v2.9.0

LinearSolve v2.9.0

Diff since v2.8.1

Merged pull requests:

• Make AppleAccelerate the default algorithm when available (#386) (@ChrisRackauckas)
• Make MKL the default when it's available (#387) (@ChrisRackauckas)
• Fix GPU tests (#388) (@ChrisRackauckas)

v2.8.1

LinearSolve v2.8.1

Diff since v2.8.0

v2.8.0

LinearSolve v2.8.0

Diff since v2.7.0

Merged pull requests:

• try documenter 1.0 upgrade (#371) (@ArnoStrouwen)
• Add Enzyme extension (#377) (@wsmoses)
• [skip ci] change master to main FormatCheck.yml (#379) (@ArnoStrouwen)

v2.7.0

LinearSolve v2.7.0

Diff since v2.6.0

v2.6.0

LinearSolve v2.6.0

Diff since v2.5.1

Merged pull requests:

• Add SimpleGMRES implementation (#366) (@avik-pal)
• Bump actions/checkout from 3 to 4 (#368) (@dependabot[bot])
• Recompile invalidations post LoopVectorization (#374) (@ChrisRackauckas)
• Update Project.toml (#376) (@avik-pal)

v2.5.1

LinearSolve v2.5.1

Diff since v2.5.0

Closed issues:

• function overwritten warning (#353)

Merged pull requests:

• Consolidate init (#360) (@ChrisRackauckas)
• Setup Accelerate and MKL for 32-bit, MKL getrf, and Metal.jl integration (#361) (@ChrisRackauckas)
• Add sparselu benchmark (#365) (@j-fu)
• Fix setting of Pardiso iparm and dparm parameters (#367) (@BeatHubmann)

v2.5.0

LinearSolve v2.5.0

Diff since v2.4.2

Closed issues:

• use AppleAccelerate as default on Mac (#354)

Merged pull requests:

• Add missing backticks in Pardiso docstring (#344) (@Vaibhavdixit02)
• Setup MKL direct factorizations (#349) (@ChrisRackauckas)
• Support Apple Accelerate and improve MKL integration (#355) (@ChrisRackauckas)
• Add a standard LU performance benchmark (#356) (@ChrisRackauckas)

v2.4.2

LinearSolve v2.4.2

Diff since v2.4.1

Closed issues:

• All extension packages behind requires for support in <1.9? (#346)
• KrylovJL solvers returning incorrect answer when u0 is a SubArray (#347)

Merged pull requests:

• Update basictests.jl (#345) (@vpuri3)
• Fiox solution handling for KrylovJL on subarrays (#348) (@ChrisRackauckas)

v2.4.1

LinearSolve v2.4.1

Diff since v2.4.0

v2.4.0

LinearSolve v2.4.0

Diff since v2.3.0

Closed issues:

• LinearSolve.solve much slower than A\b for sparse matrices (#340)
• Solving a system of linear equations LinearSolve.solve uses up more memory when A is sparse then when A is dense (#342)

Merged pull requests:

• Set a sparsity threshold for KLU (#343) (@ChrisRackauckas)