Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
Announcements
This community is designed for sharing of public information. Please do not share Intel or third-party confidential information here.
6590 Discussions

MKL fatal error when trying to compile fortran code

Nasim_hossain
Beginner
272 Views

Hello, I am trying to run a CFD code that is written in fortran. I use ifort for compiling all of my fortran codes(Installed one api baseline toolkit and hpc toolkit latest version 2022.2). Previously I ran the same code without any issue. But recently whenever I am trying to run the code I am getting an MKL error. I have included the whole error below:

//

Intel MKL FATAL ERROR: This system does not meet the minimum requirements for use of the Intel(R) Math Kernel Library.
The processor must support the Intel(R) Supplemental Streaming SIMD Extensions 3 (Intel(R) SSSE3) instructions.
The processor must support the Intel(R) Streaming SIMD Extensions 4.2 (Intel(R) SSE4.2) instructions.
The processor must support the Intel(R) Advanced Vector Extensions (Intel(R) AVX) instructions.

//

Previously I ran the code without any issue but now I am wondering what could be the potential cause behind this.

I am using Mac OS Monterey, Version 12.3.1, Chip Apple M1

Thank you in advance.

0 Kudos
6 Replies
Gajanan_Choudhary
255 Views

Hi Nasim,

 

Thanks for reaching out to us about this issue.

I see you mentioned the following:


I am using Mac OS Monterey, Version 12.3.1, Chip Apple M1

 

and


Previously I ran the same code without any issue.

Do you happen to have recently moved to a newer Apple machine from an older one?

Older Macs had Intel processors based on x86/x86_64 architecture/instruction sets; the Intel® oneAPI Math Kernel Library (oneMKL) is optimized for performance on these architectures. If you were using an older Mac with Intel processors, you would be able to use oneMKL on it. Recently released Macs with the Apple M1 chips have an ARMv8.5-A architecture/instruction set. My understanding is that oneMKL does not support this instruction set, and therefore cannot be used with the Apple M1 chip that you are using. If you have recently moved from an older Mac to a newer one, that would explain your situation, and why you are now getting the errors of missing instruction sets.

 

Please note that my knowledge on this topic may be limited, and do wait for a support engineer from the Intel oneMKL team to respond to your question.

Nasim_hossain
Beginner
253 Views

Thank you for your reply. No, I have used the same machine to run my code few months ago(around 3 months). At that time I was using intel one api baseline tool kit 2021 version. Then I updated to 2022 version. This is the only change that I could remember.

mecej4
Black Belt
232 Views

Intel compilers output X64/X86 object code. MKL libraries contain X64/X86 object code. In order to run a.out executables and shared libraries that use X64 instruction sets on ARM cpu-s such as M1, Apple provides a translation/emulation layer called Rosetta<something>. I suspect that Apple changed something in that layer, and the failure that you are seeing is a consequence of such changes.

You should not expect Intel to do anything about this, since this is entirely up to Apple to fix (or do nothing about, as they please).

ShanmukhS_Intel
Moderator
172 Views

Hi,


It seems the OS environment details shared by you doesn't meet the prerequisites for the Intel® oneAPI Math Kernel Library. Please find the system requirements information under below link.


https://www.intel.com/content/www/us/en/developer/articles/system-requirements/oneapi-math-kernel-li...


Best Regards,

Shanmukh.SS 


ShanmukhS_Intel
Moderator
132 Views

Hi,


Reminder: Has the information provided helped? Kindly let us know if we could close this thread at our end.


Best Regards,

Shanmukh.SS


ShanmukhS_Intel
Moderator
108 Views

Hi,


We assume that your issue is resolved. If you need any additional information, please post a new question as this thread will no longer be monitored by Intel.


Best Regards,

Shanmukh.SS


Reply