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.

LAPACKE_xerbla

Gert-Jan
Beginner
158 Views
The xerbla error handling function can be replaced by your own xerbla function as described here: https://software.intel.com/en-us/node/522122. However, the LAPACK_ functions, like LAPACKE_sgels(), use LAPACKE_xerbla() to print error messages instead of xerbla(). This is different than documented at https://software.intel.com/en-us/node/522122. Unfortunately, LAPACKE_xerbla() does not seem to be documented, as far as I can see. I would like to replace LAPACKE_xerbla() with my own function, but to do so I need the function declaration. I only found a function description at netlib.org: http://www.netlib.org/lapack/lapacke_utils.h, which declares LAPACKE_xerbla() as "void LAPACKE_xerbla( const char *name, lapack_int info );" Is this the correct definition for the LAPACKE_xerbla function in MKL? And is this documented somewhere?
0 Kudos
1 Reply
Konstantin_A_Intel
158 Views

Hi Gert-Jan,

Indeed, LAPACKE_xerbla is not documented in MKL. We'll try to address this ASAP.

Regarding the interface, it's identical in MKL and Netlib LAPACK. Moreover, the function bodies are equal as well.

Best regards,

Konstantin

Reply