MKL Feast (inner memory problem: info=-2)

[Di_Luca__Daniele]

Hi,

i'm trying to solve (sparse) symmetric generalized eigenvalue problems using dfeast_scrgv().

It always worked fine with relatively small problems (up to 2000*2000 sparse matrix), but it turned out i can't solve the bigger ones (about 60000*60000 sparse matrix) as it always returns 'info = -2", that I know it refers to inner memory problems, but I don't know how to fix it.

The problem is solvable using other algorithms (not  included in MKL).

I'm using the latest Intel MKL 2019 library and Intel C++ Compiler 19.0 on Visual Studio 2013 Platform Toolset.

May I ask you for some help?

Thanks in advance,

Daniele

[Gennady_F_Intel]

please try to use ILP64 API. 

[Di_Luca__Daniele]

Thank you Gennady for the support. I'm sorry I forgot to write it before, but I need to compile a 32-bit project using that Feast routine, is there any other way to fix it besides running it in 64-bit, or is it a known issue? I can share the data I used if needed.

Thanks,

Daniele

Gennady F. (Intel) wrote:

please try to use ILP64 API.

[Gennady_F_Intel]

as i guess that's not possible because 32 bit version of library supports 32bit integer only and for cases exceed 2^31 there is only one chose - moving to the 64 bit version of mkl.

[Di_Luca__Daniele]

Thank you Gennady,

as you said, using the ILP64 API it works fine, though I had some problem setting the compiler (Intel C++ 19.0 on VS2017) to use the MKL 64bit libraries, and that’s why it needed so much to reply to you. Actually, I even opened a new topic (https://software.intel.com/en-us/forums/intel-math-kernel-library/topic/805556) asking for help, as I went through some memory access violation (which I still didn’t solve at all).

Thanks again,

Daniele