Skip to content

Bump versions: NonlinearSolveBase 2.55.1 - #1378

Draft
ChrisRackauckas wants to merge 2 commits into
masterfrom
bump-NonlinearSolveBase-2.55.1
Draft

ChrisRackauckas wants to merge 2 commits into
masterfrom
bump-NonlinearSolveBase-2.55.1

Conversation

@ChrisRackauckas

@ChrisRackauckas ChrisRackauckas commented Oct 16, 2026 •

Copy link
Copy Markdown
Member

🚧 UNREVIEWED — awaiting review by @ChrisRackauckas. Filed by an AI agent running as @ChrisRackauckas; Chris has not seen this.
Harness: Claude Code 2.1.302 · Model: claude-sonnet-5-5
Conversation: https://claude.ai/code/session_01VZjwf7Ux9YtTzwHGHMqnsq

NonlinearSolveBase 2.55.0 → 2.55.1 (lib/NonlinearSolveBase)

Bumps version in lib/NonlinearSolveBase/Project.toml. Merged since v2.55.0 (the latest version in the General registry):

Compare: 5476c46...e903b6b

Patch: DampedNewtonDescent forms the sparse J'J in place (#1363); no API change.

Risk assessment

  • Risk: low
  • Blast radius: version field only; the release ships the already-merged commits listed above.
  • Evidence: version-line-only diff; default-branch CI on the shipped commit checked for regressions against the last release commit.
  • Independent review: Devin SWE-2 High (devin CLI, read-only; Codex over pace and Grok exhausted, so a non-Claude free model) rated it low, medium confidence, verdict MERGE: Internal performance fix forming sparse JᵀJ in place with allocating fallback; no public API or behavior change, patch is correct for ≥1.0. (review output is local to the fleet master; the justification is quoted here)
  • Merge: auto-merge candidate (version bump, master release duty)

🤖 Posted by an AI agent — harness: Claude Code 2.1.302 · model: claude-sonnet-5-5
Conversation: https://claude.ai/code/session_01VZjwf7Ux9YtTzwHGHMqnsq

🤖 Generated with Claude Code

https://claude.ai/code/session_01VZjwf7Ux9YtTzwHGHMqnsq

🤖 Automated review section from an AI agent running as Chris Rackauckas's account — not written or reviewed by Chris.

Fix by Codex gpt-6-luna (head ec3e5a5)

Merged the current default branch into this branch with a normal merge. It contains the polyalgorithm solve!() inference fix and regression coverage. The failing PR jobs errored in solve!() inference on Linux and macOS; the latest default-branch Core LTS run passes both platforms (PR failure, default branch run).

Validation:

  • NONLINEARSOLVE_TEST_GROUP=Core julia --project -e 'using Pkg; Pkg.test()' — PolyAlgorithm Type Inference | 16 16; final output: Testing NonlinearSolve tests passed (Julia 1.12.4, 4 threads).
  • GROUP=QA julia --project -e 'using Pkg; Pkg.test()' — QA | 28 28; final output: Testing NonlinearSolve tests passed (Julia 1.12.4, 4 threads). The initial QA invocation was interrupted during Aqua setup; the retry completed successfully.
  • Runic --inplace and --check completed successfully on each changed Julia file.
  • git diff HEAD | typos - exited 0 with no spelling findings.

The pushed branch CI had started when checked; multiple workflows were in progress and IntegrationTest was queued. CI was not awaited.

Next counted review must come from a smart model in a fresh session, not this fix session (risk.md § Independent reviews).


🤖 Posted by an AI agent — harness: Codex CLI · model: gpt-6-luna
Conversation: local fixer job SciML__NonlinearSolve.jl__1378 (no session URL)

Co-Authored-By: Chris Rackauckas <accounts@chrisrackauckas.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Agent-Harness: Claude Code 2.1.302
Agent-Model: claude-sonnet-5-5
Claude-Session: https://claude.ai/code/session_01VZjwf7Ux9YtTzwHGHMqnsq
@ChrisRackauckas-Claude

ChrisRackauckas-Claude commented Oct 17, 2026 •

Copy link
Copy Markdown
Member

🤖 Automated comment from an AI agent running as @ChrisRackauckas — not written or reviewed by Chris.

Core LTS classification at 27a76aa4: pre-existing on master, fixed on current master. Both Linux and macOS fail PolyAlgorithm Type Inference → solve!() inference (test/Core/core_tests__item3.jl:27), matching the earlier master failure. The head lacks the merged retcode type assertion in #1347. The first completed master Core LTS jobs after that fix pass on both platforms.

The SCC isconcretetype(eltype(scc_sol_debug.original)) failure is a separate assertion already failing on master; existing work is tracked at #1332.

Links:


🤖 Posted by an AI agent — harness: Codex CLI · model: gpt-6.1-sol
Conversation: local headless job codex-nls-core-lts-20261017 (no session URL)

Co-Authored-By: Chris Rackauckas <accounts@chrisrackauckas.com>
Co-Authored-By: Codex <noreply@openai.com>
Agent-Harness: Codex CLI
Agent-Model: gpt-6-luna
Agent-Session: local fixer job SciML__NonlinearSolve.jl__1378 (no session URL)

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants