Project

General

Profile

Bug #2876

2019.1 make check fails on AVX and AVX2 (Intel 2018u3)

Added by Stefan Becuwe about 1 year ago. Updated about 1 year ago.

Status:
New
Priority:
Normal
Assignee:
Erik Lindahl
Category:
-
Target version:
-
Affected version - extra info:
Affected version:
2019.1
Difficulty:
uncategorized
Close

Description

Hello

make check fails in 2019.1 on two different platforms. In both cases I used Intel 2018 update 3 compilers.

- on AVX, 3 SIMD tests fail, but only in single precision
- on AVX2, 1 MdrunNonIntegratorTests test (and consequently MdrunMpiTests) fails, but only in double precision

I attach for both cases the log file from the error, as well as the output from gmx* -version

FYI: I used the same Intel compiler with GROMACS 2018.3 without problem. On AVX, I had to add "-no-ftz" as a compiler flag to pass the tests.

Kind regards
Stefan

gmx-2019-1-version-variant-1-avx-info.txt (2.7 KB) gmx-2019-1-version-variant-1-avx-info.txt gmx -version Stefan Becuwe, 03/05/2019 02:37 PM
gmx-2019-1-version-variant-1-avx-log.txt (28.8 KB) gmx-2019-1-version-variant-1-avx-log.txt SIMD Stefan Becuwe, 03/05/2019 02:37 PM
gmx-2019-1-version-variant-2-avx2-info.txt (2.78 KB) gmx-2019-1-version-variant-2-avx2-info.txt gmx_mpi_d -version Stefan Becuwe, 03/05/2019 02:37 PM
gmx-2019-1-version-variant-2-avx2-log.txt (264 KB) gmx-2019-1-version-variant-2-avx2-log.txt MdrunNonIntegratorTests Stefan Becuwe, 03/05/2019 02:37 PM
regression-avx-2019.1-intel-2019u2.txt (1.75 KB) regression-avx-2019.1-intel-2019u2.txt regression AVX Intel 2019u2 Stefan Becuwe, 03/07/2019 01:34 PM
regression-avx2-2019.1-intel-2019u2.txt (2.45 KB) regression-avx2-2019.1-intel-2019u2.txt regression AVX2 Intel 2019u2 Stefan Becuwe, 03/07/2019 01:34 PM
freeenergy.tgz (6.19 KB) freeenergy.tgz md* files from freeenergy/expanded en freeenergy/simtemp Stefan Becuwe, 03/08/2019 10:27 AM
gmx-avx2-interactive-logfile (4.21 KB) gmx-avx2-interactive-logfile relevant part from regression report Stefan Becuwe, 03/08/2019 10:27 AM

Related issues

Related to GROMACS - Bug #2883: essentialdynamics fail with Intel MPINew

History

#1 Updated by Erik Lindahl about 1 year ago

These look very benign, so it's likely just a matter of allowing a bit more freedom in the tests for the Intel compiler.

#2 Updated by Erik Lindahl about 1 year ago

  • Assignee set to Erik Lindahl

#3 Updated by Stefan Becuwe about 1 year ago

Hello Erik

Update: I also tried the Intel 2019 update 2 compilers. Adding the '-no-ftz' flag on the AVX machine, all tests pass on both platforms!

I also tried the regression tests with Intel 2019 update 2. I found one case that goes wrong on both AVX and AVX2 in both single and double precision, and one extra case that fails on AVX2. See attachment for used flags and errors.

Kind regards
Stefan

#4 Updated by Roland Schulz about 1 year ago

The freeenergy error with ICC19u2 with AVX2 I can't reproduce. Could you please attach the mdrun.out, md.log files for those failing freeenergy tests?
The essentialdynamics error I can reproduce. Thanks for the report. The error is in the testing harness not in GROMACS. So you don't have to worry about the error. I crated #2883 for this bug.

#5 Updated by Roland Schulz about 1 year ago

  • Related to Bug #2883: essentialdynamics fail with Intel MPI added

#6 Updated by Stefan Becuwe about 1 year ago

Hello Roland

It seems those two specific tests (expanded and simtemp) from freeenergy are influenced by the environment in one way or another.
If I login directly to a compute node or login node, all regression tests pass.
If I start an interactive job (qsub -I -lnodes=1:ppn=28 (all cores)) and run the regression tests, those two tests fail. I reproduced this situation twice.
Let's hope this is useful information.

Kind regards
Stefan

Also available in: Atom PDF