MKL.jl

Using Julia with Intel's MKL

MKL.jl is a Julia package that allows users to use the Intel MKL library for Julia's underlying BLAS and LAPACK, instead of OpenBLAS, which Julia ships with by default. On Julia v1.7 and later, Julia includes libblastrampoline, which enables picking a BLAS and LAPACK library at runtime. A JuliaCon 2021 talk provides details on this mechanism.

On Julia 1.6 and earlier, adding this package will rebuild Julia's system image with MKL support built in.

for users cannot access github raw content: if you find you have trouble downloading/installing/precompiling this package on 1.6 or earlier, it may because the internet policy blocks github raw content. We strongly suggest you to upgrade your Julia to 1.7 or later versions to fix this problem. (see also #93)

无法下载GitHub文件的用户: 如果您发现无法在1.6及更早的版本的Julia中安装这个包，有可能是因为一部分编译脚本被墙了，我们强烈建议您升级到Julia 1.7或者之后的版本以修复这个问题。(参见 #93)

Usage

If you want to use MKL.jl in your project, make sure it is the first package you load before any other package. It is essential that MKL be loaded before other packages so that it can find the Intel OMP library and avoid issues resulting out of GNU OMP being loaded first.

To Install:

On Julia 1.7 and later, adding the package will replace the system BLAS and LAPACK with MKL provided ones at runtime. Note that the MKL package has to be loaded in every new Julia process. Upon quitting and restarting, Julia will start with the default OpenBLAS.

julia> using Pkg; Pkg.add("MKL")

On 1.6 and earlier, a new system image is built upon installing this package. If it doesn't happen successfully (happens when MKL_jll.jl has been installed before), run the following command to build a new system image and restart Julia.

julia> using Pkg; Pkg.build("MKL")

To Check Installation:

On Julia 1.7 and later:

julia> using LinearAlgebra

julia> BLAS.get_config()
LinearAlgebra.BLAS.LBTConfig
Libraries: 
└ [ILP64] libopenblas64_.0.3.13.dylib

julia> using MKL

julia> BLAS.get_config()
LinearAlgebra.BLAS.LBTConfig
Libraries: 
└ [ILP64] libmkl_rt.1.dylib

On Julia 1.6 and earlier:

julia> BLAS.vendor()
:mkl

Using the 64-bit vs 32-bit version of MKL

On Julia v1.7 and later, we use ILP64 by default on 64-bit systems, and LP64 on 32-bit systems.

On Julia 1.6 and earlier, when building MKL.jl, the 32-bit version of MKL is installed. This is due to frequently encountered compatibility issues with the MKL version linked to numpy, which by default is shipped with the 32-bit version of MKL. To use the 64-bit version of MKL, set the environment variable ENV["USE_BLAS64"] = true before building MKL.jl.