Fix CI docs build #822
Open
Fix CI docs build #822
+15
−10
Conversation
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
Kept throwing errors when building docs
Files generated when running docs/make.jl
Created during doc building
Contributor
Author
|
The BVP Ascher thing fails on LinearSolve v3.46.0, but works with 3.45.0. MWE script: using Pkg
Pkg.activate(temp=true)
ENV["JULIA_PKG_PRECOMPILE_AUTO"] = false
Pkg.add(name="LinearSolve", version=ARGS[1])
Pkg.add("BoundaryValueDiffEq")
import BoundaryValueDiffEq as BVP
Pkg.status()
function f!(du, u, p, t)
e = 2.7
du[1] = (1 + u[2] - sin(t)) * u[4] + cos(t)
du[2] = cos(t)
du[3] = u[4]
du[4] = (u[1] - sin(t)) * (u[4] - e^t)
end
function bc!(res, u, p, t)
res[1] = u[1]
res[2] = u[3] - 1
res[3] = u[2] - sin(1.0)
end
u0 = [0.0, 0.0, 0.0, 0.0]
tspan = (0.0, 1.0)
fun = BVP.BVPFunction(f!, bc!, mass_matrix = [1 0 0 0; 0 1 0 0; 0 0 1 0; 0 0 0 0])
prob = BVP.BVProblem(fun, u0, tspan)
sol = BVP.solve(prob,
BVP.Ascher4(; zeta = [0.0, 0.0, 1.0], jac_alg = BVP.BVPJacobianAlgorithm(BVP.AutoForwardDiff()));
dt = 0.01)run as: |
Contributor
Author
|
The problems from DiffEqProblemLibrary.jl appear to be the older ones. I think a new version is needed for (ODEProblemLibrary, SDEProblemLibrary, DDEProblemLibrary, DAEProblemLibrary, BVProblemLibrary) |
Member
|
What's the error? We should make the docs build fail on error |
Contributor
Author
|
Running with 3.46 gives $ julia mwe.jl 3.46.0
Activating new project at `/tmp/jl_eW8Fpj`
Resolving package versions...
Installed LinearSolve ─ v3.46.0
Updating `/tmp/jl_eW8Fpj/Project.toml`
⌃ [7ed4a6bd] + LinearSolve v3.46.0
Updating `/tmp/jl_eW8Fpj/Manifest.toml`
[... Pkg output ...]
Resolving package versions...
Updating `/tmp/jl_eW8Fpj/Project.toml`
[764a87c0] + BoundaryValueDiffEq v5.18.0
Updating `/tmp/jl_eW8Fpj/Manifest.toml`
[... Pkg output ...]
Precompiling BoundaryValueDiffEq finished.
15 dependencies successfully precompiled in 425 seconds. 198 already precompiled.
Status `/tmp/jl_eW8Fpj/Project.toml`
[764a87c0] BoundaryValueDiffEq v5.18.0
⌃ [7ed4a6bd] LinearSolve v3.46.0
Info Packages marked with ⌃ have new versions available and may be upgradable.
┌ Error: BLAS/LAPACK dgetrf: Matrix is singular
│ Details: U(1,1) is exactly zero. The factorization has been completed, but U is singular and division by U will produce infinity.
│ Return code (info): 1
│ matrix_size: (303, 303)
│ rhs_type: Vector{Float64}
│ rhs_size: (303,)
│ memory_usage_MB: 0.7
│ matrix_type: Matrix{Float64}
│ element_type: Float64
└ @ LinearSolve ~/.julia/packages/LinearSolve/V1Dfx/src/mkl.jl:267
ERROR: LoadError: BLAS/LAPACK dgetrf: Matrix is singular
Details: U(1,1) is exactly zero. The factorization has been completed, but U is singular and division by U will produce infinity.
Return code (info): 1
matrix_size: (303, 303)
rhs_type: Vector{Float64}
rhs_size: (303,)
memory_usage_MB: 0.7
matrix_type: Matrix{Float64}
element_type: Float64
Stacktrace:
[1] emit_message(message::String, level::Base.CoreLogging.LogLevel, file::String, line::Int64, _module::Module)
@ SciMLLogging ~/.julia/packages/SciMLLogging/3uoc4/src/utils.jl:57
[2] macro expansion
@ ~/.julia/packages/SciMLLogging/3uoc4/src/utils.jl:147 [inlined]
[3] solve!(cache::LinearSolve.LinearCache{Matrix{Float64}, Vector{Float64}, Vector{Float64}, SciMLBase.NullParameters, LinearSolve.DefaultLinearSolver, [types elided]...)
@ LinearSolve ~/.julia/packages/LinearSolve/V1Dfx/src/mkl.jl:267
[4] solve!
@ ~/.julia/packages/LinearSolve/V1Dfx/src/mkl.jl:239 [inlined]
[5] macro expansion
@ ~/.julia/packages/LinearSolve/V1Dfx/src/default.jl:514 [inlined]
[6] solve!(::LinearSolve.LinearCache{Matrix{Float64}, Vector{Float64}, Vector{Float64}, SciMLBase.NullParameters, LinearSolve.DefaultLinearSolver, [types elided]...)
@ LinearSolve ~/.julia/packages/LinearSolve/V1Dfx/src/default.jl:503
[7] solve!
@ ~/.julia/packages/LinearSolve/V1Dfx/src/default.jl:503 [inlined]
[8] #solve!#21
@ ~/.julia/packages/LinearSolve/V1Dfx/src/common.jl:441 [inlined]
[9] solve!
@ ~/.julia/packages/LinearSolve/V1Dfx/src/common.jl:440 [inlined]
[10] #_#1
@ ~/.julia/packages/NonlinearSolveBase/mJsVc/ext/NonlinearSolveBaseLinearSolveExt.jl:24 [inlined]
[11] LinearSolveJLCache
@ ~/.julia/packages/NonlinearSolveBase/mJsVc/ext/NonlinearSolveBaseLinearSolveExt.jl:14 [inlined]
[12] solve!(cache::NonlinearSolveBase.NewtonDescentCache{Vector{Float64}, Nothing, [types elided]...)
@ NonlinearSolveBase ~/.julia/packages/NonlinearSolveBase/mJsVc/src/descent/newton.jl:119
[13] solve! (repeats 2 times)
@ ~/.julia/packages/NonlinearSolveBase/mJsVc/src/descent/newton.jl:94 [inlined]
[14] step!(cache::NonlinearSolveFirstOrder.GeneralizedFirstOrderAlgorithmCache{Vector{Float64}, Vector{Float64}, Vector{Float64}, SciMLBase.NullParameters, [types elided]...)
@ NonlinearSolveFirstOrder ~/.julia/packages/NonlinearSolveFirstOrder/inFag/src/solve.jl:266
[15] step!
@ ~/.julia/packages/NonlinearSolveFirstOrder/inFag/src/solve.jl:244 [inlined]
[16] #step!#180
@ ~/.julia/packages/NonlinearSolveBase/mJsVc/src/solve.jl:559 [inlined]
[17] step!
@ ~/.julia/packages/NonlinearSolveBase/mJsVc/src/solve.jl:553 [inlined]
[18] solve!(cache::NonlinearSolveFirstOrder.GeneralizedFirstOrderAlgorithmCache{Vector{Float64}, Vector{Float64}, Vector{Float64}, SciMLBase.NullParameters, [types elided]...)
@ NonlinearSolveBase ~/.julia/packages/NonlinearSolveBase/mJsVc/src/solve.jl:287
[19] #__solve#155
@ ~/.julia/packages/NonlinearSolveBase/mJsVc/src/solve.jl:273 [inlined]
[20] __solve
@ ~/.julia/packages/NonlinearSolveBase/mJsVc/src/solve.jl:270 [inlined]
[21] macro expansion
@ ~/.julia/packages/NonlinearSolveBase/mJsVc/src/solve.jl:478 [inlined]
[22] __generated_polysolve(::SciMLBase.NonlinearProblem{Vector{Float64}, true, SciMLBase.NullParameters, [types elided]...)
@ NonlinearSolveBase ~/.julia/packages/NonlinearSolveBase/mJsVc/src/solve.jl:423
[23] __generated_polysolve
@ ~/.julia/packages/NonlinearSolveBase/mJsVc/src/solve.jl:423 [inlined]
[24] #__solve#166
@ ~/.julia/packages/NonlinearSolveBase/mJsVc/src/solve.jl:396 [inlined]
[25] __perform_ascher_iteration(cache::BoundaryValueDiffEqAscher.AscherCache{true, Float64, SciMLBase.BVProblem{Vector{Float64}, Tuple{Float64, Float64}, [types elided]...)
@ BoundaryValueDiffEqAscher ~/.julia/packages/BoundaryValueDiffEqAscher/CPeel/src/ascher.jl:162
[27] #__solve#34
@ ~/.julia/packages/BoundaryValueDiffEqCore/hFduL/src/BoundaryValueDiffEqCore.jl:38 [inlined]
[28] __solve
@ ~/.julia/packages/BoundaryValueDiffEqCore/hFduL/src/BoundaryValueDiffEqCore.jl:35 [inlined]
[29] #solve_call#23
@ ~/.julia/packages/DiffEqBase/aB45d/src/solve.jl:127 [inlined]
[30] solve_call
@ ~/.julia/packages/DiffEqBase/aB45d/src/solve.jl:84 [inlined]
[31] solve_up(prob::SciMLBase.BVProblem{Vector{Float64}, Tuple{Float64, Float64}, true, Nothing, SciMLBase.NullParameters, SciMLBase.BVPFunction{true, [types elided]...)
@ DiffEqBase ~/.julia/packages/DiffEqBase/aB45d/src/solve.jl:563
[32] solve_up
@ ~/.julia/packages/DiffEqBase/aB45d/src/solve.jl:540 [inlined]
[33] #solve#29
@ ~/.julia/packages/DiffEqBase/aB45d/src/solve.jl:530 [inlined]
[34] top-level scope
@ ~/mwe.jl:25
[35] include(mod::Module, _path::String)
@ Base ./Base.jl:306
[36] exec_options(opts::Base.JLOptions)
@ Base ./client.jl:317
[37] _start()
@ Base ./client.jl:550
in expression starting at mwe.jl:25
|
Contributor
Author
|
Created SciML/LinearSolve.jl#830 for the BVP example failing |
5 tasks
Contributor
Author
|
A local build was successful, and the LinearSolve.jl issue (#830) got solved... somehow, so I'm marking this as ready for review 👍 |
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.
Checklist
Additional context
Fixes errors in buildkite CI docs build.
Currently in draft mode: the example in tutorials/bvp_example.md (Ascher method for solving BVDAEs) fails on the newest LinearSolve.jl (3.47.0), but if pinned to LinearSolve.jl v3.40.0, it does produce results. Not sure what the underlying issues are yet.