Automatically convert incompatible vector problems to svector

Required for diffeqr and diffeqpy. Note that the `f` functions are okay when generated by MTK, since it generates a dispatch for SA -> SA, so therefore all that's needed is the u0 and p conversion for this to work downstream.

src/algorithms.jl

@@ -99,8 +99,8 @@ imparts some extra limitations on the use.
 ## Positional Arguments
 
   - `backend`: the KernelAbstractions backend for performing the computation.
-  - `cpu_offload`: the percentage of trajectories to offload to the CPU. Default is 0.2 or
-    20% of trajectories.
+  - `cpu_offload`: the percentage of trajectories to offload to the CPU. Default is 0.0 or
+    0% of trajectories.
 
 ## Limitations
 
@@ -165,7 +165,7 @@ function EnsembleGPUArray(dev)
 end
 
 function EnsembleGPUKernel(dev)
-    EnsembleGPUKernel(dev, 0.2)
+    EnsembleGPUKernel(dev, 0.0)
 end
 
 function ChainRulesCore.rrule(::Type{<:EnsembleGPUArray})

src/ensemblegpukernel/lowerlevel_solve.jl

@@ -34,6 +34,8 @@ function vectorized_solve(probs, prob::ODEProblem, alg;
     timeseries = prob.tspan[1]:dt:prob.tspan[2]
     nsteps = length(timeseries)
 
+    prob = convert(ImmutableODEProblem, prob)
+
     dt = convert(eltype(prob.tspan), dt)
 
     if saveat === nothing
@@ -154,6 +156,8 @@ function vectorized_asolve(probs, prob::ODEProblem, alg;
     backend = get_backend(probs)
     backend = maybe_prefer_blocks(backend)
 
+    prob = convert(ImmutableODEProblem, prob)
+
     dt = convert(eltype(prob.tspan), dt)
     abstol = convert(eltype(prob.tspan), abstol)
     reltol = convert(eltype(prob.tspan), reltol)

src/ensemblegpukernel/problems/ode_problems.jl

@@ -95,11 +95,17 @@ function ImmutableODEProblem(f, u0, tspan, p = NullParameters(); kwargs...)
     ImmutableODEProblem(_f, _u0, _tspan, p; kwargs...)
 end
 
+staticarray_itize(x) = x
+staticarray_itize(x::Vector) = SVector{length(x)}(x)
+staticarray_itize(x::SizedVector) = SVector{length(x)}(x)
+staticarray_itize(x::Matrix) = SMatrix{size(x)...}(x)
+staticarray_itize(x::SizedMatrix) = SMatrix{size(x)...}(x)
+
 function Base.convert(::Type{ImmutableODEProblem}, prob::T) where {T <: ODEProblem}
     ImmutableODEProblem(prob.f,
-        prob.u0,
+        staticarray_itize(prob.u0),
         prob.tspan,
-        prob.p,
+        staticarray_itize(prob.p),
         prob.problem_type;
         prob.kwargs...)
 end

test/gpu_kernel_de/conversions.jl

@@ -0,0 +1,22 @@
+using DiffEqGPU, OrdinaryDiffEq, StaticArrays, LinearAlgebra
+include("../utils.jl")
+
+function lorenz(u, p, t)
+    σ = p[1]
+    ρ = p[2]
+    β = p[3]
+    du1 = σ * (u[2] - u[1])
+    du2 = u[1] * (ρ - u[3]) - u[2]
+    du3 = u[1] * u[2] - β * u[3]
+    return SVector{3}(du1, du2, du3)
+end
+
+u0 = [1.0f0; 0.0f0; 0.0f0]
+tspan = (0.0f0, 10.0f0)
+p = [10.0f0, 28.0f0, 8 / 3.0f0]
+prob = ODEProblem{false}(lorenz, u0, tspan, p)
+prob_func = (prob, i, repeat) -> remake(prob, p = (@SVector rand(Float32, 3)) .* p)
+monteprob = EnsembleProblem(prob, prob_func = prob_func, safetycopy = false)
+sol = solve(monteprob, GPUTsit5(), EnsembleGPUKernel(backend),
+    trajectories = 10_000,
+    saveat = 1.0f0);

test/runtests.jl

@@ -43,6 +43,10 @@ end
     include("gpu_kernel_de/finite_diff.jl")
 end
 
+@time @safetestset "GPU Kernelized Auto-Conversion tests" begin
+    include("gpu_kernel_de/conversions.jl")
+end
+
 if GROUP in SUPPORTS_LUFACT
     @time @safetestset "EnsembleGPUArray" begin
         include("ensemblegpuarray.jl")