Need adjoint for reinterpret SVector

Minimum working example:

f(x) = sum(sum(reinterpret(SVector{size(x, 1), eltype(x)}, x)))
Zygote.gradient(f, rand(3, 10))

ERROR: Need an adjoint for constructor Base.ReinterpretArray{SVector{3, Float64}, 2, Float64, Matrix{Float64}, false}. Gradient is of type FillArrays.Fill{FillArrays.Fill{Float64, 1, Tuple{Base.OneTo{Int64}}}, 2, Tuple{Base.OneTo{Int64}, Base.OneTo{Int64}}}
Stacktrace:
 [1] error(s::String)
   @ Base ./error.jl:33
 [2] (::Zygote.Jnew{Base.ReinterpretArray{SVector{3, Float64}, 2, Float64, Matrix{Float64}, false}, Nothing, false})(Δ::FillArrays.Fill{FillArrays.Fill{Float64, 1, Tuple{Base.OneTo{Int64}}}, 2, Tuple{Base.OneTo{Int64}, Base.OneTo{Int64}}})
   @ Zygote ~/.julia/packages/Zygote/ajuwN/src/lib/lib.jl:323
 [3] (::Zygote.var"#1811#back#235"{Zygote.Jnew{Base.ReinterpretArray{SVector{3, Float64}, 2, Float64, Matrix{Float64}, false}, Nothing, false}})(Δ::FillArrays.Fill{FillArrays.Fill{Float64, 1, Tuple{Base.OneTo{Int64}}}, 2, Tuple{Base.OneTo{Int64}, Base.OneTo{Int64}}})
   @ Zygote ~/.julia/packages/ZygoteRules/AIbCs/src/adjoint.jl:67
 [4] Pullback
   @ ./reinterpretarray.jl:47 [inlined]
 [5] (::typeof(∂(reinterpret)))(Δ::FillArrays.Fill{FillArrays.Fill{Float64, 1, Tuple{Base.OneTo{Int64}}}, 2, Tuple{Base.OneTo{Int64}, Base.OneTo{Int64}}})
   @ Zygote ~/.julia/packages/Zygote/ajuwN/src/compiler/interface2.jl:0
 [6] Pullback
   @ ./REPL[95]:1 [inlined]
 [7] (::Zygote.var"#52#53"{typeof(∂(f))})(Δ::Float64)
   @ Zygote ~/.julia/packages/Zygote/ajuwN/src/compiler/interface.jl:41
 [8] gradient(::Function, ::Matrix{Float64}, ::Vararg{Any})
   @ Zygote ~/.julia/packages/Zygote/ajuwN/src/compiler/interface.jl:76
 [9] top-level scope
   @ REPL[98]:1

Start with the supplied minimum working example and the Base.ReinterpretArray constructor shown in the error. Investigate how reinterpret(SVector{size(x, 1), eltype(x)}, x) is handled during reverse-mode differentiation. Done means the example runs without the missing-adjoint error and produces a gradient for the input matrix.