{chem}[foss/2021a] CP2K v9.1

[branfosj]

(created using eb --new-pr)

Draft while we check the tests and dependencies.

[branfosj]

Test report by @branfosj
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
bear-pg0211u03a.bear.cluster - Linux RHEL 8.5, x86_64, Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz (cascadelake), Python 3.6.8
See https://gist.github.com/0c28f1d8afcfeb13e83afd7a9e56e597 for a full test report.

[branfosj]

Test report by @branfosj
SUCCESS
Build succeeded for 6 out of 6 (1 easyconfigs in total)
bear-pg0105u36b.bear.cluster - Linux RHEL 8.5, x86_64, Intel(R) Xeon(R) Platinum 8360Y CPU @ 2.40GHz (icelake), Python 3.6.8
See https://gist.github.com/34cd135db70f5e2a8db99f74bed7d763 for a full test report.

[boegel]

@branfosj Any updates on this?

[branfosj]

--------------------------------- Summary --------------------------------
Number of FAILED  tests 18
Number of WRONG   tests 6
Number of CORRECT tests 3604
Total number of   tests 3628
--------------------------------------------------------------------------
Number of LEAKING tests 0
Number of memory  leaks 0
--------------------------------------------------------------------------
GREPME 18 6 3604 0 3628 0

Summary: correct: 3604 / 3628; wrong: 6; failed: 18; 128min
Status: FAILED

The 18 failed look to be SIGSEGV's, with backtraces similar to:

[bear-pg0211u03a:2616826:0:2616850] Caught signal 11 (Segmentation fault: address not mapped to object at address 0xfffffffc0e325670)
==== backtrace (tid:2616850) ====
 0 0x00000000000211fe ucs_debug_print_backtrace()  /dev/shm/build-branfosj-admin/branfosj-admin-up/UCX/1.10.0/GCCcore-10.3.0/ucx-1.10.0/src/ucs/debug/debug.c:656
 1 0x0000000000012c20 .annobin_sigaction.c()  sigaction.c:0
 2 0x000000000118bb6f __hfx_load_balance_methods_MOD_estimate_block_cost.constprop.0()  hfx_load_balance_methods.F90:0
 3 0x000000000119441d __hfx_load_balance_methods_MOD_hfx_load_balance()  ???:0
 4 0x0000000000b716ea __hfx_energy_potential_MOD_integrate_four_center._omp_fn.0()  hfx_energy_potential.F90:0
 5 0x000000000001a046 gomp_thread_start()  /dev/shm/build-branfosj-admin/branfosj-admin-up/GCCcore/10.3.0/system-system/gcc-10.3.0/stage3_obj/x86_64-pc-linux-gnu/libgomp/../../../libgomp/team.c:123
 6 0x000000000000817a start_thread()  pthread_create.c:0
 7 0x00000000000fcdc3 __GI___clone()  :0

6 wrong - 4 are just outside tolerance and 2 are quite a bit out:

• QS/regtest-gpw-4/H2O-debug-5.inp.out: relative error : 1.52789104e-07 > numerical tolerance = 4e-10
• QS/regtest-gpw-4/H2O-debug-6.inp.out: relative error : 1.52790209e-07 > numerical tolerance = 4e-10
• QS/regtest-mp2-grad/H2O_grad_mme.inp.out: relative error : 6.67224302e-09 > numerical tolerance = 6e-09
• QS/regtest-mp2-grad/H2O_grad_gpw.inp.out: relative error : 7.32319726e-08 > numerical tolerance = 7e-08
• QS/regtest-mp2-grad/H2O_grad_ri-hfx.inp.out: relative error : 6.40217302e-09 > numerical tolerance = 6e-09
• Fist/regtest-1-4/multipole_dipole.dbg_f_real.inp.out: relative error : 1.03449783e-14 > numerical tolerance = 1.0E-14

[boegel]

Test report by @boegel
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
node3519.doduo.os - Linux RHEL 8.4, x86_64, AMD EPYC 7552 48-Core Processor (zen2), Python 3.6.8
See https://gist.github.com/bb6101b81f21d3c7c11c7e22bb7a6c9b for a full test report.

[alinelena]

ok I am doing 9.1 and 2022.1 for 2022a I see similar issues with tolerance misses...
the segfaults run fine if I run it by hand so I suspect is just something wrong the way in which tests are run.

[alinelena]

see #16007

[branfosj]

There are PRs for more recent versions, so closing this one