Missing adapt for sparse and CUDABackend

[hexaeder]

Describe the bug

There is a difference between adapting a sparse matrix to CuArray vs CUDABackend. I would expect both to return a CuSparseMatrixCSC.

To reproduce

using Pkg
pkg"activate --temp"
pkg"add SparseArrays, CUDA, Adapt"

using SparseArrays, CUDA, Adapt, CUDA.CUSPARSE

A = sprand(Float32, 10, 10, 0.1)
adapt(CUDABackend(), A) isa SparseMatrixCSC # true
adapt(CuArray, A) isa CuSparseMatrixCSC     #true

Manifest.toml

(jl_5A56jY) pkg> st -m
Status `/tmp/jl_5A56jY/Manifest.toml`
  [621f4979] AbstractFFTs v1.5.0
  [79e6a3ab] Adapt v4.0.4
  [a9b6321e] Atomix v0.1.0
  [ab4f0b2a] BFloat16s v0.5.0
  [fa961155] CEnum v0.5.0
  [052768ef] CUDA v5.4.3
  [1af6417a] CUDA_Runtime_Discovery v0.3.4
  [3da002f7] ColorTypes v0.11.5
  [5ae59095] Colors v0.12.11
  [34da2185] Compat v4.15.0
  [a8cc5b0e] Crayons v4.1.1
  [9a962f9c] DataAPI v1.16.0
  [a93c6f00] DataFrames v1.6.1
  [864edb3b] DataStructures v0.18.20
  [e2d170a0] DataValueInterfaces v1.0.0
  [e2ba6199] ExprTools v0.1.10
  [53c48c17] FixedPointNumbers v0.8.5
  [0c68f7d7] GPUArrays v10.3.0
  [46192b85] GPUArraysCore v0.1.6
  [61eb1bfa] GPUCompiler v0.26.7
  [842dd82b] InlineStrings v1.4.2
  [41ab1584] InvertedIndices v1.3.0
  [82899510] IteratorInterfaceExtensions v1.0.0
  [692b3bcd] JLLWrappers v1.5.0
  [63c18a36] KernelAbstractions v0.9.22
  [929cbde3] LLVM v8.1.0
  [8b046642] LLVMLoopInfo v1.0.0
  [b964fa9f] LaTeXStrings v1.3.1
  [1914dd2f] MacroTools v0.5.13
  [e1d29d7a] Missings v1.2.0
  [5da4648a] NVTX v0.3.4
  [bac558e1] OrderedCollections v1.6.3
  [2dfb63ee] PooledArrays v1.4.3
  [aea7be01] PrecompileTools v1.2.1
  [21216c6a] Preferences v1.4.3
  [08abe8d2] PrettyTables v2.3.2
  [74087812] Random123 v1.7.0
  [e6cf234a] RandomNumbers v1.5.3
  [189a3867] Reexport v1.2.2
  [ae029012] Requires v1.3.0
  [6c6a2e73] Scratch v1.2.1
  [91c51154] SentinelArrays v1.4.5
  [a2af1166] SortingAlgorithms v1.2.1
  [90137ffa] StaticArrays v1.9.7
  [1e83bf80] StaticArraysCore v1.4.3
  [10745b16] Statistics v1.11.1
  [892a3eda] StringManipulation v0.3.4
  [3783bdb8] TableTraits v1.0.1
  [bd369af6] Tables v1.12.0
  [a759f4b9] TimerOutputs v0.5.24
  [013be700] UnsafeAtomics v0.2.1
  [d80eeb9a] UnsafeAtomicsLLVM v0.1.5
  [4ee394cb] CUDA_Driver_jll v0.9.1+1
  [76a88914] CUDA_Runtime_jll v0.14.1+0
  [9c1d0b0a] JuliaNVTXCallbacks_jll v0.2.1+0
  [dad2f222] LLVMExtra_jll v0.0.31+0
  [e98f9f5b] NVTX_jll v3.1.0+2
  [0dad84c5] ArgTools v1.1.2
  [56f22d72] Artifacts v1.11.0
  [2a0f44e3] Base64 v1.11.0
  [ade2ca70] Dates v1.11.0
  [f43a241f] Downloads v1.6.0
  [7b1f6079] FileWatching v1.11.0
  [9fa8497b] Future v1.11.0
  [b77e0a4c] InteractiveUtils v1.11.0
  [4af54fe1] LazyArtifacts v1.11.0
  [b27032c2] LibCURL v0.6.4
  [76f85450] LibGit2 v1.11.0
  [8f399da3] Libdl v1.11.0
  [37e2e46d] LinearAlgebra v1.11.0
  [56ddb016] Logging v1.11.0
  [d6f4376e] Markdown v1.11.0
  [ca575930] NetworkOptions v1.2.0
  [44cfe95a] Pkg v1.11.0
  [de0858da] Printf v1.11.0
  [3fa0cd96] REPL v1.11.0
  [9a3f8284] Random v1.11.0
  [ea8e919c] SHA v0.7.0
  [9e88b42a] Serialization v1.11.0
  [6462fe0b] Sockets v1.11.0
  [2f01184e] SparseArrays v1.11.0
  [f489334b] StyledStrings v1.11.0
  [fa267f1f] TOML v1.0.3
  [a4e569a6] Tar v1.10.0
  [8dfed614] Test v1.11.0
  [cf7118a7] UUIDs v1.11.0
  [4ec0a83e] Unicode v1.11.0
  [e66e0078] CompilerSupportLibraries_jll v1.1.1+0
  [deac9b47] LibCURL_jll v8.6.0+0
  [e37daf67] LibGit2_jll v1.7.2+0
  [29816b5a] LibSSH2_jll v1.11.0+1
  [c8ffd9c3] MbedTLS_jll v2.28.6+0
  [14a3606d] MozillaCACerts_jll v2023.12.12
  [4536629a] OpenBLAS_jll v0.3.27+1
  [bea87d4a] SuiteSparse_jll v7.7.0+0
  [83775a58] Zlib_jll v1.2.13+1
  [8e850b90] libblastrampoline_jll v5.10.1+0
  [8e850ede] nghttp2_jll v1.59.0+0
  [3f19e933] p7zip_jll v17.4.0+2

Version info

Details on Julia:

# please post the output of:
julia> versioninfo()
Julia Version 1.11.0-rc2
Commit 34c3a63147b (2024-07-29 06:24 UTC)
Build Info:
  Official https://julialang.org/ release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: 16 × AMD Ryzen 9 8945HS w/ Radeon 780M Graphics
  WORD_SIZE: 64
  LLVM: libLLVM-16.0.6 (ORCJIT, znver3)
Threads: 16 default, 0 interactive, 8 GC (on 16 virtual cores)
Environment:
  JULIA_NUM_THREADS = auto

Details on CUDA:

julia> CUDA.versioninfo()
CUDA runtime 12.5, artifact installation
CUDA driver 12.5
NVIDIA driver 555.99.0

CUDA libraries:
- CUBLAS: 12.5.3
- CURAND: 10.3.6
- CUFFT: 11.2.3
- CUSOLVER: 11.6.3
- CUSPARSE: 12.5.1
- CUPTI: 2024.2.1 (API 23.0.0)
- NVML: 12.0.0+555.52.1

Julia packages:
- CUDA: 5.4.3
- CUDA_Driver_jll: 0.9.1+1
- CUDA_Runtime_jll: 0.14.1+0

Toolchain:
- Julia: 1.11.0-rc2
- LLVM: 16.0.6

1 device:
  0: ������������������������������������������������������������ (sm_89, 7.444 GiB / 7.996 GiB available)

[vchuravy]

Hm these are the only adapt rules defined for the CUDABackend

CUDA.jl/src/CUDAKernels.jl

Lines 31 to 33 in d7077da

	Adapt.adapt_storage(::CUDABackend, a::Array) = Adapt.adapt(CuArray, a)
	Adapt.adapt_storage(::CUDABackend, a::CuArray) = a
	Adapt.adapt_storage(::KA.CPU, a::CuArray) = convert(Array, a)

[maleadt]

Is adapt(::Backend, x) even a public interface right now? It's mentioned nowhere in the docs, and also doesn't seem to be used by KA.jl at a quick glance, so I don't think you can expect the operation to be fleshed out.

[vchuravy]

I added it in JuliaGPU/KernelAbstractions.jl#364 inspired by DiffEqGPU and JuliaGPU/KernelAbstractions.jl#353

Ideally it would just mean the same as CuArray

[maleadt]

Right, but as it stands it's not part of the public API for KA.jl users, so @hexaeder I would recommend just using adapt(CuArray, x) which is documented and tested. If you think adapt(::Backend, x) should become a thing, please open an issue on KA.jl and we can consider extending the behavior of the adaptors that back-ends need to implement (right now, it's explicitly only Array <-> GPUArray).

[hexaeder]

Allright, I'll change it in my code! I probably stumbled over it by accident. adapt(backend, x) seems natural, and since it worked for simple cases, I just assumed that it is part of the official interface without rechecking the documentation. Sorry for the noise!

[vchuravy]

I just assumed that it is part of the official interface without rechecking the documentation. Sorry for the noise!

It's a fine assumption. I should have added it to the docs at the time.