Refresh AutomaticDifferentiation under v7 stack#1571
Open
ChrisRackauckas-Claude wants to merge 1 commit into
Open
Refresh AutomaticDifferentiation under v7 stack#1571ChrisRackauckas-Claude wants to merge 1 commit into
ChrisRackauckas-Claude wants to merge 1 commit into
Conversation
This was referenced May 20, 2026
83b87bb to
7f04b99
Compare
ChrisRackauckas-Claude
pushed a commit
to ChrisRackauckas-Claude/SciMLBenchmarks.jl
that referenced
this pull request
May 22, 2026
Pkg.update on benchmarks/ModelingToolkit refreshes the Manifest under the OrdinaryDiffEq v7 / SciMLBase v3 / ModelingToolkit 11 stack. Notable bumps: - SymbolicUtils 4.26.0 -> 4.30.1 (clears the precompile method-overwrite bug that PR SciML#1571 also hit on the AutomaticDifferentiation refresh). - ModelingToolkit 11.24.1 -> 11.26.4, ModelingToolkitBase 1.33.1 -> 1.36.3, Symbolics 7.21.0 -> 7.24.2. - SciMLBase 3.7.1 -> 3.13.0, LinearSolve 3.75.0 -> 3.80.0, SciMLLogging 1.9.1 -> 2.0.0. - OrdinaryDiffEq subpackages (Core, Default, BDF, Rosenbrock, SDIRK, Tsit5, Verner, Differentiation) bumped to current point releases on the v2/v3/v4 lines. No .jmd source changes needed: static scan found no surviving v7-era patterns (no bool autodiff/verbose, no top-level precs, no sol[i] timestep indexing, no old prob_func signature). Local Pkg.instantiate clean; Pkg.update produced the expected Manifest diff and precompile of the refreshed graph succeeded (the residual XSteam method-overwrite warning is the upstream XSteam bug that the ThermalFluid.jmd already works around by pinning the fixed rev at script-runtime). The folder's prior CI red (job 75199969458, run 25374549824) was a runner-workspace corruption on amdci1-1 ("Can't find action.yml under _actions/actions/checkout/v6", "No such file or directory" for bash working directory), not a real benchmark failure. Co-Authored-By: Chris Rackauckas <accounts@chrisrackauckas.com>
Run `Pkg.update()` against the v7 stack baseline. Notable bumps: SymbolicUtils 4.26.0 → 4.27.0 (fixes precompilation method-overwrite of `apply_optimization_rules`), DifferentiationInterface 0.7.17 → 0.7.18, Enzyme 0.13.140 → 0.13.147, Mooncake 0.5.27 → 0.5.28, OrdinaryDiffEqRosenbrock 2.0.0 → 2.2.0, SciMLSensitivity 7.106.0 → 7.109.0, ModelingToolkit 11.24.1 → 11.26.3, LinearSolve 3.75.0 → 3.80.0, plus assorted transitive bumps. Local smoke: instantiated under Julia 1.11.9, all dependencies precompile cleanly, and `weave_file(:script)` tangles all three jmd files (BrussScaling, JuliaAD, SimpleODEAD) without error. Project.toml [compat] is unchanged. Co-Authored-By: Chris Rackauckas <accounts@chrisrackauckas.com>
ChrisRackauckas-Claude
pushed a commit
to ChrisRackauckas-Claude/SciMLBenchmarks.jl
that referenced
this pull request
May 31, 2026
Pkg.update on benchmarks/ModelingToolkit refreshes the Manifest under the OrdinaryDiffEq v7 / SciMLBase v3 / ModelingToolkit 11 stack. Notable bumps: - SymbolicUtils 4.26.0 -> 4.30.1 (clears the precompile method-overwrite bug that PR SciML#1571 also hit on the AutomaticDifferentiation refresh). - ModelingToolkit 11.24.1 -> 11.26.4, ModelingToolkitBase 1.33.1 -> 1.36.3, Symbolics 7.21.0 -> 7.24.2. - SciMLBase 3.7.1 -> 3.13.0, LinearSolve 3.75.0 -> 3.80.0, SciMLLogging 1.9.1 -> 2.0.0. - OrdinaryDiffEq subpackages (Core, Default, BDF, Rosenbrock, SDIRK, Tsit5, Verner, Differentiation) bumped to current point releases on the v2/v3/v4 lines. No .jmd source changes needed: static scan found no surviving v7-era patterns (no bool autodiff/verbose, no top-level precs, no sol[i] timestep indexing, no old prob_func signature). Local Pkg.instantiate clean; Pkg.update produced the expected Manifest diff and precompile of the refreshed graph succeeded (the residual XSteam method-overwrite warning is the upstream XSteam bug that the ThermalFluid.jmd already works around by pinning the fixed rev at script-runtime). The folder's prior CI red (job 75199969458, run 25374549824) was a runner-workspace corruption on amdci1-1 ("Can't find action.yml under _actions/actions/checkout/v6", "No such file or directory" for bash working directory), not a real benchmark failure. Co-Authored-By: Chris Rackauckas <accounts@chrisrackauckas.com>
7f04b99 to
a643f4b
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
benchmarks/AutomaticDifferentiation/Manifest.tomlviaPkg.update()against the v7 stack baseline (post-Update benchmarks to OrdinaryDiffEq v7 / SciMLBase v3 / RecursiveArrayTools v4 stack #1551).Project.tomland[compat]entries are unchanged; this is a pure manifest bump.apply_optimization_rulesmethod-overwrite error that broke loading of ParameterizedFunctions / SymbolicUtils-dependent benchmarks under 4.26.0.Local smoke test
julia +1.11 --project=benchmarks/AutomaticDifferentiation -e 'using Pkg; Pkg.update()'— clean, all 107 newly-built dependencies precompile (with the SymbolicUtils 4.27.0 bump that was the actual root cause of the precompile spam under the old manifest).julia +1.11 --threads=auto --project=benchmarks/AutomaticDifferentiation -e 'using SciMLBenchmarks; for f in readdir(\"benchmarks/AutomaticDifferentiation\"); endswith(f, \".jmd\") && SciMLBenchmarks.weave_file(\"benchmarks/AutomaticDifferentiation\", f, (:script,)); end'— all three.jmdfiles (BrussScaling, JuliaAD, SimpleODEAD) tangle cleanly to.jlscripts.autodiff=true/false,verbose=true/false,precs=,sol.minimum,u_modified!).25912449328against the v7 stack, so the actual benchmark execution is known-good — this PR just refreshes the pin set.Test plan
Benchmarksjob forbenchmarks/AutomaticDifferentiationto go green on amdci1-1 / equivalent.Please ignore until reviewed by @ChrisRackauckas.