Very strange behavior with ReverseDiff.jl compiled when not linking

[torfjelde]

I'm seeing some very strange behavior with compiled ReverseDiff.jl:

It gets the correct results when linking, but not when we don't link.

julia> using Turing, Zygote, ReverseDiff, LogDensityProblems, LogDensityProblemsAD

julia> # Test model.
       @model function demo(x, y)
           s ~ InverseGamma(2, 3)
           m ~ Normal(0, sqrt(s))
           x ~ Normal(m, sqrt(s))
           y ~ Normal(m, sqrt(s))
       end
demo (generic function with 2 methods)

julia> model = demo(1.5, 2.0);

julia> # Make the benchmark suite.
       backends = [
           Turing.Essential.ForwardDiffAD{0}(),
           Turing.Essential.ReverseDiffAD{false}(),
           Turing.Essential.ReverseDiffAD{true}()
       ];

julia> context = DynamicPPL.DefaultContext();

julia> varinfo = DynamicPPL.VarInfo(model);

julia> θ = varinfo[:]
2-element Vector{Float64}:
  0.877721788997423
 -0.4609132411104931

julia> for backend in backends
           f = LogDensityProblemsAD.ADgradient(
               backend,
               DynamicPPL.LogDensityFunction(
                   DynamicPPL.unflatten(varinfo, θ),
                   model,
                   context
               )
           )
           @info "$backend" LogDensityProblems.logdensity_and_gradient(f, θ)
       end
┌ Info: Turing.Essential.ForwardDiffAD{0, true}()
└   LogDensityProblems.logdensity_and_gradient(f, θ) = (-9.15196013596708, [5.331170301312024, 5.562969706960088])
┌ Info: Turing.Essential.ReverseDiffAD{false}()
└   LogDensityProblems.logdensity_and_gradient(f, θ) = (-9.15196013596708, [5.331170301312024, 5.562969706960087])
┌ Info: Turing.Essential.ReverseDiffAD{true}()
└   LogDensityProblems.logdensity_and_gradient(f, θ) = (-9.26758782412419, [6.332605592854213, 5.498758418990619])

julia> varinfo_linked = DynamicPPL.link!!(
           DynamicPPL.VarInfo(model),
           model
       );

julia> θ_linked = varinfo_linked[:]
2-element Vector{Float64}:
 -0.41309731702034985
 -0.2283878061115611

julia> for backend in backends
           f_linked = LogDensityProblemsAD.ADgradient(
               backend,
               DynamicPPL.LogDensityFunction(
                   DynamicPPL.unflatten(varinfo_linked, θ_linked),
                   model,
                   context
               )
           )

           @info "$backend" LogDensityProblems.logdensity_and_gradient(f_linked, θ_linked)
       end
┌ Info: Turing.Essential.ForwardDiffAD{0, true}()
└   LogDensityProblems.logdensity_and_gradient(f_linked, θ_linked) = (-9.698123276822646, [7.084372864116072, 6.325841497112945])
┌ Info: Turing.Essential.ReverseDiffAD{false}()
└   LogDensityProblems.logdensity_and_gradient(f_linked, θ_linked) = (-9.698123276822646, [7.084372864116072, 6.325841497112945])
┌ Info: Turing.Essential.ReverseDiffAD{true}()
└   LogDensityProblems.logdensity_and_gradient(f_linked, θ_linked) = (-9.698123276822646, [7.084372864116072, 6.325841497112945])

Notice the differences in the first computations.

What in the world is this? Seen this before @devmotion @willtebbutt @yebai ?

Environment

julia> versioninfo()
Julia Version 1.9.3
Commit bed2cd540a1 (2023-08-24 14:43 UTC)
Build Info:
  Official https://julialang.org/ release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: 12 × Intel(R) Core(TM) i7-10710U CPU @ 1.10GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-14.0.6 (ORCJIT, skylake)
  Threads: 1 on 12 virtual cores

(jl_f3oaDw) pkg> st --manifest
Status `/tmp/jl_f3oaDw/Manifest.toml`
  [47edcb42] ADTypes v0.2.4
  [621f4979] AbstractFFTs v1.5.0
  [80f14c24] AbstractMCMC v4.4.2
  [7a57a42e] AbstractPPL v0.6.2
  [1520ce14] AbstractTrees v0.4.4
  [79e6a3ab] Adapt v3.6.2
  [0bf59076] AdvancedHMC v0.5.5
  [5b7e9947] AdvancedMH v0.7.5
  [576499cb] AdvancedPS v0.4.3
  [b5ca4192] AdvancedVI v0.2.4
  [dce04be8] ArgCheck v2.3.0
  [4fba245c] ArrayInterface v7.4.11
  [a9b6321e] Atomix v0.1.0
  [13072b0f] AxisAlgorithms v1.0.1
  [39de3d68] AxisArrays v0.4.7
  [198e06fe] BangBang v0.3.39
  [9718e550] Baselet v0.1.1
  [76274a88] Bijectors v0.13.7
  [fa961155] CEnum v0.4.2
  [49dc2e85] Calculus v0.5.1
  [082447d4] ChainRules v1.54.0
  [d360d2e6] ChainRulesCore v1.16.0
  [9e997f8a] ChangesOfVariables v0.1.8
  [861a8166] Combinatorics v1.0.2
  [38540f10] CommonSolve v0.2.4
  [bbf7d656] CommonSubexpressions v0.3.0
  [34da2185] Compat v4.9.0
  [a33af91c] CompositionsBase v0.1.2
  [88cd18e8] ConsoleProgressMonitor v0.1.2
  [187b0558] ConstructionBase v1.5.4
  [a8cc5b0e] Crayons v4.1.1
  [9a962f9c] DataAPI v1.15.0
  [864edb3b] DataStructures v0.18.15
  [e2d170a0] DataValueInterfaces v1.0.0
  [244e2a9f] DefineSingletons v0.1.2
  [8bb1440f] DelimitedFiles v1.9.1
  [b429d917] DensityInterface v0.4.0
  [163ba53b] DiffResults v1.1.0
  [b552c78f] DiffRules v1.15.1
  [31c24e10] Distributions v0.25.100
  [ced4e74d] DistributionsAD v0.6.52
  [ffbed154] DocStringExtensions v0.9.3
  [fa6b7ba4] DualNumbers v0.6.8
  [366bfd00] DynamicPPL v0.23.17
  [cad2338a] EllipticalSliceSampling v1.1.0
  [4e289a0a] EnumX v1.0.4
  [e2ba6199] ExprTools v0.1.10
  [7a1cc6ca] FFTW v1.7.1
  [1a297f60] FillArrays v1.6.1
  [59287772] Formatting v0.4.2
  [f6369f11] ForwardDiff v0.10.36
  [069b7b12] FunctionWrappers v1.1.3
  [77dc65aa] FunctionWrappersWrappers v0.1.3
  [d9f16b24] Functors v0.4.5
  [0c68f7d7] GPUArrays v9.0.0
  [46192b85] GPUArraysCore v0.1.5
  [34004b35] HypergeometricFunctions v0.3.23
  [7869d1d1] IRTools v0.4.10
  [22cec73e] InitialValues v0.3.1
  [505f98c9] InplaceOps v0.3.0
  [a98d9a8b] Interpolations v0.14.7
  [8197267c] IntervalSets v0.7.7
  [3587e190] InverseFunctions v0.1.12
  [41ab1584] InvertedIndices v1.3.0
  [92d709cd] IrrationalConstants v0.2.2
  [c8e1da08] IterTools v1.8.0
  [82899510] IteratorInterfaceExtensions v1.0.0
  [692b3bcd] JLLWrappers v1.5.0
  [63c18a36] KernelAbstractions v0.9.8
  [5ab0869b] KernelDensity v0.6.7
  [929cbde3] LLVM v6.2.1
  [8ac3fa9e] LRUCache v1.5.0
  [b964fa9f] LaTeXStrings v1.3.0
  [50d2b5c4] Lazy v0.15.1
  [1d6d02ad] LeftChildRightSiblingTrees v0.2.0
  [6f1fad26] Libtask v0.8.6
  [6fdf6af0] LogDensityProblems v2.1.1
⌃ [996a588d] LogDensityProblemsAD v1.5.0
  [2ab3a3ac] LogExpFunctions v0.3.26
  [e6f89c97] LoggingExtras v1.0.2
  [c7f686f2] MCMCChains v6.0.3
  [be115224] MCMCDiagnosticTools v0.3.5
  [e80e1ace] MLJModelInterface v1.9.2
  [1914dd2f] MacroTools v0.5.11
  [dbb5928d] MappedArrays v0.4.2
  [128add7d] MicroCollections v0.1.4
  [e1d29d7a] Missings v1.1.0
  [872c559c] NNlib v0.9.5
  [77ba4419] NaNMath v1.0.2
  [86f7a689] NamedArrays v0.10.0
  [c020b1a1] NaturalSort v1.0.0
  [6fe1bfb0] OffsetArrays v1.12.10
⌅ [3bd65402] Optimisers v0.2.20
  [bac558e1] OrderedCollections v1.6.2
  [90014a1f] PDMats v0.11.17
  [aea7be01] PrecompileTools v1.2.0
  [21216c6a] Preferences v1.4.0
  [08abe8d2] PrettyTables v2.2.7
  [33c8b6b6] ProgressLogging v0.1.4
  [92933f4c] ProgressMeter v1.9.0
  [1fd47b50] QuadGK v2.8.2
  [74087812] Random123 v1.6.1
  [e6cf234a] RandomNumbers v1.5.3
  [b3c3ace0] RangeArrays v0.3.2
  [c84ed2f1] Ratios v0.4.5
  [c1ae055f] RealDot v0.1.0
  [3cdcf5f2] RecipesBase v1.3.4
  [731186ca] RecursiveArrayTools v2.38.10
  [189a3867] Reexport v1.2.2
  [ae029012] Requires v1.3.0
  [37e2e3b7] ReverseDiff v1.15.1
  [79098fc4] Rmath v0.7.1
  [f2b01f46] Roots v2.0.19
  [7e49a35a] RuntimeGeneratedFunctions v0.5.12
  [0bca4576] SciMLBase v1.98.0
  [c0aeaf25] SciMLOperators v0.3.6
  [30f210dd] ScientificTypesBase v3.0.0
  [efcf1570] Setfield v1.1.1
  [ce78b400] SimpleUnPack v1.1.0
  [a2af1166] SortingAlgorithms v1.1.1
  [dc90abb0] SparseInverseSubset v0.1.1
  [276daf66] SpecialFunctions v2.3.1
  [171d559e] SplittablesBase v0.1.15
  [90137ffa] StaticArrays v1.6.3
  [1e83bf80] StaticArraysCore v1.4.2
  [64bff920] StatisticalTraits v3.2.0
  [82ae8749] StatsAPI v1.7.0
  [2913bbd2] StatsBase v0.34.0
  [4c63d2b9] StatsFuns v1.3.0
  [892a3eda] StringManipulation v0.3.4
  [09ab397b] StructArrays v0.6.16
  [2efcf032] SymbolicIndexingInterface v0.2.2
  [3783bdb8] TableTraits v1.0.1
  [bd369af6] Tables v1.11.0
  [5d786b92] TerminalLoggers v0.1.7
  [9f7883ad] Tracker v0.2.26
  [28d57a85] Transducers v0.4.78
  [410a4b4d] Tricks v0.1.7
  [781d530d] TruncatedStacktraces v1.4.0
  [fce5fe82] Turing v0.29.1
  [013be700] UnsafeAtomics v0.2.1
  [d80eeb9a] UnsafeAtomicsLLVM v0.1.3
  [efce3f68] WoodburyMatrices v0.5.5
  [e88e6eb3] Zygote v0.6.64
  [700de1a5] ZygoteRules v0.2.3
  [f5851436] FFTW_jll v3.3.10+0
  [1d5cc7b8] IntelOpenMP_jll v2023.2.0+0
  [dad2f222] LLVMExtra_jll v0.0.25+0
  [856f044c] MKL_jll v2023.2.0+0
  [efe28fd5] OpenSpecFun_jll v0.5.5+0
  [f50d1b31] Rmath_jll v0.4.0+0
  [0dad84c5] ArgTools v1.1.1
  [56f22d72] Artifacts
  [2a0f44e3] Base64
  [ade2ca70] Dates
  [8ba89e20] Distributed
  [f43a241f] Downloads v1.6.0
  [7b1f6079] FileWatching
  [9fa8497b] Future
  [b77e0a4c] InteractiveUtils
  [4af54fe1] LazyArtifacts
  [b27032c2] LibCURL v0.6.3
  [76f85450] LibGit2
  [8f399da3] Libdl
  [37e2e46d] LinearAlgebra
  [56ddb016] Logging
  [d6f4376e] Markdown
  [a63ad114] Mmap
  [ca575930] NetworkOptions v1.2.0
  [44cfe95a] Pkg v1.9.2
  [de0858da] Printf
  [3fa0cd96] REPL
  [9a3f8284] Random
  [ea8e919c] SHA v0.7.0
  [9e88b42a] Serialization
  [1a1011a3] SharedArrays
  [6462fe0b] Sockets
  [2f01184e] SparseArrays
  [10745b16] Statistics v1.9.0
  [4607b0f0] SuiteSparse
  [fa267f1f] TOML v1.0.3
  [a4e569a6] Tar v1.10.0
  [8dfed614] Test
  [cf7118a7] UUIDs
  [4ec0a83e] Unicode
  [e66e0078] CompilerSupportLibraries_jll v1.0.5+0
  [deac9b47] LibCURL_jll v7.84.0+0
  [29816b5a] LibSSH2_jll v1.10.2+0
  [c8ffd9c3] MbedTLS_jll v2.28.2+0
  [14a3606d] MozillaCACerts_jll v2022.10.11
  [4536629a] OpenBLAS_jll v0.3.21+4
  [05823500] OpenLibm_jll v0.8.1+0
  [bea87d4a] SuiteSparse_jll v5.10.1+6
  [83775a58] Zlib_jll v1.2.13+0
  [8e850b90] libblastrampoline_jll v5.8.0+0
  [8e850ede] nghttp2_jll v1.48.0+0
  [3f19e933] p7zip_jll v17.4.0+0
Info Packages marked with ⌃ and ⌅ have new versions available, but those with ⌅ are restricted by compatibility constraints from upgrading. To see why use `status --outdated -m`

[torfjelde]

This was resolve by #2097