Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
Announcements
The Intel sign-in experience has changed to support enhanced security controls. If you sign in, click here for more information.

Issues with MKL PARDISO

SergP
Beginner
776 Views

Hi,

 

We have found several issues with MKL PARDISO for improved two-level factorization algorithm (iparm[23]=10), namely:

1. CNR functionality works incorrectly, solutions could be incorrect;

2. pardiso_getdiag - segmentation faulted in some cases;

3. user permutation vector (iparm[4]) works incorrectly in multi-threading mode.

MKL information:

Linux

Major version:     2021

Minor version:     0

Update version:    1

Build:             20201104

We have prepared reproducers with specific matrices, how can we share them for confirmation of the issues?

 

Regards,

Sergey

0 Kudos
6 Replies
Gennady_F_Intel
Moderator
768 Views

Sergey,

You could attach this case to this thread or you may send us this reproducer via the private communication.

SergP
Beginner
729 Views

Gennady,

 

I'd prefer to use private communications. Could you please send me instructions on how to do it?

 

Sergey

Gennady_F_Intel
Moderator
711 Views

Sergey,

All of the issues you reported were reproduced with MKL 2021.3 and will be escalated against the Sparse Solver Team. We will keep this thread informed of the status of these issues.

Thanks,

Gennady


SergP
Beginner
557 Views

Hi Gennady,

 

Am I right that 

MKLD-11665 Fixed the incorrect result produced by MKL PARDISO when a permutation vectore is used

is not related to issues, reported by me? 

Thanks,

Sergey

Gennady_F_Intel
Moderator
526 Views

Hello Sergey,

it is very similar but not exactly. You might check if this update somehow fixes this particular problem.

-Gennady


SergP
Beginner
372 Views

Hi Gennady,

 

We have checked this update but our issues in it are still reproduced. 

Do you have any estimation of the time when they will be fixed?

 

Regards,

Sergey

Reply