Skip to content
This repository has been archived by the owner on Mar 12, 2021. It is now read-only.

CUBLAS initialization #697

Closed
erathorn opened this issue Apr 29, 2020 · 1 comment
Closed

CUBLAS initialization #697

erathorn opened this issue Apr 29, 2020 · 1 comment
Labels

Comments

@erathorn
Copy link

Thank you for creating this amazing library.

However, I encounter a problem with CUBLAS. Using Matrix multiplication does not work for me. I get the following problem:

using CuArrays

CuArrays.seed!(42)

A = CuArrays.rand(5,5)

A*A
ERROR: CUBLASError: the library was not initialized (code 1, CUBLAS_STATUS_NOT_INITIALIZED)
Stacktrace:
 [1] throw_api_error(::CuArrays.CUBLAS.cublasStatus_t) at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/error.jl:52
 [2] macro expansion at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/error.jl:65 [inlined]
 [3] cublasCreate_v2 at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/libcublas.jl:9 [inlined]
 [4] cublasCreate_v2 at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/wrappers.jl:8 [inlined]
 [5] (::CuArrays.CUBLAS.var"#354#356"{CUDAdrv.CuContext})() at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/CUBLAS.jl:40
 [6] get! at ./abstractdict.jl:661 [inlined]
 [7] handle() at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/CUBLAS.jl:39
 [8] top-level scope at REPL[3]:1

I tried to search for the error, and I ended up here:

julia> using CuArrays

julia> CuArrays.CUBLAS.handle()
ERROR: CUBLASError: the library was not initialized (code 1, CUBLAS_STATUS_NOT_INITIALIZED)
Stacktrace:
 [1] throw_api_error(::CuArrays.CUBLAS.cublasStatus_t) at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/error.jl:52
 [2] macro expansion at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/error.jl:65 [inlined]
 [3] cublasCreate_v2 at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/libcublas.jl:9 [inlined]
 [4] cublasCreate_v2 at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/wrappers.jl:8 [inlined]
 [5] (::CuArrays.CUBLAS.var"#354#356"{CUDAdrv.CuContext})() at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/CUBLAS.jl:40
 [6] get! at ./abstractdict.jl:661 [inlined]
 [7] handle() at /home/jo/.julia/packages/CuArrays/1njKF/src/blas/CUBLAS.jl:39
 [8] top-level scope at REPL[3]:1

Here is all the information about the julia version and the packages:

Julia Version 1.3.1
Commit 2d5741174c (2019-12-30 21:36 UTC)
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: AMD Ryzen 9 3900X 12-Core Processor
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-6.0.1 (ORCJIT, znver1)
Environment:
  JULIA_NUM_THREADS = 4
nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Sun_Jul_28_19:07:16_PDT_2019
Cuda compilation tools, release 10.1, V10.1.243
  [b19378d9] ArrayFire v1.0.6
  [c52e3926] Atom v0.12.8
  [6e4b80f9] BenchmarkTools v0.5.0
  [336ed68f] CSV v0.5.22
  [3895d2a7] CUDAapi v2.1.0
  [c5f51814] CUDAdrv v5.0.1
  [be33ccc6] CUDAnative v2.8.1
  [49dc2e85] Calculus v0.5.1
  [a81c6b42] Compose v0.8.0
  [3a865a2d] CuArrays v1.7.0
  [a93c6f00] DataFrames v0.20.0
  [31c24e10] Distributions v0.22.3
  [587475ba] Flux v0.10.1
  [f6369f11] ForwardDiff v0.10.9
  [0c68f7d7] GPUArrays v2.0.1
  [c91e804a] Gadfly v1.0.1
  [e5e0dc1b] Juno v0.7.2
  [093fc24a] LightGraphs v1.3.0
  [2774e3e8] NLsolve v4.3.0
  [90014a1f] PDMats v0.9.11
  [189a3867] Reexport v0.2.0
  [992d4aef] Showoff v0.3.1
  [276daf66] SpecialFunctions v0.10.0
  [2913bbd2] StatsBase v0.32.0
  [4c63d2b9] StatsFuns v0.9.4
  [e88e6eb3] Zygote v0.4.15
  [8ba89e20] Distributed
  [37e2e46d] LinearAlgebra
  [d6f4376e] Markdown
  [9a3f8284] Random
  [10745b16] Statistics
@erathorn erathorn added the bug label Apr 29, 2020
@erathorn
Copy link
Author

The root of this issue was a broken CUDA installation. I updated reinstalled Cuda and solved the problem. Sorry for the noise.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
Projects
None yet
Development

No branches or pull requests

1 participant