Merge pull request #611 from SciML/test

Fix test setup

Author: ChrisRackauckas

.github/workflows/Tests.yml

@@ -32,7 +32,7 @@ jobs:
           - "DefaultsLoading"
           - "LinearSolveHYPRE"
           - "LinearSolvePardiso"
-          - "Enzyme"
+          - "NoPre"
     uses: "SciML/.github/.github/workflows/tests.yml@v1"
     with:
       group: "${{ matrix.group }}"

Project.toml

@@ -75,7 +75,6 @@ ChainRulesCore = "1.22"
 ConcreteStructs = "0.2.3"
 DocStringExtensions = "0.9.3"
 EnumX = "1.0.4"
-Enzyme = "0.13"
 EnzymeCore = "0.8.1"
 FastAlmostBandedMatrices = "0.1"
 FastLapackInterface = "2"
@@ -85,7 +84,6 @@ GPUArraysCore = "0.1.6, 0.2"
 HYPRE = "1.4.0"
 InteractiveUtils = "1.10"
 IterativeSolvers = "0.9.3"
-JET = "0.8.28, 0.9"
 KernelAbstractions = "0.9.27"
 Krylov = "0.10"
 KrylovKit = "0.8, 0.9"
@@ -124,15 +122,13 @@ AllocCheck = "9b6a8646-10ed-4001-bbdc-1d2f46dfbb1a"
 Aqua = "4c88cf16-eb10-579e-8560-4a9242c79595"
 BandedMatrices = "aae01518-5342-5314-be14-df237901396f"
 BlockDiagonals = "0a1fb500-61f7-11e9-3c65-f5ef3456f9f0"
-Enzyme = "7da242da-08ed-463a-9acd-ee780be4f1d9"
 FastAlmostBandedMatrices = "9d29842c-ecb8-4973-b1e9-a27b1157504e"
 FastLapackInterface = "29a986be-02c6-4525-aec4-84b980013641"
 FiniteDiff = "6a86dc24-6348-571c-b903-95158fe2bd41"
 ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210"
 HYPRE = "b5ffcf37-a2bd-41ab-a3da-4bd9bc8ad771"
 InteractiveUtils = "b77e0a4c-d291-57a0-90e8-8db25a27a240"
 IterativeSolvers = "42fd0dbc-a981-5370-80f2-aaf504508153"
-JET = "c3a54625-cd67-489e-a8e7-0a5a0ff4e31b"
 KernelAbstractions = "63c18a36-062a-441e-b654-da1e3ab1ce7c"
 KrylovKit = "0b1a1467-8014-51b9-945f-bf0ae24f4b77"
 KrylovPreconditioners = "45d422c2-293f-44ce-8315-2cb988662dec"
@@ -152,4 +148,4 @@ Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
 Zygote = "e88e6eb3-aa80-5325-afca-941959d7151f"
 
 [targets]
-test = ["Aqua", "Test", "IterativeSolvers", "InteractiveUtils", "JET", "KrylovKit", "KrylovPreconditioners", "Pkg", "Random", "SafeTestsets", "MultiFloats", "ForwardDiff", "HYPRE", "MPI", "BlockDiagonals", "Enzyme", "FiniteDiff", "BandedMatrices", "FastAlmostBandedMatrices", "StaticArrays", "AllocCheck", "StableRNGs", "Zygote", "RecursiveFactorization", "Sparspak", "FastLapackInterface", "SparseArrays"]
+test = ["Aqua", "Test", "IterativeSolvers", "InteractiveUtils", "KrylovKit", "KrylovPreconditioners", "Pkg", "Random", "SafeTestsets", "MultiFloats", "ForwardDiff", "HYPRE", "MPI", "BlockDiagonals", "FiniteDiff", "BandedMatrices", "FastAlmostBandedMatrices", "StaticArrays", "AllocCheck", "StableRNGs", "Zygote", "RecursiveFactorization", "Sparspak", "FastLapackInterface", "SparseArrays"]

docs/src/basics/FAQ.md

@@ -26,7 +26,7 @@ the performance to the expected state, please open an issue and we will improve
 This is addressed in the [JuliaCon 2022 video](https://www.youtube.com/watch?v=JWI34_w-yYw&t=182s). This happens in
 a few ways:
 
- 1. The Fortran/C code that NumPy/SciPy uses is actually slow. It's [OpenBLAS](https://github.com/xianyi/OpenBLAS),
+ 1. The Fortran/C code that NumPy/SciPy uses is actually slow. It's [OpenBLAS](https://github.com/OpenMathLib/OpenBLAS),
     a library developed in part by the Julia Lab back in 2012 as a fast open source BLAS implementation. Many
     open source environments now use this build, including many R distributions. However, the Julia Lab has greatly
     improved its ability to generate optimized SIMD in platform-specific ways. This, and improved multithreading support

docs/src/index.md

@@ -23,7 +23,7 @@ Pkg.add("LinearSolve")
 ## Contributing
 
   - Please refer to the
-    [SciML ColPrac: Contributor's Guide on Collaborative Practices for Community Packages](https://github.com/SciML/ColPrac/blob/master/README.md)
+    [SciML ColPrac: Contributor's Guide on Collaborative Practices for Community Packages](https://github.com/SciML/ColPrac)
     for guidance on PRs, issues, and other matters relating to contributing to SciML.
 
   - See the [SciML Style Guide](https://github.com/SciML/SciMLStyle) for common coding practices and other style decisions.

src/common.jl

@@ -255,7 +255,6 @@ function SciMLBase.reinit!(cache::LinearCache;
         b = cache.b,
         u = cache.u,
         p = nothing,
-        reinit_cache = false,
         reuse_precs = false)
     (; alg, cacheval, abstol, reltol, maxiters, verbose, assumptions, sensealg) = cache
 
@@ -270,23 +269,16 @@ function SciMLBase.reinit!(cache::LinearCache;
     p = isnothing(p) ? cache.p : p
     Pl = cache.Pl
     Pr = cache.Pr
-    if reinit_cache
-        return LinearCache{
-            typeof(A), typeof(b), typeof(u), typeof(p), typeof(alg), typeof(cacheval),
-            typeof(Pl), typeof(Pr), typeof(reltol), typeof(assumptions.issq),
-            typeof(sensealg)}(
-            A, b, u, p, alg, cacheval, precsisfresh, isfresh, Pl, Pr, abstol, reltol,
-            maxiters, verbose, assumptions, sensealg)
-    else
-        cache.A = A
-        cache.b = b
-        cache.u = u
-        cache.p = p
-        cache.Pl = Pl
-        cache.Pr = Pr
-        cache.isfresh = true
-        cache.precsisfresh = precsisfresh
-    end
+
+    cache.A = A
+    cache.b = b
+    cache.u = u
+    cache.p = p
+    cache.Pl = Pl
+    cache.Pr = Pr
+    cache.isfresh = true
+    cache.precsisfresh = precsisfresh
+    nothing
 end
 
 function SciMLBase.solve(prob::LinearProblem, args...; kwargs...)

test/default_algs.jl

@@ -1,4 +1,5 @@
-using LinearSolve, RecursiveFactorization, LinearAlgebra, SparseArrays, Test, JET
+using LinearSolve, RecursiveFactorization, LinearAlgebra, SparseArrays, Test
+
 @test LinearSolve.defaultalg(nothing, zeros(3)).alg ===
       LinearSolve.DefaultAlgorithmChoice.GenericLUFactorization
 prob = LinearProblem(rand(3, 3), rand(3))
@@ -55,19 +56,7 @@ solve(prob)
 A = rand(4, 4)
 b = rand(4)
 prob = LinearProblem(A, b)
-VERSION ≥ v"1.10-" && JET.@test_opt init(prob, nothing)
-JET.@test_opt solve(prob, LUFactorization())
-JET.@test_opt solve(prob, GenericLUFactorization())
-@test_skip JET.@test_opt solve(prob, QRFactorization())
-JET.@test_opt solve(prob, DiagonalFactorization())
-#JET.@test_opt solve(prob, SVDFactorization())
-#JET.@test_opt solve(prob, KrylovJL_GMRES())
-
 prob = LinearProblem(sparse(A), b)
-#JET.@test_opt solve(prob, UMFPACKFactorization())
-#JET.@test_opt solve(prob, KLUFactorization())
-#JET.@test_opt solve(prob, SparspakFactorization())
-#JET.@test_opt solve(prob)
 @inferred solve(prob)
 @inferred init(prob, nothing)
 
@@ -81,8 +70,8 @@ m, n = 2, 2
 A = rand(m, n)
 b = rand(m)
 x = rand(n)
-f = (du, u, p, t) -> mul!(du, A, u)
-fadj = (du, u, p, t) -> mul!(du, A', u)
+f = (w, v, u, p, t) -> mul!(w, A, v)
+fadj = (w, v, u, p, t) -> mul!(w, A', v)
 funcop = FunctionOperator(f, x, b; op_adjoint = fadj)
 prob = LinearProblem(funcop, b)
 sol1 = solve(prob)
@@ -93,8 +82,8 @@ m, n = 3, 2
 A = rand(m, n)
 b = rand(m)
 x = rand(n)
-f = (du, u, p, t) -> mul!(du, A, u)
-fadj = (du, u, p, t) -> mul!(du, A', u)
+f = (w, v, u, p, t) -> mul!(w, A, v)
+fadj = (w, v, u, p, t) -> mul!(w, A', v)
 funcop = FunctionOperator(f, x, b; op_adjoint = fadj)
 prob = LinearProblem(funcop, b)
 sol1 = solve(prob)
@@ -105,8 +94,8 @@ m, n = 2, 3
 A = rand(m, n)
 b = rand(m)
 x = rand(n)
-f = (du, u, p, t) -> mul!(du, A, u)
-fadj = (du, u, p, t) -> mul!(du, A', u)
+f = (w, v, u, p, t) -> mul!(w, A, v)
+fadj = (w, v, u, p, t) -> mul!(w, A', v)
 funcop = FunctionOperator(f, x, b; op_adjoint = fadj)
 prob = LinearProblem(funcop, b)
 sol1 = solve(prob)

test/nopre/Project.toml

@@ -0,0 +1,11 @@
+[deps]
+FiniteDiff = "6a86dc24-6348-571c-b903-95158fe2bd41"
+ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210"
+Enzyme = "7da242da-08ed-463a-9acd-ee780be4f1d9"
+JET = "c3a54625-cd67-489e-a8e7-0a5a0ff4e31b"
+LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
+LinearSolve = "7ed4a6bd-45f5-4d41-b270-4a48e9bafcae"
+RecursiveFactorization = "f2c3362d-daeb-58d1-803e-2bc74f2840b4"
+SparseArrays = "2f01184e-e22b-5df5-ae63-d93ebab69eaf"
+StableRNGs = "860ef19b-820b-49d6-a774-d7a799459cd3"
+StaticArrays = "90137ffa-7385-5640-81b9-e52037218182"

test/enzyme.jl renamed to test/nopre/enzyme.jl

@@ -0,0 +1,19 @@
+using LinearSolve, RecursiveFactorization, LinearAlgebra, SparseArrays, Test
+using JET
+
+A = rand(4, 4)
+b = rand(4)
+prob = LinearProblem(A, b)
+JET.@test_opt init(prob, nothing)
+JET.@test_opt solve(prob, LUFactorization())
+JET.@test_opt solve(prob, GenericLUFactorization())
+@test_skip JET.@test_opt solve(prob, QRFactorization())
+JET.@test_opt solve(prob, DiagonalFactorization())
+#JET.@test_opt solve(prob, SVDFactorization())
+#JET.@test_opt solve(prob, KrylovJL_GMRES())
+
+prob = LinearProblem(sparse(A), b)
+#JET.@test_opt solve(prob, UMFPACKFactorization())
+#JET.@test_opt solve(prob, KLUFactorization())
+#JET.@test_opt solve(prob, SparspakFactorization())
+#JET.@test_opt solve(prob)

test/nopre/jet.jl

@@ -17,7 +17,7 @@ for alg in (nothing, LUFactorization(), SVDFactorization(), CholeskyFactorizatio
     sol = solve(LinearProblem(A, b), alg)
     @inferred solve(LinearProblem(A, b), alg)
     @test norm(A * sol .- b) < 1e-10
-    if alg isa KrylovJL{typeof(LinearSolve.Krylov.gmres!)}
+    if alg isa KrylovJL{typeof(LinearSolve.Krylov.gmres!)} && isempty(VERSION.prerelease)
         @test_broken __solve_no_alloc(A, b, alg) isa SciMLBase.LinearSolution
     else
         @test_nowarn __solve_no_alloc(A, b, alg) isa SciMLBase.LinearSolution