Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
Announcements
Welcome to the Intel Community. If you get an answer you like, please mark it as an Accepted Solution to help others. Thank you!
For the latest information on Intel’s response to the Log4j/Log4Shell vulnerability, please see Intel-SA-00646

MKL FATAL Error - Error on loading function mkl_vml_serv_threader_c_1i_2o.

des_c_
Beginner
354 Views

Hi, i am trying out the latest intel python with TBB. i tried 'python -m TBB test.py' which test is a program with lots of PANDAS dataframe and random forest regressors. Any advice what should i be looking at? This is on windows 7 64 bits

0 Kudos
4 Replies
Oleksandr_P_Intel
354 Views

Hi,

Could you please confirm whether your program uses NumExpr? Additionally, please provide the output of  ``conda list mkl``

Thank you,
Oleksandr

des_c_
Beginner
354 Views

hi, yes i use a lot of scikit-learn, statsmodels and scipy, not sure NumExpr is involve but i dont use it directly.

for mkl

mkl                       2017.0.3                intel_5  [intel]  intel

numpy config as such: -

lapack_mkl_info:
    define_macros = [('SCIPY_MKL_H', None), ('HAVE_CBLAS', None)]
    library_dirs = ['c:/IntelPython3\\Library\\lib']
    include_dirs = ['c:/IntelPython3\\Library\\include']
    libraries = ['mkl_core_dll', 'mkl_intel_lp64_dll', 'mkl_intel_thread_dll']
mkl_info:
    define_macros = [('SCIPY_MKL_H', None), ('HAVE_CBLAS', None)]
    library_dirs = ['c:/IntelPython3\\Library\\lib']
    include_dirs = ['c:/IntelPython3\\Library\\include']
    libraries = ['mkl_core_dll', 'mkl_intel_lp64_dll', 'mkl_intel_thread_dll']
blas_opt_info:
    define_macros = [('SCIPY_MKL_H', None), ('HAVE_CBLAS', None)]
    library_dirs = ['c:/IntelPython3\\Library\\lib']
    include_dirs = ['c:/IntelPython3\\Library\\include']
    libraries = ['mkl_core_dll', 'mkl_intel_lp64_dll', 'mkl_intel_thread_dll']
lapack_opt_info:
    define_macros = [('SCIPY_MKL_H', None), ('HAVE_CBLAS', None)]
    library_dirs = ['c:/IntelPython3\\Library\\lib']
    include_dirs = ['c:/IntelPython3\\Library\\include']
    libraries = ['mkl_core_dll', 'mkl_intel_lp64_dll', 'mkl_intel_thread_dll']
blas_mkl_info:
    define_macros = [('SCIPY_MKL_H', None), ('HAVE_CBLAS', None)]
    library_dirs = ['c:/IntelPython3\\Library\\lib']
    include_dirs = ['c:/IntelPython3\\Library\\include']
    libraries = ['mkl_core_dll', 'mkl_intel_lp64_dll', 'mkl_intel_thread_dll']

Jingwei_Z_Intel
Employee
354 Views

We are able to reproduce the error on our side, and we are working on a fix.

Thank you,

Jingwei Zhang

Oleksandr_P_Intel
354 Views

Hi des, 

The crash is caused by use of vector math library (VML) using TBB as MKL's threading and is specific to Windows only. As Jingwei said, the fix is underway. 

The current work-around is not to use the TBB package on Windows, since it makes MKL dynamic runtime use TBB threading layer.

Per our findings the crash happens through the use of numexpr, but it would be helpful to confirm that on your end. If it possible to share a reproducer with us, please let us know. Otherwise, please provide the output of `conda list --explicit` for the environment where you experience the issue.

It would also be helpful to know if you are using `numpy.random_intel`?

Thank you,
Oleksandr

Reply