topic Check the discussions on in Intel® oneAPI Math Kernel Library

MKL giving different results for different argument alignments

Tom_C_2 — Wed, 16 Dec 2015 09:34:33 GMT

We're seeing DSYEVX in MKL give different results when the A matrix is aligned on a 16-byte boundary or not. A very simple test case is available at https://github.com/tomkcook/dsyevx_bug.

We're using the following product versions:

Microsoft Visual Studio 2013 Version 12.0.40629.00 Update 5
Intel(R) Visual Fortran Composer XE 2013 SP1 Update 2 Integration for Microsoft Visual Studio* 2013, 14.0.0086.12
MKL 11.1 Update 2

We've got some more info on this, but before I waste anyone's time with that, I'd appreciate if someone could confirm this is actually a bug an not just me doing something daft!

The source file at Github has

mecej4 — Wed, 16 Dec 2015 11:44:45 GMT

The source file at Github has several lines with over 300 characters. Many compilers cannot handle more than 132 characters per line.

Thanks for the comment, but I

Tom_C_2 — Wed, 16 Dec 2015 11:50:25 GMT

Thanks for the comment, but I'm fairly confident that's not what's happening here. Even if the lines were being silently truncated, that should cause a compile error as the line-continuation ampersands are at the ends of lines.

Check the discussions on

TimP — Wed, 16 Dec 2015 13:08:00 GMT

Check the discussions on this subject on the MKL forum, and consider whether you want to try the conditional numerical reproducibility options.

If you wish to minimize alignment dependencies in your own compiled code, set -fp:source or equivalent VS property.

Aligned results may be slightly more accurate, with differences indicative of the level of accuracy.

Tom, Could you check the

Gennady_F_Intel — Wed, 16 Dec 2015 18:48:19 GMT

Tom, Could you check the problem with the latest version of MKL 11.3? This version fully supports unaligned input. But, please take into account - the #of threads should be the same from run to run.