Fix GPU Higham2005 LinearSolve fallback#246
Merged
ChrisRackauckas merged 1 commit intoJul 12, 2026
Merged
Conversation
Co-Authored-By: Chris Rackauckas <accounts@chrisrackauckas.com>
ChrisRackauckas
marked this pull request as ready for review
July 12, 2026 04:45
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR should be ignored until reviewed by @ChrisRackauckas.
This is separate from PR #245.
Root cause
PR #244 moved the generated ExpMethodHigham2005 Padé denominator solve through a cached LinearSolve workspace. The code intended GPU arrays to keep the existing lu! fallback, but the GPU CI stack shows CuArray entering LinearSolve.DefaultLinearSolver and then QRFactorization, failing in cuSOLVER ldiv! with:
The failing stack is from test/gpu/gputests.jl:22 through src/exp_noalloc.jl:64 and LinearSolve factorization.jl.
Evidence
Change
Make _pade_linsolve return nothing for GPUArraysCore.AbstractGPUArray values even when the array also satisfies the strided BLAS matrix method, forcing the generated code back onto its intended GPU lu! fallback.
Validation
Local GPU verification was not possible because this machine has no /dev/nvidia* device; the pushed draft PR is intended to let the self-hosted GPU job verify the fix.