MKL_VML: Not a F90 module -- oversight, or informed choice?

[ereisch]

I'm using routines out of the VML (VCCONJ, VSSINCOS), but it appears as though, unlike most other MKL interfaces, the MKL_VML group of routines are not organized into an F90 module.  Was this an informed programming decision, or simply an oversight?  I can't wrap a module around it and/or include this file in my subprogram either, because the syntax rules are different between my fixed-form source file that is run through the preprocessor, and the free-form F90 source of the mkl_vml.f90 interface definition file.

If this was an oversight, can I request the next version of MKL (I'm using 2023.1, which I know is out of date) wrap this into a module so I can just USE MKL_VML in my subprogram?  If it was a deliberate choice, what is the "proper" way of accessing these functions in a fixed-form Fortran subprogram?

Thanks in advance.

[mecej4]

Simple solution: create a free form source file with three lines:

module mkl_vml
   include mkl_vml.f90
end module

Give the file a name ending with .f90 . Compiling this file will generate the module file mkl_vml.mod . 

[ShanmukhS_Intel]

Hi Eric,

Thanks for posting in Intel Communities.

Has the information provided earlier helped? Could you please let us know if you need any other information?

Best Regards,

Shanmukh.SS

[ereisch]

This wasn't really an issue, per se; I was just bringing to light that this one include file was not a module, whereas most/all of the other MKL include files were.  Thanks.

[ShanmukhS_Intel]

Hi,

Thanks for helping us improve our products! We’ve submitted the feature request to the dev team, they will consider it based on multiple factors including, but not limited to priority and criticality of the feature. Once it is included in an upcoming release, it would be documented in the release notes.

Best Regards,

Shanmukh.SS