v0.2.0

KrylovPreconditioners v0.2.0

Diff since v0.1.0

Merged pull requests:

• Use in-place sparse triangular solves for CUDA.jl (#24) (@amontoison)
• Add a KrylovOperator for sparse matrices on GPUs (#25) (@amontoison)
• Fix leak in block-Jacobi (#26) (@michel2323)
• Add a function update_operator! (#27) (@amontoison)
• Update Project.toml (#28) (@amontoison)
• Add timer and update API (#29) (@michel2323)
• Update IC(0) and ILU(0) for CUDA GPUs (#30) (@amontoison)
• Update gpu.jl (#32) (@amontoison)
• [GPU] Update IC(0) and ILU(0) for AMD GPUs (#35) (@amontoison)
• Rename update_operator! into update! (#36) (@amontoison)
• Fix update! on CPU (#37) (@amontoison)
• Adding a scaling (#38) (@michel2323)

Closed issues:

• [GPU] Create linear operators to optimize sparse matrix-vector and matrix-matrix products (#22)
• Tests fail on AMDGPU (#31)
• [GPU] Update IC(0) and ILU(0) for AMD GPUs (#33)