JuliaGPU / JuliaGPU/GPUCompiler.jl

"Unexpected internal error" caused by mapping over CuArray

Open
#170 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Julia
Stars
187
Forks
68
Avg merge
1d 12h
Merged PRs (30d)
28

Description

I started experimenting with CUDA.jl yesterday and managed to produce an error with a simple example. The error message requested that I file an issue here. My apologies if this is caused by my doing something naive -- I'm just following orders! A minimal example that produces the error is this:

```
module MinimalFailure

using CUDA

function badfunc!(i::Int, results::CuArray{Float32,1})
results[i] = 0.0f0
return
end

n=1024
data = CuArray{Float32,1}(undef,n)
idcs = CuArray(collect(1:n))
map(i -> badfunc!(i,data), idcs)

end
```

The resulting error message, system info and backtrace are below:

```
WARNING: replacing module MinimalFailure.
ERROR: LoadError: GPUCompiler.jl encountered an unexpected internal error.
Please file an issue attaching the following information, including the backtrace,
as well as a reproducible example (if possible).

InternalCompilerError: length(frames) == 1, at /home/dahlbom/.julia/packages/GPUCompiler/uTpNx/src/validation.jl:246

Compiler invocation: PTX CompilerJob of kernel broadcast_kernel(CUDA.CuKernelContext, CuDeviceArray{Nothing,1,1}, Base.Broadcast.Broadcasted{Nothing,Tuple{Base.OneTo{Int64}},Main.MinimalFailure.var"#1#2",Tuple{Base.Broadcast.Extruded{CuDeviceArray{Int64,1,1},Tuple{Bool},Tuple{Int64}}}}, Int64) for sm_61

Additional information:
- frames = svec(svec(:jl_symbol_name_, Symbol("/buildworker/worker/package_linux64/build/src/julia.h"), 970, nothing, true, true), svec(:jl_symbol_name, Symbol("/buildworker/worker/package_linux64/build/src/rtutils.c"), 333, nothing, true, false))

Installed packages:
- CommonSolve = v"0.2.0"
- DelNetExperiment = v"0.1.0"
- Requires = v"1.1.3"
- Gtk = v"1.1.7"
- GPUArrays = v"6.2.1"
- ArrayLayouts = v"0.3.8"
- CommonSubexpressions = v"0.3.0"
- IteratorInterfaceExtensions = v"1.0.0"
- Xorg_libXtst_jll = v"1.2.3+4"
- Graphics = v"1.1.0"
- GeometryBasics = v"0.2.15"
- CodeTracking = v"1.0.5"
- Makie = v"0.11.0"
- Libiconv_jll = v"1.16.0+7"
- JuliaVariables = v"0.2.4"
- NLSolversBase = v"7.5.0"
- FileIO = v"1.6.5"
- Base64 = nothing
- GLM = v"1.4.1"
- Wayland_protocols_jll = v"1.18.0+4"
- Colors = v"0.12.7"
- IndirectArrays = v"0.5.1"
- PyPlot = v"2.9.0"
- Profile = nothing
- PDMats = v"0.10.1"
- WAV = v"1.1.0"
- LibGit2 = nothing
- PyCall = v"1.92.2"
- VectorizationBase = v"0.16.2"
- Graphene_jll = v"1.10.0+2"
- IfElse = v"0.1.0"
- OrderedCollections = v"1.4.0"
- TimerOutputs = v"0.5.8"
- LibVPX_jll = v"1.9.0+1"
- Tables = v"1.4.1"
- StatsMakie = v"0.2.3"
- OpenSSL_jll = v"1.1.1+6"
- ImageCore = v"0.8.21"
- VersionParsing = v"1.2.0"
- Ogg_jll = v"1.3.4+2"
- CairoMakie = v"0.3.4"
- Juno = v"0.8.4"
- ColorVectorSpace = v"0.8.7"
- LeftChildRightSiblingTrees = v"0.1.2"
- CodeTools = v"0.7.1"
- MakieLayout = v"0.9.10"
- LNR = v"0.2.1"
- Libepoxy_jll = v"1.5.4+1"
- libvorbis_jll = v"1.3.6+6"
- RecursiveFactorization = v"0.1.11"
- MosaicViews = v"0.2.4"
- GPUCompiler = v"0.8.3"
- VertexSafeGraphs = v"0.1.2"
- Xorg_xkbcomp_jll = v"1.4.2+4"
- Combinatorics = v"1.0.2"
- Zlib_jll = v"1.2.11+18"
- Markdown = nothing
- Xorg_libXcomposite_jll = v"0.4.5+4"
- NNlib = v"0.7.11"
- Formatting = v"0.4.2"
- GTK3_jll = v"3.24.11+5"
- Adapt = v"3.2.0"
- Opus_jll = v"1.3.1+3"
- Plots = v"1.11.2"
- SharedArrays = nothing
- Dbus_jll = v"1.12.16+3"
- ProgressMeter = v"1.5.0"
- Pidfile = v"1.2.0"
- Missings = v"0.4.5"
- Xorg_libXext_jll = v"1.3.4+4"
- AssetRegistry = v"0.1.0"
- StructArrays = v"0.4.2"
- SpecialFunctions = v"0.10.3"
- Observables = v"0.3.3"
- Libtiff_jll = v"4.1.0+2"
- RandomExtensions = v"0.4.3"
- FuzzyCompletions = v"0.4.1"
- DocSeeker = v"0.4.3"
- Blosc_jll = v"1.14.3+1"
- IOCapture = v"0.1.1"
- ModelingToolkit = v"3.20.1"
- GeneralizedGenerated = v"0.2.8"
- FunctionWrappers = v"1.1.2"
- Xorg_libXinerama_jll = v"1.1.4+4"
- FunctionalCollections = v"0.5.0"
- MacroTools = v"0.5.6"
- EzXML = v"1.1.0"
- ImageIO = v"0.2.0"
- DiffEqCallbacks = v"2.13.5"
- WoodburyMatrices = v"0.5.3"
- WebIO = v"0.8.15"
- NameResolution = v"0.1.5"
- Test = nothing
- PkgTemplates = v"0.7.16"
- hicolor_icon_theme_jll = v"0.17.0+3"
- Hexagons = v"0.2.0"
- DataFrames = v"0.21.8"
- BFloat16s = v"0.1.0"
- JuliaInterpreter = v"0.8.11"
- ThreadingUtilities = v"0.2.3"
- FixedPointNumbers = v"0.8.4"
- OrdinaryDiffEq = v"5.39.1"
- LZO_jll = v"2.10.0+3"
- FFMPEG_jll = v"4.3.1+4"
- SHA = nothing
- ExprTools = v"0.1.3"
- TableTraitsUtils = v"1.0.1"
- at_spi2_atk_jll = v"2.34.1+4"
- xkbcommon_jll = v"0.9.1+5"
- Hwloc = v"1.3.0"
- RecursiveArrayTools = v"2.11.1"
- LoopVectorization = v"0.10.0"
- Xorg_libXfixes_jll = v"5.0.3+4"
- Xorg_libXcursor_jll = v"1.2.0+4"
- Parsers = v"1.1.0"
- FreeTypeAbstraction = v"0.8.4"
- ZygoteRules = v"0.2.1"
- FlameGraphs = v"0.2.5"
- Serialization = nothing
- ForwardDiff = v"0.10.17"
- MLStyle = v"0.4.9"
- Xorg_libXdmcp_jll = v"1.1.3+4"
- Grisu = v"1.0.0"
- Fontconfig_jll = v"2.13.1+14"
- libpng_jll = v"1.6.37+6"
- Qt5Base_jll = v"5.15.2+0"
- PositiveFactorizations = v"0.2.4"
- Hiccup = v"0.2.2"
- Xorg_xtrans_jll = v"1.4.0+3"
- FillArrays = v"0.8.14"
- QuadGK = v"2.4.1"
- Zstd_jll = v"1.4.8+0"
- MbedTLS = v"1.0.3"
- AbstractFFTs = v"1.0.1"
- JLLWrappers = v"1.2.0"
- adwaita_icon_theme_jll = v"3.33.92+5"
- ExponentialUtilities = v"1.8.1"
- Future = nothing
- Artifacts = v"1.3.0"
- Xorg_xcb_util_jll = v"0.4.0+1"
- PaddedViews = v"0.5.8"
- Mmap = nothing
- ParameterizedFunctions = v"5.6.0"
- SignedDistanceFields = v"0.4.0"
- FFMPEG = v"0.3.0"
- Random123 = v"1.3.1"
- Sockets = nothing
- Xorg_libXrandr_jll = v"1.5.2+4"
- SparseDiffTools = v"1.13.0"
- Lazy = v"0.15.1"
- ShiftedArrays = v"1.0.0"
- LazyArtifacts = v"1.3.0"
- CEnum = v"0.4.1"
- DataAPI = v"1.6.0"
- Statistics = nothing
- Cairo_jll = v"1.16.0+6"
- DiffRules = v"1.0.2"
- CompilerSupportLibraries_jll = v"0.3.4+0"
- METIS_jll = v"5.1.0+5"
- DataValues = v"0.4.13"
- NLsolve = v"4.5.1"
- AxisAlgorithms = v"1.0.0"
- Expat_jll = v"2.2.7+6"
- DiffEqFinancial = v"2.4.0"
- Sundials_jll = v"5.2.0+1"
- CategoricalArrays = v"0.8.3"
- AbstractTrees = v"0.3.4"
- GenericSVD = v"0.3.0"
- Contour = v"0.5.7"
- Logging = nothing
- AuditoryFilters = v"0.1.0"
- PrettyPrint = v"0.2.0"
- LoggingExtras = v"0.4.6"
- ArnoldiMethod = v"0.1.0"
- MultiScaleArrays = v"1.8.1"
- GR_jll = v"0.57.1+0"
- SteadyStateDiffEq = v"1.6.2"
- CanonicalTraits = v"0.2.4"
- libfdk_aac_jll = v"0.1.6+4"
- Dates = nothing
- ChainRulesCore = v"0.9.37"
- LAME_jll = v"3.100.0+3"
- Showoff = v"0.3.2"
- gdk_pixbuf_jll = v"2.38.2+9"
- Blosc = v"0.7.0"
- FFTW_jll = v"3.3.9+7"
- Optim = v"0.20.1"
- HTTP = v"0.8.19"
- Pkg = nothing
- Xorg_libxkbfile_jll = v"1.1.0+4"
- Interpolations = v"0.12.10"
- FoldingTrees = v"1.0.1"
- GLFW_jll = v"3.3.3+0"
- Match = v"1.1.0"
- SLEEFPirates = v"0.6.8"
- Lz4_jll = v"1.9.2+2"
- LightGraphs = v"1.3.5"
- HDF5_jll = v"1.12.0+1"
- Xorg_libX11_jll = v"1.6.9+4"
- DataStructures = v"0.17.20"
- Rmath_jll = v"0.2.2+2"
- StatsFuns = v"0.9.7"
- DataFramesMeta = v"0.6.0"
- FiniteDiff = v"2.8.0"
- Libffi_jll = v"3.2.1+4"
- Compat = v"2.2.1"
- Documenter = v"0.26.3"
- Conda = v"1.5.1"
- REPL = nothing
- ATK_jll = v"2.34.1+5"
- Xorg_libXdamage_jll = v"1.1.5+4"
- LoweredCodeUtils = v"1.2.9"
- AbstractPlotting = v"0.11.2"
- FileWatching = nothing
- MKL_jll = v"2021.1.1+1"
- DocStringExtensions = v"0.8.4"
- Animations = v"0.4.1"
- Random = nothing
- DimensionalPlotRecipes = v"1.2.0"
- OpenSpecFun_jll = v"0.5.3+4"
- Libglvnd_jll = v"1.3.0+3"
- Distributions = v"0.23.8"
- Libdl = nothing
- BandedMatrices = v"0.15.15"
- Xorg_xcb_util_keysyms_jll = v"0.4.0+1"
- OpenBLAS_jll = v"0.3.9+5"
- PooledArrays = v"0.5.3"
- WebSockets = v"1.5.9"
- MFCC = v"0.3.1"
- DiffEqPhysics = v"3.9.0"
- Xorg_xcb_util_renderutil_jll = v"0.3.9+1"
- Xorg_libXrender_jll = v"0.9.10+4"
- StimAnaGen = v"0.1.0"
- DelimitedFiles = nothing
- ResettableStacks = v"1.1.0"
- SparseArrays = nothing
- Ratios = v"0.4.0"
- InvertedIndices = v"1.0.0"
- Graphite2_jll = v"1.3.13+4"
- PCRE_jll = v"8.42.0+4"
- Roots = v"1.0.8"
- UnPack = v"1.0.2"
- Intervals = v"1.5.0"
- SimpleTraits = v"0.9.3"
- LaTeXStrings = v"1.2.1"
- PyPeriModels = v"0.1.0"
- Measures = v"0.3.1"
- Glib_jll = v"2.59.0+4"
- GLMakie = v"0.1.5"
- Xorg_libXau_jll = v"1.0.9+4"
- GridLayoutBase = v"0.3.7"
- x264_jll = v"2020.7.14+2"
- DifferentialEquations = v"6.16.0"
- PlotUtils = v"1.0.10"
- Libuuid_jll = v"2.34.0+7"
- Atom = v"0.12.30"
- StringDistances = v"0.10.0"
- RecipesBase = v"1.1.1"
- GLFW = v"3.4.1"
- DelayDiffEq = v"5.24.1"
- XSLT_jll = v"1.1.33+4"
- LabelledArrays = v"1.6.0"
- Mustache = v"1.0.10"
- Cthulhu = v"1.6.1"
- Pandas = v"1.4.0"
- JpegTurbo_jll = v"2.0.1+3"
- ImageMagick_jll = v"6.9.10-12+3"
- ConsoleProgressMonitor = v"0.1.2"
- Revise = v"3.1.14"
- SafeTestsets = v"0.0.1"
- XML2_jll = v"2.9.10+3"
- Libgcrypt_jll = v"1.8.5+4"
- DSP = v"0.6.10"
- DataValueInterfaces = v"1.0.0"
- Crayons = v"4.0.4"
- AbstractAlgebra = v"0.11.2"
- ColorSchemes = v"3.11.0"
- SymbolicUtils = v"0.5.2"
- SciMLBase = v"1.10.1"
- Libgpg_error_jll = v"1.36.0+3"
- DiffResults = v"1.0.3"
- JuliaFormatter = v"0.12.3"
- Unitful = v"1.7.0"
- FreeType = v"3.0.1"
- IniFile = v"0.5.0"
- x265_jll = v"3.0.0+3"
- TableTraits = v"1.0.1"
- LinearAlgebra = nothing
- MeshIO = v"0.4.0"
- Pixman_jll = v"0.40.0+0"
- libass_jll = v"0.14.0+4"
- FFTW = v"1.3.2"
- IterativeSolvers = v"0.8.5"
- GR = v"0.57.1"
- Xorg_xkeyboard_config_jll = v"2.27.0+4"
- UnicodeFun = v"0.4.1"
- Packing = v"0.4.1"
- PNGFiles = v"0.2.1"
- Sundials = v"4.2.6"
- TimeZones = v"1.5.3"
- Xorg_xcb_util_image_jll = v"0.4.0+1"
- CommonMark = v"0.6.4"
- InteractiveUtils = nothing
- StatsModels = v"0.6.21"
- Polynomials = v"1.2.1"
- Distributed = nothing
- RecipesPipeline = v"0.3.2"
- SuiteSparse = nothing
- SortingAlgorithms = v"0.3.1"
- CSTParser = v"2.5.0"
- nghttp2_jll = v"1.40.0+2"
- JLD = v"0.10.0"
- KernelDensity = v"0.5.1"
- DiffEqNoiseProcess = v"5.5.2"
- Wayland_jll = v"1.17.0+4"
- TerminalLoggers = v"0.1.3"
- OffsetArrays = v"1.6.2"
- Reexport = v"0.2.0"
- JSON = v"0.21.1"
- ColorTypes = v"0.10.12"
- Inflate = v"0.1.2"
- Tokenize = v"0.5.15"
- Xorg_libXi_jll = v"1.7.10+4"
- StaticArrays = v"0.12.5"
- Media = v"0.5.0"
- EllipsisNotation = v"1.0.0"
- StatsBase = v"0.32.2"
- URIParser = v"0.4.1"
- iso_codes_jll = v"4.3.0+4"
- DiffEqDiffTools = v"1.7.0"
- ArrayInterface = v"2.14.17"
- Xorg_libpthread_stubs_jll = v"0.1.0+3"
- Unicode = nothing
- LineSearches = v"7.1.1"
- ConstructionBase = v"1.1.0"
- Pango_jll = v"1.42.4+10"
- CoupledFields = v"0.2.0"
- MuladdMacro = v"0.2.2"
- BoundaryValueDiffEq = v"2.7.1"
- BenchmarkTools = v"0.7.0"
- Wavelets = v"0.9.2"
- at_spi2_core_jll = v"2.34.0+4"
- StructTypes = v"1.5.2"
- Compose = v"0.9.2"
- Libmount_jll = v"2.34.0+3"
- ColorBrewer = v"0.4.0"
- Rmath = v"0.6.1"
- Loess = v"0.5.3"
- CUDA = v"2.4.1"
- ICU_jll = v"67.1.0+3"
- LibCURL_jll = v"7.70.0+2"
- FreeType2_jll = v"2.10.1+5"
- MappedArrays = v"0.3.0"
- StochasticDiffEq = v"6.25.0"
- Xorg_libxcb_jll = v"1.13.0+3"
- ModernGL = v"1.1.2"
- Gettext_jll = v"0.20.1+7"
- ShaderAbstractions = v"0.2.5"
- Distances = v"0.10.2"
- SuiteSparse_jll = v"5.4.0+9"
- FriBidi_jll = v"1.0.5+6"
- Cairo = v"1.0.5"
- MbedTLS_jll = v"2.16.8+1"
- Scratch = v"1.0.3"
- LLVM = v"3.6.0"
- Mocking = v"0.7.1"
- IntervalSets = v"0.5.3"
- LibSSH2_jll = v"1.9.0+3"
- Widgets = v"0.6.2"
- DiffEqBase = v"6.44.3"
- TreeViews = v"0.3.0"
- Hwloc_jll = v"2.4.1+0"
- UUIDs = nothing
- PoissonRandom = v"0.4.0"
- IntelOpenMP_jll = v"2018.0.3+2"
- NaNMath = v"0.3.5"
- ImageMagick = v"1.2.0"
- Latexify = v"0.14.12"
- Calculus = v"0.5.1"
- Xorg_xcb_util_wm_jll = v"0.4.1+1"
- IterTools = v"1.3.0"
- Parameters = v"0.12.2"
- Bzip2_jll = v"1.0.6+5"
- HDF5 = v"0.13.6"
- HarfBuzz_jll = v"2.6.1+10"
- Gadfly = v"1.3.2"
- PlotThemes = v"2.0.1"
- Printf = nothing
- DiffEqJump = v"6.10.0"
- ProgressLogging = v"0.1.4"
- RandomNumbers = v"1.4.0"

Julia Version 1.5.3
Commit 788b2c77c1 (2020-11-09 13:37 UTC)
Platform Info:
OS: Linux (x86_64-pc-linux-gnu)
CPU: Intel(R) Core(TM) i7-6900K CPU @ 3.20GHz
WORD_SIZE: 64
LIBM: libopenlibm
LLVM: libLLVM-9.0.1 (ORCJIT, broadwell)

Stacktrace:
[1] check_ir!(::GPUCompiler.CompilerJob{GPUCompiler.PTXCompilerTarget,CUDA.CUDACompilerParams}, ::Array{Tuple{String,Array{Base.StackTraces.StackFrame,1},Any},1}, ::LLVM.CallInst) at /home/dahlbom/.julia/packages/GPUCompiler/uTpNx/src/validation.jl:246
[2] check_ir!(::GPUCompiler.CompilerJob{GPUCompiler.PTXCompilerTarget,CUDA.CUDACompilerParams}, ::Array{Tuple{String,Array{Base.StackTraces.StackFrame,1},Any},1}, ::LLVM.Function) at /home/dahlbom/.julia/packages/GPUCompiler/uTpNx/src/validation.jl:140
[3] check_ir!(::GPUCompiler.CompilerJob{GPUCompiler.PTXCompilerTarget,CUDA.CUDACompilerParams}, ::Array{Tuple{String,Array{Base.StackTraces.StackFrame,1},Any},1}, ::LLVM.Module) at /home/dahlbom/.julia/packages/GPUCompiler/uTpNx/src/validation.jl:131
[4] check_ir(::GPUCompiler.CompilerJob{GPUCompiler.PTXCompilerTarget,CUDA.CUDACompilerParams}, ::LLVM.Module) at /home/dahlbom/.julia/packages/GPUCompiler/uTpNx/src/validation.jl:120
[5] macro expansion at /home/dahlbom/.julia/packages/GPUCompiler/uTpNx/src/driver.jl:239 [inlined]
[6] macro expansion at /home/dahlbom/.julia/packages/TimerOutputs/4QAIk/src/TimerOutput.jl:206 [inlined]
[7] codegen(::Symbol, ::GPUCompiler.CompilerJob; libraries::Bool, deferred_codegen::Bool, optimize::Bool, strip::Bool, validate::Bool, only_entry::Bool) at /home/dahlbom/.julia/packages/GPUCompiler/uTpNx/src/driver.jl:237
[8] compile(::Symbol, ::GPUCompiler.CompilerJob; libraries::Bool, deferred_codegen::Bool, optimize::Bool, strip::Bool, validate::Bool, only_entry::Bool) at /home/dahlbom/.julia/packages/GPUCompiler/uTpNx/src/driver.jl:39
[9] compile at /home/dahlbom/.julia/packages/GPUCompiler/uTpNx/src/driver.jl:35 [inlined]
[10] cufunction_compile(::GPUCompiler.FunctionSpec; kwargs::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}) at /home/dahlbom/.julia/packages/CUDA/wTQsK/src/compiler/execution.jl:302
[11] cufunction_compile(::GPUCompiler.FunctionSpec) at /home/dahlbom/.julia/packages/CUDA/wTQsK/src/compiler/execution.jl:297
[12] check_cache(::Dict{UInt64,Any}, ::Any, ::Any, ::GPUCompiler.FunctionSpec{GPUArrays.var"#broadcast_kernel#14",Tuple{CUDA.CuKernelContext,CuDeviceArray{Nothing,1,1},Base.Broadcast.Broadcasted{Nothing,Tuple{Base.OneTo{Int64}},Main.MinimalFailure.var"#1#2",Tuple{Base.Broadcast.Extruded{CuDeviceArray{Int64,1,1},Tuple{Bool},Tuple{Int64}}}},Int64}}, ::UInt64; kwargs::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}) at /home/dahlbom/.julia/packages/GPUCompiler/uTpNx/src/cache.jl:40
[13] broadcast_kernel at /home/dahlbom/.julia/packages/GPUArrays/bjw3g/src/host/broadcast.jl:57 [inlined]
[14] cached_compilation at /home/dahlbom/.julia/packages/GPUCompiler/uTpNx/src/cache.jl:65 [inlined]
[15] cufunction(::GPUArrays.var"#broadcast_kernel#14", ::Type{Tuple{CUDA.CuKernelContext,CuDeviceArray{Nothing,1,1},Base.Broadcast.Broadcasted{Nothing,Tuple{Base.OneTo{Int64}},Main.MinimalFailure.var"#1#2",Tuple{Base.Broadcast.Extruded{CuDeviceArray{Int64,1,1},Tuple{Bool},Tuple{Int64}}}},Int64}}; name::Nothing, kwargs::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}) at /home/dahlbom/.julia/packages/CUDA/wTQsK/src/compiler/execution.jl:289
[16] cufunction at /home/dahlbom/.julia/packages/CUDA/wTQsK/src/compiler/execution.jl:286 [inlined]
[17] macro expansion at /home/dahlbom/.julia/packages/CUDA/wTQsK/src/compiler/execution.jl:100 [inlined]
[18] #launch_heuristic#857 at /home/dahlbom/.julia/packages/CUDA/wTQsK/src/gpuarrays.jl:17 [inlined]
[19] launch_heuristic at /home/dahlbom/.julia/packages/CUDA/wTQsK/src/gpuarrays.jl:17 [inlined]
[20] copyto! at /home/dahlbom/.julia/packages/GPUArrays/bjw3g/src/host/broadcast.jl:63 [inlined]
[21] copyto! at ./broadcast.jl:886 [inlined]
[22] copy at /home/dahlbom/.julia/packages/GPUArrays/bjw3g/src/host/broadcast.jl:47 [inlined]
[23] materialize(::Base.Broadcast.Broadcasted{CUDA.CuArrayStyle{1},Nothing,Main.MinimalFailure.var"#1#2",Tuple{CuArray{Int64,1}}}) at ./broadcast.jl:837
[24] map(::Function, ::CuArray{Int64,1}) at /home/dahlbom/.julia/packages/GPUArrays/bjw3g/src/host/broadcast.jl:86
[25] top-level scope at /home/dahlbom/research/jcuda/MinimalFailure.jl:13
[26] include(::String) at ./client.jl:457
[27] top-level scope at REPL[4]:1
in expression starting at /home/dahlbom/research/jcuda/MinimalFailure.jl:13

```

Contributor guide

No contributing guide indexed for this repository

Research direction

Reproduce the minimal map-over-CuArray example using the reported CUDA v2.4.1 and GPUCompiler v0.8.3 context. Start by inspecting GPUCompiler's validation.jl at line 246 and the reported PTX compiler invocation; done means this input no longer produces an unexpected internal compiler error.

Written by the indexing model from the issue text.

Assessment

Tech stack
julia
Domain
backend, compilers
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.