Gromacs 2025.3 installation, make check regression tests /complex failed

User discussions
1

GROMACS version: 2025.3
GROMACS modification: No

g ++ 13.30, Mpirun 4.1.6, cmake 3.28.3, Cuda 13.0.48, Python 3.12.3

My flags are

cmake .. -DGMX_GPU=CUDA DGMX_BUILD_OWN_FFTW=ON -DREGRESSIONTEST_DOWNLOAD=ON

when I run

make check

I get that regressiontest/complex failed (as shown below). All other test are passed. I would really like some advise on how to proceed for finding out, what I did wrong. Thank you! /Marie

93/94 Test #93: regressiontests/complex …***Failed 375.93 sec
Re-running aminoacids using only CPU-based non-bonded kernels
Re-running aminoacids using CPU-based update
Re-running awh_multidim using only CPU-based non-bonded kernels
Re-running awh_multidim using CPU-based update
Re-running butane using only CPU-based non-bonded kernels
Re-running butane using CPU-based update
Re-running dd121 using only CPU-based non-bonded kernels
Mdrun cannot use the requested (or automatic) number of ranks, retrying with 8.

Abnormal return value for ’ gmx mdrun -nb cpu -notunepme >mdrun.out 2>&1’ was 1
Retrying mdrun with better settings…
Re-running distance_restraints using only CPU-based non-bonded kernels
Re-running distance_restraints using CPU-based update
Re-running ethyleenglycol using only CPU-based non-bonded kernels
Mdrun cannot use the requested (or automatic) number of ranks, retrying with 8.

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was 1
Retrying mdrun with better settings…
Re-running ethyleenglycol using CPU-based update
Re-running field using only CPU-based non-bonded kernels
Re-running field using CPU-based update
Re-running nacl using only CPU-based non-bonded kernels
Re-running nacl using CPU-based update

Abnormal return value for ’ gmx mdrun -nb cpu -notunepme >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in nbnxn-energy-groups for nbnxn-energy-groups
Re-running nbnxn-free-energy using only CPU-based non-bonded kernels
Mdrun cannot use the requested (or automatic) number of ranks, retrying with 8.

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was 1
Retrying mdrun with better settings…
Re-running nbnxn-free-energy using CPU-based update
Re-running nbnxn-free-energy-vv using only CPU-based non-bonded kernels
Mdrun cannot use the requested (or automatic) number of ranks, retrying with 8.

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was 1
Retrying mdrun with better settings…
Re-running nbnxn-ljpme-geometric using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in nbnxn-ljpme-geometric/nb-cpu for nbnxn-ljpme-geometric-nb-cpu
Re-running nbnxn-ljpme-geometric using CPU-based update
Re-running nbnxn-ljpme-LB using only CPU-based non-bonded kernels
Re-running nbnxn-ljpme-LB-geometric using only CPU-based non-bonded kernels
Re-running nbnxn-vdw-force-switch using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in nbnxn-vdw-force-switch/nb-cpu for nbnxn-vdw-force-switch-nb-cpu
Re-running nbnxn-vdw-force-switch using CPU-based update
Re-running nbnxn-vdw-potential-switch using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in nbnxn-vdw-potential-switch/nb-cpu for nbnxn-vdw-potential-switch-nb-cpu
Re-running nbnxn-vdw-potential-switch using CPU-based update
Re-running nbnxn-vdw-potential-switch-argon using only CPU-based non-bonded kernels
Re-running nbnxn-vdw-potential-switch-argon using CPU-based update
Re-running nbnxn_pme using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in nbnxn_pme/nb-cpu for nbnxn_pme-nb-cpu
Re-running nbnxn_pme using CPU-based update
Re-running nbnxn_pme_order5 using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in nbnxn_pme_order5/nb-cpu for nbnxn_pme_order5-nb-cpu
Re-running nbnxn_pme_order5 using CPU-based update
Re-running nbnxn_pme_order6 using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in nbnxn_pme_order6/nb-cpu for nbnxn_pme_order6-nb-cpu
Re-running nbnxn_pme_order6 using CPU-based update
Re-running nbnxn_rf using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in nbnxn_rf/nb-cpu for nbnxn_rf-nb-cpu
Re-running nbnxn_rf using CPU-based update
Re-running nbnxn_rzero using only CPU-based non-bonded kernels
Re-running nbnxn_rzero using CPU-based update
Re-running nbnxn_vsite using only CPU-based non-bonded kernels
Re-running nst_mismatch using only CPU-based non-bonded kernels
Re-running nst_mismatch using CPU-based update
Re-running octahedron using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in octahedron/nb-cpu for octahedron-nb-cpu
Re-running octahedron using CPU-based update
Re-running orientation-restraints using only CPU-based non-bonded kernels
Re-running position-restraints using only CPU-based non-bonded kernels
Re-running position-restraints using CPU-based update
Re-running pr-vrescale using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in pr-vrescale/nb-cpu for pr-vrescale-nb-cpu
Re-running pr-vrescale using CPU-based update
Re-running pull_constraint using only CPU-based non-bonded kernels
Re-running pull_cylinder using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in pull_cylinder/nb-cpu for pull_cylinder-nb-cpu
Re-running pull_cylinder using CPU-based update
Re-running pull_geometry_angle using only CPU-based non-bonded kernels
Re-running pull_geometry_angle using CPU-based update
Re-running pull_geometry_angle-axis using only CPU-based non-bonded kernels
Re-running pull_geometry_angle-axis using CPU-based update
Re-running pull_geometry_dihedral using only CPU-based non-bonded kernels
Re-running pull_geometry_dihedral using CPU-based update
Re-running sw using only CPU-based non-bonded kernels
Re-running swap_x using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in swap_x/nb-cpu for swap_x-nb-cpu
Re-running swap_y using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in swap_y/nb-cpu for swap_y-nb-cpu
Re-running swap_z using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in swap_z/nb-cpu for swap_z-nb-cpu
Re-running tip4p using only CPU-based non-bonded kernels
Re-running tip4pflex using only CPU-based non-bonded kernels

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was -1
FAILED. Check mdrun.out, md.log file(s) in tip4pflex/nb-cpu for tip4pflex-nb-cpu
Re-running urea using only CPU-based non-bonded kernels
Mdrun cannot use the requested (or automatic) number of ranks, retrying with 8.

Abnormal return value for ’ gmx mdrun -notunepme -nb cpu >mdrun.out 2>&1’ was 1
Retrying mdrun with better settings…
Re-running urea using CPU-based update
Re-running walls using only CPU-based non-bonded kernels
Re-running walls using CPU-based update
15 out of 115 complex tests FAILED

  Start 94: regressiontests/essentialdynamics

94/94 Test #94: regressiontests/essentialdynamics … Passed 25.71 sec

99% tests passed, 1 tests failed out of 94

Label Time Summary:
GTest = 662.01 secproc (90 tests)
IntegrationTest = 357.41 secproc (29 tests)
MpiTest = 452.11 secproc (21 tests)
QuickGpuTest = 142.36 secproc (23 tests)
SlowGpuTest = 902.36 secproc (16 tests)
SlowTest = 288.30 secproc (14 tests)
UnitTest = 16.29 sec*proc (47 tests)

Total Test time (real) = 760.76 sec

The following tests FAILED:
93 - regressiontests/complex (Failed)
Errors while running CTest
make[3]: *** [CMakeFiles/run-ctest-nophys.dir/build.make:71: CMakeFiles/run-ctest-nophys] Error 8
make[2]: *** [CMakeFiles/Makefile2:3383: CMakeFiles/run-ctest-nophys.dir/all] Error 2
make[1]: *** [CMakeFiles/Makefile2:3416: CMakeFiles/check.dir/rule] Error 2
make: *** [Makefile:628: check] Error 2

2

I don’t fully understand this output. Maybe the tests fail because you are asking for too many ranks (but then our tests should handle this better). @al42and do you have an idea what is going wrong here?

Anyhow, I think the issue is with the test setup, your GROMACS installation should be fine.

3

@Marie, can you attach the files mentioned in the error messages?

E.g., for

FAILED. Check mdrun.out, md.log file(s) in pull_cylinder/nb-cpu for pull_cylinder-nb-cpu,

that would be complex/pull_cylinder/nb-cpu/mdrun.out and complex/pull_cylinder/nb-cpu/md.log

4

md.log (19.6 KB) - pull cylinder

mdrun.txt (2.4 KB) - pull cylinder

These are the files for 1 of the error messages, I can attach more if needed.

Thank you for looking into this!

5

Thanks.

You can check if the errors in other files are any different, but what you attached confirms Berk’s suspicion:

Fatal error:
The number of ranks selected for particle-particle work (22) contains a large
prime factor 11. In most cases this will lead to bad performance. Choose a
number with smaller prime factors or set the decomposition (option -dd)
manually.

@Marie: this error should be safe to ignore. You can also try adding -DREGRESSIONTEST_EXTRA_ARGS='-nt;2' to your CMake command; then run make check again; it should now run with fixed configuration, avoiding this error.

@hess: CPUs with 22 threads are fun. We have 6 P-cores (with 2 threads each), 8 E-cores and 2 LPE-cores (with 1 thread each). Yet GROMACS incorrectly reports them as 11 cores with 2 threads each. Package 0: [ 0 1] [ 2 3] [ 4 5] [ 6 7] [ 8 9] [ 10 11] [ 12 13] [ 14 15] [ 16 17] [ 18 19] [ 20 21].

@Marie: To help us improve hardware detection for new Intel CPUs, could you run the following command in your GROMACS source root directory?

./src/gromacs/hardware/tests/capture-topology.sh CoreUltra9185H

This script will create a directory named CoreUltra9185H containing hardware topology information for your CPU. Please archive this directory (e.g., using tar -czvf CoreUltra9185H.tar.gz CoreUltra9185H) and share the resulting file with us.

6

I checked the other files and they all contain the same fatal error.

Good to know that it shouldn’t affect the GROMACS installation

Link for CoreUltra9185H.tar

Thank you for your help!

1 Like