over a week Newbie friendliness 25/100

exanauts/SparseDirectSolver.jl#27 · Opened Oct 1, 2026

Maintainers usually reply within 1 day
T26 — Segmented fused factorization with dependency counters (issues #82, #109, #110)
Open
- Is something wrong?
task

Difficulty 5/5 Over a week Newbie friendliness 25/100

exanauts/SparseDirectSolver.jl#26 · Opened Oct 1, 2026

Maintainers usually reply within 1 day
T25 — Split factorization: tiled SYRK, chunked TRSM, split regime-C fronts
Open
- Is something wrong?
task

Difficulty 5/5 Over a week Newbie friendliness 20/100

exanauts/SparseDirectSolver.jl#25 · Opened Oct 1, 2026

Maintainers usually reply within 1 day
T24 — Partitioned-inverse fused solve (`solve_alg = "algo1"`, issue #82 solve half)
Open
- Is something wrong?
on-hold task

Difficulty 4/5 3-5 days Newbie friendliness 35/100

exanauts/SparseDirectSolver.jl#24 · 2 comments · Opened Oct 1, 2026

Maintainers usually reply within 1 day

exanauts/SparseDirectSolver.jl

Is exanauts/SparseDirectSolver.jl beginner-friendly?

How to contribute to exanauts/SparseDirectSolver.jl

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Get new issues in your inbox