topic Hello, please check if the in Intel® oneAPI Math Kernel Library

cluster_sparse_solver iparm(2)

asd__asdqwe — Mon, 08 Jan 2018 21:24:29 GMT

Hello,

How is the feature iparm(2) = 10 supposed to work with cluster_sparse_solver? If I adapt cl_solver_unsym_distr_c.c by changing the value of iparm[ 1] to 10, with 2 processes, it hangs on both Linux with IntelMPI and macOS with MPICH, and it segfaults with more than 2 processes.

Moreover, there is some unwanted output:

$ mpirun -np 2 ./a.out
RANK # 0 Total mem 0
RANK # 1 Total mem 0

$ mpirun -np 4 ./a.out

===================================================================================
= BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
= PID 48932 RUNNING AT
= EXIT CODE: 11
= CLEANING UP REMAINING PROCESSES
= YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
===================================================================================
YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Segmentation fault: 11 (signal 11)

Could this be fixed please?

Thanks.

Have you checked the problem

Gennady_F_Intel — Tue, 09 Jan 2018 04:10:56 GMT

Have you checked the problem with the latest MKl 2018 u1 ( released at Nov '17)?

I've checked this problem

asd__asdqwe — Tue, 09 Jan 2018 06:41:09 GMT

I've checked this problem with Intel(R) Math Kernel Library Version 2018.0.1 Product Build 20171007 for Intel(R) 64 architecture applications on all OSes.

With the Fortran example, and

asd__asdqwe — Tue, 09 Jan 2018 07:02:07 GMT

With the Fortran example, and iparm(2) = 10, it also segfaults with more than 2 processes, but it seems OK with 2 processes, though there is a lot of unwanted output.

$ mpirun -np 4 _results/intel_intelmpi_lp64_intel64_a/cl_solver_sym_distr_f.exe
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
cl_solver_sym_dis 000000000652FC3D Unknown Unknown Unknown
libpthread-2.26.s 00007FC1AB87F160 Unknown Unknown Unknown

$ mpirun -np 2 _results/intel_intelmpi_lp64_intel64_a/cl_solver_sym_distr_f.exe
RANK # 0 Total mem 0
RANK # 1 Total mem 0
Memory allocated on phase 11 on Rank # 0 0.0000 Gb
Memory allocated on phase 11 on Rank # 1 0.0000 Gb

Can you reproduce these

asd__asdqwe — Thu, 11 Jan 2018 06:45:19 GMT

Can you reproduce these errors on both Linux and macOS with both C and Fortran?

Thanks.

yes, we see the problem on

Gennady_F_Intel — Fri, 12 Jan 2018 03:51:20 GMT

yes, we see the problem on our side. We will record it and fix it in future release. the next time, If any new issue, please feel free to submit to Online Service Center

Hello,

Gennady_F_Intel — Fri, 12 Jan 2018 18:39:12 GMT

Hello,

we fixed the problem and if you have access to the Online Service Center, then you may submit ticket there and I will give you the engineering build to check if the problem exists on your side.

thanks,

Gennady

Hello, please check if the

Gennady_F_Intel — Mon, 07 May 2018 18:00:38 GMT

Hello, please check if the issue still exists into the latest MKL 2018 update 2 and let us know the result.