Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

T24 — Partitioned-inverse fused solve (`solve_alg = "algo1"`, issue #82 solve half)

Open
  • Is something wrong?
#24 2 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

  • #120 by @github-actions — closed without merging

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
35/100
Issue type
Feature
Clarity
Needs clarification
Activity status
Active
Tech stack
julia
Domain
backend

Research direction

Read the T24 section in TASKS.md together with the named PLAN.md sections and previous-task Report blocks; the issue identifies nd_partition_tree and user_nd_partition_tree as the relevant entry points. Run every test listed by T24 on the KA CPU and CUDA/AMDGPU runners, then verify the exported tree imports with the same supernode partition and nnz_L and that Run tests and Aqua pass.

Written by the indexing model from the issue text.

Description

on-hold task

Task T24 of TASKS.md — milestone M11 — Performance. (Renamed on 2026-10-09: the ND partition-tree task moved to T30, see STATE.md §8.)

Depends on: #23 (T23) must be merged first; tasks are strictly sequential.

Summary (from TASKS.md): adopt the PR #107 solve prototype (bench/solve_proto_*.jl): per-front L₁₁ inverses turning every TRSV into a GEMV, one fused dependency-counter kernel per direction for fronts of width ≤ 256 (atomic-free variant kept), strategy chosen per schedule (it regresses on lap3d_40 and apache2), Cholesky and LDLᵀ, nrhs > 1; every solve suite repeated under "algo1" with a residual gate; Report: solve table vs the cuDSS baseline. Refs #82.

Scope and deliverables: the section T24 of TASKS.md on main is the single source of truth; it is not copied here so it cannot drift.

Definition of done:

  • Every test listed in the task passes on the KA CPU backend and on the CUDA self-hosted runner (Run tests and Aqua green on the PR).
  • The task's status marker in TASKS.md is set and its Report block is filled in honestly (CUDA counts come from CI).
  • A PR titled T24: … from branch task/T24-<slug> with Closes #24 in its body, reviewed by the Claude reviewer with VERDICT: APPROVE.
  • Out-of-scope bugs or follow-ups found on the way are opened as separate issues (label found-by-agent), not fixed silently.

How it starts: add the label claude:implement (the pipeline does this automatically after the previous task's PR merges; label on-hold to prevent that).

Dominant language
Julia
Stars
3
Forks
0
Avg merge
3h 49m
Merged PRs (30d)
68

Getting set up

  • No Dockerfile or Docker Compose file
  • Has a pull request template
  • No contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from exanauts/SparseDirectSolver.jl

All issues in exanauts/SparseDirectSolver.jl

Similar issues

More Julia issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.