Repository navigation
Require the FiniteDiff release that scales the JVP step - #4583
Merged
ChrisRackauckas merged 1 commit intoSep 17, 2026
Merged
ChrisRackauckas merged 1 commit into
ChrisRackauckas merged 1 commit into
Conversation
singhharsh1708
deleted the
differentiation-require-finitediff-jvp-fix
branch
September 23, 2026 12:56
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
OrdinaryDiffEqDifferentiationnow requires FiniteDiff 2.33. Below that the finite difference JVP takes an unscaled step, andRadauIIA3withKrylovJL_GMRESandAutoFiniteDiffstalls until dt falls below floating point epsilon.That is what the red
downgrade-sublibraries (lib/OrdinaryDiffEqFIRK)lane reports on every PR:firk_krylov_tests.jlfails "RadauIIA3 counts matrix-free Jacobian-vector products" at t = 0.1191619.Reproduced locally at the versions that lane resolves: FiniteDiff 2.31.1 and 2.32.0 fail 97 of 98, 2.33.0 passes 98 of 98, with LinearSolve 5.15.0, Krylov 0.10.10, SciMLBase 3.53.0, SciMLOperators 1.30.0 and ForwardDiff 1.3.3 held at the lane's versions. Holding each of those old while FiniteDiff is current keeps the suite green, so FiniteDiff is the one that matters.
FiniteDiff 2.33.0 carries a single change, JuliaDiff/FiniteDiff.jl#230, which scales the JVP step by
norm(x)/norm(v);JVPCachein this package is what calls it.This touches the same
Project.tomlversion line as #4582, so whichever lands second needs the bump re-derived.AI Disclosure: Claude assisted with this change.