Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Pinning blas implementation not working?

See original GitHub issue

The documention on pining the blas implementation in https://conda-forge.org/docs/maintainer/knowledge_base.html#switching-blas-implementation doesn’t seem to work:

run conda install "liblapack=*=*mkl"
Add blas=*=mkl to pinned
run conda update --all and the blas implementation switch back to netlib:

Collecting package metadata (current_repodata.json): done
Solving environment: done

## Package Plan ##

  environment location: /opt/miniconda3/envs/blas_issue


The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    _openmp_mutex-4.5          |            1_gnu          22 KB  conda-forge
    libblas-3.9.0              |1_h86c2bf4_netlib         199 KB  conda-forge
    libgcc-ng-9.3.0            |      h5dbcf3e_17         7.8 MB  conda-forge
    libgfortran-ng-9.3.0       |      he4bcb1c_17          22 KB  conda-forge
    libgfortran5-9.3.0         |      he4bcb1c_17         2.0 MB  conda-forge
    libgomp-9.3.0              |      h5dbcf3e_17         378 KB  conda-forge
    liblapack-3.9.0            |1_h86c2bf4_netlib         3.0 MB  conda-forge
    ------------------------------------------------------------
                                           Total:        13.3 MB

The following NEW packages will be INSTALLED:

  libgcc-ng          conda-forge/linux-64::libgcc-ng-9.3.0-h5dbcf3e_17
  libgfortran-ng     conda-forge/linux-64::libgfortran-ng-9.3.0-he4bcb1c_17
  libgfortran5       conda-forge/linux-64::libgfortran5-9.3.0-he4bcb1c_17
  libgomp            conda-forge/linux-64::libgomp-9.3.0-h5dbcf3e_17

The following packages will be REMOVED:

  llvm-openmp-11.0.0-hfc4b9b4_1
  mkl-2020.2-256

The following packages will be UPDATED:

  _openmp_mutex                                  4.5-1_llvm --> 4.5-1_gnu
  libblas                                      3.8.0-16_mkl --> 3.9.0-1_h86c2bf4_netlib
  liblapack                                    3.8.0-16_mkl --> 3.9.0-1_h86c2bf4_netlib

Environment (conda list):

$ conda list
# packages in environment at /opt/miniconda3/envs/blas_issue:
#
# Name                    Version                   Build  Channel
_libgcc_mutex             0.1                 conda_forge    conda-forge
_openmp_mutex             4.5                      1_llvm    conda-forge
blas                      1.0                         mkl  
libblas                   3.8.0                    16_mkl    conda-forge
liblapack                 3.8.0                    16_mkl    conda-forge
llvm-openmp               11.0.0               hfc4b9b4_1    conda-forge
mkl                       2020.2                      256    conda-forge

Details about conda and system ( conda info after step 1):

$ conda info
     active environment : blas_issue
    active env location : /opt/miniconda3/envs/blas_issue
            shell level : 2
       user config file : /home/eric/.condarc
 populated config files : /home/eric/.condarc
          conda version : 4.9.0
    conda-build version : 3.20.4
         python version : 3.7.9.final.0
       virtual packages : __cuda=11.1=0
                          __glibc=2.31=0
                          __unix=0=0
                          __archspec=1=x86_64
       base environment : /opt/miniconda3  (writable)
           channel URLs : https://conda.anaconda.org/conda-forge/linux-64
                          https://conda.anaconda.org/conda-forge/noarch
                          https://repo.anaconda.com/pkgs/main/linux-64
                          https://repo.anaconda.com/pkgs/main/noarch
                          https://repo.anaconda.com/pkgs/r/linux-64
                          https://repo.anaconda.com/pkgs/r/noarch
          package cache : /opt/miniconda3/pkgs
                          /home/eric/.conda/pkgs
       envs directories : /opt/miniconda3/envs
                          /home/eric/.conda/envs
               platform : linux-64
             user-agent : conda/4.9.0 requests/2.24.0 CPython/3.7.9 Linux/5.8.15-201.fc32.x86_64 fedora/32 glibc/2.31
                UID:GID : 1000:1000
             netrc file : None
           offline mode : False