Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add triangular operators for AMD and NVIDIA GPUs #41

Merged
merged 5 commits into from
Dec 7, 2023

Conversation

amontoison
Copy link
Member

No description provided.

@amontoison amontoison force-pushed the triangular_operators branch from fe2b14e to 136292e Compare December 5, 2023 20:11
@amontoison
Copy link
Member Author

@michel2323 Can you test this branch on AMD GPUs?

@michel2323
Copy link
Collaborator

michel2323 commented Dec 7, 2023

I get this error below only in

rocSPARSE.rocsparse_spsv(rocSPARSE.handle(), A.transa, alpha, A.descA, descX, descY, T,
. I looked at the other spsv's and I can't figure out what the problem is. The test for X and Y in line 160 works.

  ROCSPARSEError: (code 3, rocsparse_status_invalid_pointer)
  Stacktrace:
    [1] macro expansion
      @ /lustre/orion/csc359/scratch/mschanen/julia_depot/dev/AMDGPU/src/sparse/error.jl:62 [inlined]
    [2] rocsparse_spsv(handle::Ptr{AMDGPU.rocSPARSE._rocsparse_handle}, trans::Char, alpha::Base.RefValue{Float64}, mat::AMDGPU.rocSPARSE.ROCSparseMatrixDescriptor, x::AMDGPU.rocSPARSE.ROCDenseVectorDescriptor, y::AMDGPU.rocSPARSE.ROCDenseVectorDescriptor, compute_type::Type, alg::AMDGPU.rocSPARSE.rocsparse_spsv_alg_, stage::AMDGPU.rocSPARSE.rocsparse_spsv_stage_, buffer_size::Base.RefValue{UInt64}, temp_buffer::ROCArray{UInt8, 1, AMDGPU.Runtime.Mem.HIPBuffer})
      @ AMDGPU.rocSPARSE /lustre/orion/csc359/scratch/mschanen/julia_depot/dev/AMDGPU/src/sparse/librocsparse.jl:5137
    [3] ldiv!(y::ROCArray{Float64, 1, AMDGPU.Runtime.Mem.HIPBuffer}, A::KrylovPreconditionersAMDGPUExt.AMD_TriangularOperator{Float64}, x::ROCArray{Float64, 1, AMDGPU.Runtime.Mem.HIPBuffer})
      @ KrylovPreconditionersAMDGPUExt /lustre/orion/csc359/scratch/mschanen/git/KrylovPreconditioners.jl/ext/AMDGPU/operators.jl:150

@amontoison amontoison merged commit 2781c1d into main Dec 7, 2023
9 of 10 checks passed
@amontoison amontoison deleted the triangular_operators branch December 7, 2023 17:57
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants