Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.

Is C++ DFT ready to use?

sudoLife
New Contributor I
911 Views

Hi,

 

I am trying to integrate the C++ version of the OneMKL library. Specifically, I need DFT.

I am using this as a reference: https://spec.oneapi.io/versions/1.0-rev-3/elements/oneMKL/source/domains/dft/dft.html

 

However, it seems that the library is incomplete, things that are in the documentation do not exist in reality, like enum config_value, for example. Due to this, it feels unusable.

 

Am I looking at the wrong documentation? Should I use the C interface for full functionality?

 

The installed library version is 2022.0.2, OS is Ubuntu, compiler is dpcpp.

Labels (3)
0 Kudos
1 Solution
ShanmukhS_Intel
Moderator
840 Views

Hi,

 

Regrets for difference in understanding.

 

Kindly refer to the below link for computing a discrete Fourier transform through the fast Fourier transform algorithm with respect to DPC++ interface.

https://www.intel.com/content/www/us/en/develop/documentation/oneapi-mkl-dpcpp-developer-reference/top/fourier-transform-functions.html

 

The DPC++ configuration parameters have type mkl::dft::config_param. Each parameter mkl::dft::config_param::<PARAMETER_NAME> has the same meaning and possible values as the C interface’s DFTI_<PARAMETER_NAME>. Some options corresponding to C interface might not be available on DPC++ interface yet. For a full description of each parameter’s meaning and options, see Config Params.

 

Kindly let us know if you need any details further.

 

Best Regards,

Shanmukh.SS

 

View solution in original post

0 Kudos
6 Replies
ShanmukhS_Intel
Moderator
881 Views

Hi,

 

Thank you for posting on Intel Communities.


>> Am I looking at the wrong documentation?

We would like to inform you to refer the below developer reference link which elaborates the configuration settings of MKL_DFTI module.

Each of the configuration parameters is identified by a named constant in the MKL_DFTI module. These named constants have the enumeration type DFTI_CONFIG_PARAM.

 

https://www.intel.com/content/www/us/en/develop/documentation/onemkl-developer-reference-c/top/fourier-transform-functions/fft-functions/configuration-settings.html

 

Best Regards,

Shanmukh.SS

 

 

 


0 Kudos
sudoLife
New Contributor I
875 Views

Hi,

 

Thank you for the reply. You are referring me to the C API.

I am trying to use the C++ version. Please check my link to understand what I am talking about.

0 Kudos
ShanmukhS_Intel
Moderator
841 Views

Hi,

 

Regrets for difference in understanding.

 

Kindly refer to the below link for computing a discrete Fourier transform through the fast Fourier transform algorithm with respect to DPC++ interface.

https://www.intel.com/content/www/us/en/develop/documentation/oneapi-mkl-dpcpp-developer-reference/top/fourier-transform-functions.html

 

The DPC++ configuration parameters have type mkl::dft::config_param. Each parameter mkl::dft::config_param::<PARAMETER_NAME> has the same meaning and possible values as the C interface’s DFTI_<PARAMETER_NAME>. Some options corresponding to C interface might not be available on DPC++ interface yet. For a full description of each parameter’s meaning and options, see Config Params.

 

Kindly let us know if you need any details further.

 

Best Regards,

Shanmukh.SS

 

0 Kudos
sudoLife
New Contributor I
792 Views

Some of the options available in the C interface are not fully supported in the DPC++ interface yet

Got it, thanks.

 

So it is, indeed, not ready. and mkl::dft::config_value really doesn't exist, I guess.

0 Kudos
ShanmukhS_Intel
Moderator
809 Views

Hi,

 

Reminder:

Has the information provided helped? Is your issue resolved?

 

Kindly let us know if we could close this case at our end.

 

Best regards,

Shanmukh.SS

 

0 Kudos
ShanmukhS_Intel
Moderator
781 Views

Hi,

 

Thanks for accepting our solution. If you need any additional information, please post a new question as this thread will no longer be monitored by Intel.

 

Kindly refer to example mentioned under link below for the input parameters, "param" which means configuration parameter and "value" which mean configuration value.

 

https://www.intel.com/content/www/us/en/develop/documentation/oneapi-mkl-dpcpp-developer-reference/top/fourier-transform-functions/descriptor-precision-domain-set_value.html

 

Best Regards,

Shanmukh.SS

 

0 Kudos
Reply