nek5000/nekexamples

Examples for Nek5000

Nek5000

Short Tests	Examples

Nek5000 is a fast and scalable open source CFD solver.

Release Notes

Make sure to read the release notes before using the code.

Getting Started

See here.

Troubleshooting

If you run into problems compiling, installing, or running Nek5000, please send a message to the User's Group mailing list.

Reporting Bugs

Nek5000 is hosted on GitHub and all bugs are reported and tracked through the Issues feature on GitHub. However, GitHub Issues should not be used for common troubleshooting purposes. If you are having trouble installing the code or getting your model to run properly, you should first send a message to the User's Group mailing list. If it turns out your issue really is a bug in the code, an issue will then be created on GitHub. If you want to request that a feature be added to the code, you may create an Issue on GitHub.

Contributing

Our project is hosted on GitHub. If you are planning a large contribution, we encourage you to discuss the concept here on GitHub and interact with us frequently to ensure that your effort is well-directed.

nekexamples's Issues

Where is the Kovasznay example?

Add overview to README

We should add an overview of all examples to the README.

Example	Description	Used Features
lowMach
eddy
...

Fs2_St2, PnPn-2 failure

The Fs2_St2, PnPn-2 test fails with the following message:

1:     FAILURE: amp: Test value 0.6376301 exceeds target value 0.6376171 +/- 1e-06

Nek5000/Nek5000@f0266aa is the first bad commit

Add example for new RANS implementation

Eddy_PsiOmega

The Eddy_PsiOmega PnPn and PnPn-2 tests fail at runtime with an apparent segfault during or after userchk:

done :: set initial conditions
  
 call userchk

===================================================================================
=   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
=   PID 15410 RUNNING AT churn
=   EXIT CODE: 139
=   CLEANING UP REMAINING PROCESSES
=   YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
===================================================================================
YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Segmentation fault (signal 11)
This typically refers to a problem with your application.
Please see the FAQ page for debugging suggestions
Finished running nek5000!

FsHydro, PnPn failure

The FsHydro, PnPn example fails with the following error:

1:     FAILURE: gmres: Test value 520.0 exceeds target value 0.0 +/- 108.0
2:     FAILURE: AMP: Test value -75.894666 exceeds target value -6.4616452e-05 +/- 0.002

Note that these reference values for have not been established. See #22 for further discussion.

Various Jenkins tests fail

https://jenkins-ci.cels.anl.gov/job/Nek5000/

Trapping Fortran errors in subprocesses

In lib/nekBinRun.py, the functions run_meshgen`` andrun_nekdon't trap the Fortran runtime errors in the called subprocesses. This should be fixed; otherwise,NekTests.py``` will incorrectly report that subprocesses (esp. genmap and genbox) have succeeded.

lowMach_test.rea misnamed

The test case for lowMach_test is looking for an .rea file misnamed "lowmach_test.rea" (note capitalization). The test case fails when it can't find the misnamed .rea file.

Renaming TestCase subclasses

@oanaoana requested that we rename the TestCase subclasses to exactly match the example subdirectories. For example, we would rename the "FsHydro" subclass to "fs_hyrdo".

This change would disobey naming conventions (not syntax rules) for class names. However, this would make things easier for users who don't care about the internals.

For example, right now, you would run a test case like:

python -m 'unittest' NekTests.FsHydro 

and with the proposed change, you would run a test case like this:

python -m 'unittest' NekTests.fs_hydro

That is, you don't need to remember separate class names and directory names.

Describe workflow for manually running examples

The README exclusively describes the usage of NekTests.py. This is an artifact of the development process. There was never a proper README for the NekExamples repo, but there was a README for the deprecated NekTests repo. When I moved the relevant bits of NekTests to NekExamples, I just copied the NekTests' README, too.

As it is, users might mistakenly believe that NekTests.py is the intended way to run examples, rather than a tool for software testing. This was certainly the case in a recent post to the mailing list.

We should describe how to manually run the examples, and we should do so before we describe the usage of NekTests.py.

lowMach_test is out of sync with Nek5000 repo

This PR
Nek5000/Nek5000#101
introduced some necessary fixes in:
https://github.com/Nek5000/Nek5000/tree/develop/short_tests/lowMach_test

Now, the SIZE and .usr files in this repo are not up to date with the main repo.

@stgeke

Peris, PnPn

The Peris, PnPn test case fails with the following error:

1:     FAILURE: gmres: Test value 25.0 exceeds target value 0.0 +/- 18.0

new SIZE for mhd tests Mhd_GpfB and Mhd_GpfM

mhd tests Mhd_GpfB and Mhd_GpfM use fast diagonalization method that require lfdm to be set to 1. Right now it is set to 0 by default in SIZE.inc, which be default should not be changed. We should move lfdm to SIZE for this tests

New tests for incl_def example

NekTests.py needs to be updated to run the incl_def/ example, which was recently merged in PR #25

cc: @nicooff

expected failures

A number of tests are "expected" to fail, in that they have failed in Nek5000 for some time now (dating back to the SVN repo).

Until we fix these tests, I'd like to mark them as "expected failures" in NekTests.py using one of these decorators:
https://docs.python.org/2/library/unittest.html#skipping-tests-and-expected-failures

Fs2_St1, PnPn-2 failure

The Fs2_St1, PnPn-2 test fails with the following message:

1:     FAILURE: amp: Test value 0.6382535 exceeds target value 0.6382414 +/- 1e-06

Nek5000/Nek5000@f0266aa is the first bad commit

Show how to use history points

Add short-test to examples

In case they are useful.

Don't run tests in serial on Jenkins

I think we can safely drop all serial runs. The functionality to run Nek5000 w/o MPI is tested with our short_tests already.

Rayleigh_Ray1, PnPn-2

The Rayleigh_Ray1, PnPn-2 test case fails with the following error:

1:     FAILURE: umax: Test value 0.004814308 exceeds target value 0.004831113 +/- 1e-05

Nek5000/Nek5000@ab95320 is the first bad commit

Rayleigh_Ray2, PnPn failure

The Rayleigh_Ray2, PnPn test fails with the following error:

1:     FAILURE: umax: Test value 0.005519951 exceeds target value 0.004549071 +/- 1e-05

Nek5000/Nek5000@ab95320 is the first bad commit

Allow user to set MPI ranks in test suite

Currently, the parallel test cases in NekTests.py are always run with 4 MPI ranks. We should allow the user to set the number of MPI ranks.

Fs2_StdWv, PnPn-2

The Fs2_StdWv, PnPn test case fails with the following error:

1:     FAILURE: amp: Test value 0.1403051 exceeds target value 0.1403287 +/- 1e-06

Nek5000/Nek5000@f0266aa is the first bad commit

Demonstrate how to use exo2nek

Show to run NekNek in MPMD mode

• demonstrate session specific settings in usr file
• setup different case file structure to run in MPMD mode

Hemi, PnPn-2

The Hemi, PnPn-2 test case fails at link time with the following error:

mpif77 -o nek5000 obj/hemi.o obj/drive.o obj/drive1.o obj/drive2.o obj/plan4.o obj/bdry.o obj/coef.o obj/conduct.o obj/connect1.o obj/connect2.o obj/dssum.o obj/edgec.o obj/eigsolv.o obj/gauss.o obj/genxyz.o obj/navier1.o obj/makeq.o obj/navier0.o obj/navier2.o obj/navier3.o obj/navier4.o obj/prepost.o obj/speclib.o obj/map2.o obj/turb.o obj/mvmesh.o obj/ic.o obj/gfldr.o obj/ssolv.o obj/planx.o obj/math.o obj/mxm_wrapper.o obj/hmholtz.o obj/gfdm_par.o obj/gfdm_op.o obj/gfdm_solve.o obj/subs1.o obj/subs2.o obj/genbox.o obj/gmres.o obj/hsmg.o obj/convect.o obj/induct.o obj/perturb.o obj/navier5.o obj/navier6.o obj/navier7.o obj/navier8.o obj/fast3d.o obj/fasts.o obj/calcz.o obj/byte.o obj/chelpers.o obj/byte_mpi.o obj/postpro.o obj/intp_usr.o obj/cvode_driver.o obj/nek_comm.o obj/vprops.o obj/makeq_aux.o obj/papi.o obj/nek_in_situ.o obj/readat_new.o obj/reader_re2.o obj/finiparser.o obj/iniparser.o obj/dictionary.o obj/mxm_std.o obj/comm_mpi.o obj/singlmesh.o obj/blas.o obj/dsygv.o -L/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/3rd_party/gslib/src -lgs 
obj/hemi.o: In function `particle_out_':
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/NekExamples/hemi/hemi.f:469:(.text+0x14f): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/NekExamples/hemi/hemi.f:470:(.text+0x16d): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/NekExamples/hemi/hemi.f:475:(.text+0x227): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
obj/drive2.o: In function `dofcnt_':
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/drive2.f:1585:(.text+0x9314): relocation truncated to fit: R_X86_64_PC32 against symbol `scrns_' defined in COMMON section in obj/convect.o
obj/plan4.o: In function `crespsp_':
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/plan4.f:237:(.text+0xc65): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/plan4.f:252:(.text+0xd18): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/plan4.f:263:(.text+0xdab): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/plan4.f:246:(.text+0x10fb): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
obj/bdry.o: In function `trst3d_':
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/bdry.f:1662:(.text+0x387): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
obj/coef.o: In function `area3_':
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/coef.f:1221:(.text+0x3ce): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/coef.f:1221:(.text+0x3d5): additional relocation overflows omitted from the output
collect2: error: ld returned 1 exit status
make: *** [nek5000] Error 1

implementing NekTestCase.run_neknek()

In #48, I introduced nekBinRun.run_neknek(). We should also implement NekTestCase.run_neknek() as a wrapper.

cc: @kmittal2

Add remaining test cases

We still need to implement some test cases from the legacy test suite. These are trickier than the others, so I've been procrastinating. :)

• 2d_eigtest/eig1.rea
• benard/ray9.rea
• benard/ray_dn.rea
• benard/ray_nn.rea
• cone/cone.rea
• cone/cone016.rea
• cone/cone064.rea
• cone/cone256.rea
• eddy/amg_eddy.rea
• eddy/htps_ed.rea
• eddy_neknek/eddy_neknek.rea
• expansion/expansion.rea

Rayleigh_Ray2, PnPn-2 failure

The Rayleigh_Ray2, PnPn-2 test fails with the following error:

1:     FAILURE: umax: Test value 0.00662153 exceeds target value 0.006728787 +/- 1e-05

Nek5000/Nek5000@e646239 is the first bad commit

Fs2_StdWv, PnPn failure

The Fs2_StdWv, PnPn test case fails with the following error:

1:     FAILURE: gmres: Test value 520.0 exceeds target value 0.0 +/- 20.0
2:     FAILURE: amp: Test value 0.6533972 exceeds target value 0.1403287 +/- 1e-06

Note that these reference values for have not been established. See #22 for further discussion.

Reference solutions for PnPn test cases

In the legacy test suite, the following test cases were expected to ABORT.

• fs_2/st1.rea, PnPn
• fs_2/std_wv.rea, PnPn
• `fs_hydro/fs_hydro.rea``, PnPn
• peris/peris.rea, PnPn
• solid/solid.rea, PnPn
• var_vis/var_vis.rea, PnPn

The PnPn solver has since been updated, and those test cases are not expected to ABORT. Hence, we should establish reference solutions and add them to the NekTests.py

port all SIZE files

In Nek5000/short_tests/, we have already ported all the SIZE files over to the new format (which includes a longer SIZE.inc file). We should do the same with NekExamples/*

Os7000, PnPn failure

The Os7000, PnPn test fails with the following error:

1:     FAILURE: egn: Test value 8.53793291e-05 exceeds target value 4.74494769e-05 +/- 1e-06

Nek5000/Nek5000@37f8226 is the first bad commit

Fs2_St1, PnPn failure

The Fs2_St1, PnPn test fails with the following messages:

1:     FAILURE: gmres: Test value 520.0 exceeds target value 0.0 +/- 38.0
2:     FAILURE: amp: Test value 0.6977515 exceeds target value 0.6382414 +/- 1e-06

Note that these reference values for have not been established. See #22 for further discussion.

Pipe_Helix, PnPn failure

The Pipe_Helix, PnPn test case fails with the following error:

1:     FAILURE: err2: Test value 1.9077636 exceeds target value 1.9077617 +/- 1e-06

Nek5000/Nek5000@54f7f9d is the first bad commit

IFCHAR = T ??

I notice that several examples have ifchar=T.

Some don't even use a dt big enough to warrant such usage (e.g., cyl_restart case).

I think all examples (except for short tests designed to test Characteristics) should have IFCHAR F unless we really want to show off CFL >> 1.

Otherwise, jobs will run twice as slow. People often start with one of the examples and modify it to suit their needs without realizing that IFCHAR=T.

(This happened in one of the large KTH runs that George did for pipe flow. He was running with CFL=0.5 and IFCHAR T. Turning that to F sped the code up by 2X and would have saved a million core hours---it was my fault for not paying attention earlier that time.)

Can we / should we switch all the examples to have default IFCHAR F (since the decision
to use Characteristics is typically a more advanced-user decision) ?

ExtCyl, PnPn-2 failure

The ExtCyl, PnPn-2 test case fails with the following message:

1:     FAILURE: 1dragx: Test value 1.21392285787 exceeds target value 1.2138878 +/- 1e-05

The first broken commit is:
Nek5000/Nek5000@9bf13ae

Add example with recycling BC

Hemi, PnPn

The Hemi, PnPn test case fails at link time with the following error:

mpif77 -o nek5000 obj/hemi.o obj/drive.o obj/drive1.o obj/drive2.o obj/plan4.o obj/bdry.o obj/coef.o obj/conduct.o obj/connect1.o obj/connect2.o obj/dssum.o obj/edgec.o obj/eigsolv.o obj/gauss.o obj/genxyz.o obj/navier1.o obj/makeq.o obj/navier0.o obj/navier2.o obj/navier3.o obj/navier4.o obj/prepost.o obj/speclib.o obj/map2.o obj/turb.o obj/mvmesh.o obj/ic.o obj/gfldr.o obj/ssolv.o obj/planx.o obj/math.o obj/mxm_wrapper.o obj/hmholtz.o obj/gfdm_par.o obj/gfdm_op.o obj/gfdm_solve.o obj/subs1.o obj/subs2.o obj/genbox.o obj/gmres.o obj/hsmg.o obj/convect.o obj/induct.o obj/perturb.o obj/navier5.o obj/navier6.o obj/navier7.o obj/navier8.o obj/fast3d.o obj/fasts.o obj/calcz.o obj/byte.o obj/chelpers.o obj/byte_mpi.o obj/postpro.o obj/intp_usr.o obj/cvode_driver.o obj/nek_comm.o obj/vprops.o obj/makeq_aux.o obj/papi.o obj/nek_in_situ.o obj/readat_new.o obj/reader_re2.o obj/finiparser.o obj/iniparser.o obj/dictionary.o obj/mxm_std.o obj/comm_mpi.o obj/singlmesh.o obj/blas.o obj/dsygv.o -L/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/3rd_party/gslib/src -lgs 
obj/hemi.o: In function `particle_out_':
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/NekExamples/hemi/hemi.f:469:(.text+0x14f): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/NekExamples/hemi/hemi.f:470:(.text+0x16d): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/NekExamples/hemi/hemi.f:475:(.text+0x227): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
obj/drive2.o: In function `dofcnt_':
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/drive2.f:1585:(.text+0x9314): relocation truncated to fit: R_X86_64_PC32 against symbol `scrns_' defined in COMMON section in obj/convect.o
obj/plan4.o: In function `crespsp_':
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/plan4.f:237:(.text+0xc65): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/plan4.f:252:(.text+0xd18): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/plan4.f:263:(.text+0xdab): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/plan4.f:246:(.text+0x10fb): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
obj/plan4.o: In function `plan4_':
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/plan4.f:80:(.text+0x17f1): relocation truncated to fit: R_X86_64_32S against symbol `orthov_' defined in COMMON section in obj/induct.o
obj/bdry.o: In function `trst3d_':
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/bdry.f:1662:(.text+0x387): relocation truncated to fit: R_X86_64_32S against symbol `scrns_' defined in COMMON section in obj/convect.o
obj/coef.o: In function `area3_':
/sandbox/jenkins-ci/workspace/Nek5000/COMPILER/gnu/TEST_CASE/Hemi/Nek5000/core/coef.f:1221:(.text+0x3ce): additional relocation overflows omitted from the output
collect2: error: ld returned 1 exit status
make: *** [nek5000] Error 1

Hemi failure

The Hemi PnPn and PnPn-2 tests fail at link-time with these errors:

mpif77 -o nek5000 obj/hemi.o obj/drive.o obj/drive1.o obj/drive2.o obj/plan4.o obj/bdry.o obj/coef.o obj/conduct.o obj/connect1.o obj/connect2.o obj/dssum.o obj/edgec.o obj/eigsolv.o obj/gauss.o obj/genxyz.o obj/navier1.o obj/makeq.o obj/navier0.o obj/navier2.o obj/navier3.o obj/navier4.o obj/prepost.o obj/speclib.o obj/map2.o obj/turb.o obj/mvmesh.o obj/ic.o obj/gfldr.o obj/ssolv.o obj/planx.o obj/math.o obj/mxm_wrapper.o obj/hmholtz.o obj/gfdm_par.o obj/gfdm_op.o obj/gfdm_solve.o obj/subs1.o obj/subs2.o obj/genbox.o obj/gmres.o obj/hsmg.o obj/convect.o obj/induct.o obj/perturb.o obj/navier5.o obj/navier6.o obj/navier7.o obj/navier8.o obj/fast3d.o obj/fasts.o obj/calcz.o obj/byte.o obj/chelpers.o obj/byte_mpi.o obj/postpro.o obj/cvode_driver.o obj/nek_comm.o obj/vprops.o obj/makeq_aux.o obj/papi.o obj/nek_in_situ.o obj/readat_new.o obj/reader_re2.o obj/finiparser.o obj/iniparser.o obj/dictionary.o obj/mxm_std.o obj/comm_mpi.o obj/singlmesh.o obj/blas.o obj/dsygv.o obj/ssygv.o obj/jl_gs.o obj/jl_sort.o obj/jl_sarray_transfer.o obj/jl_sarray_sort.o obj/jl_gs_local.o obj/jl_crystal.o obj/jl_comm.o obj/jl_tensor.o obj/jl_fail.o obj/jl_fcrystal.o obj/jl_findpts.o obj/jl_findpts_local.o obj/jl_obbox.o obj/jl_poly.o obj/jl_lob_bnd.o obj/jl_findpts_el_3.o obj/jl_findpts_el_2.o obj/jl_sparse_cholesky.o obj/jl_xxt.o obj/jl_fcrs.o 
obj/hemi.o: In function `interp_v_':
/sandbox/jenkins-ci/workspace/Nek5000-PGI/NekExamples/hemi/hemi.f:356: undefined reference to `intpts_setup_'
/sandbox/jenkins-ci/workspace/Nek5000-PGI/NekExamples/hemi/hemi.f:359: undefined reference to `intpts_'
make: *** [nek5000] Error 2

not defined intptd in Hemi

Routines intpts and intpts_setup removed form nek core

problems tweaking mhd/SIZE

In lib/nekFileConfig.py, we tweak lx2, ly2, and lz2 in the SIZE file by doing regex substitutions:

 re.sub(r'(^ {6}parameter *\( *lx2 *= *)\S+?( *\))', r'\g<1>{0}\g<2>'.format(lx2), l)

However, in mhd/SIZE, the declarations of lx2 and ly2 are capitalized differently from normal:

      PARAMETER (lx2=lx1-2) 
      PARAMETER (ly2=ly1-2)

So these values don't get tweaked correctly.

I think we can change the regex substitution to a case-insensitive one. (Might want to do this for all substitutions in the NekTests library?)

Renaming test case methods

@oanaoana requested that we rename the test case methods. The capitalization scheme is admittedly a little confusing. I named the methods:

• test_PnPn_Serial
• test_PnPn_Parallel
• test_PnPn2_Serial
• test_PnPn2_Parallel
  And I propose changing them to
• test_pnpn_serial
• test_pnpn_parallel
• test_pnpn2_serial
• test_pnpn2_parallel

Compiling prenek fails

Hi all,

I'm not sure if NekTools has any regression testing, but I tried cloning Nek5000 to find that prenek won't compile. I'm running on Ubuntu 16.10. Here is what the problem seems to be:

gfortran -o /home/gavin/code/Nek5000/bin/prex prenek.o curve.o edit.o build.o build1.o build2.o bound.o plot.o xinterface.o glomod.o legend.o vprops.o iolib.o subs.o zipper2.o postnek6.o screen.o revert.o crs.o mxm.o xdriver.o  -L/usr/lib/X11 -lX11 -lm
curve.o: In function `hex_transition_3d_e_':
curve.f:(.text+0x6f07): relocation truncated to fit: R_X86_64_PC32 against symbol `ctmp2_' defined in COMMON section in zipper2.o
curve.f:(.text+0x6f0e): relocation truncated to fit: R_X86_64_PC32 against symbol `ctmp2_' defined in COMMON section in zipper2.o
curve.f:(.text+0x6f15): relocation truncated to fit: R_X86_64_PC32 against symbol `ctmp2_' defined in COMMON section in zipper2.o
curve.f:(.text+0x6f2f): relocation truncated to fit: R_X86_64_PC32 against symbol `ctmp2_' defined in COMMON section in zipper2.o
curve.f:(.text+0x6fb0): relocation truncated to fit: R_X86_64_PC32 against symbol `ctmp2_' defined in COMMON section in zipper2.o
curve.f:(.text+0x6fef): relocation truncated to fit: R_X86_64_PC32 against symbol `ctmp2_' defined in COMMON section in zipper2.o
curve.f:(.text+0x702e): relocation truncated to fit: R_X86_64_PC32 against symbol `ctmp2_' defined in COMMON section in zipper2.o
curve.f:(.text+0x706d): relocation truncated to fit: R_X86_64_PC32 against symbol `ctmp2_' defined in COMMON section in zipper2.o
curve.f:(.text+0x7074): relocation truncated to fit: R_X86_64_PC32 against symbol `ctmp2_' defined in COMMON section in zipper2.o
curve.f:(.text+0x70c0): relocation truncated to fit: R_X86_64_PC32 against symbol `ctmp2_' defined in COMMON section in zipper2.o
curve.f:(.text+0x70c7): additional relocation overflows omitted from the output
collect2: error: ld returned 1 exit status
makefile:34: recipe for target 'prex' failed

Can anyone provide some guidance on this? It would be much appreciated.

Have a nice day,

Gavin Ridley

online documentation - small correction

In the instructions -- it says type "ethier" ... but I guess it should be type "turbChannel" ?

cd ~
tar -xvzf Nek5000_17.0.0.tar.gz
export PATH=$HOME/Nek5000/bin:$PATH
cd ~/Nek5000/tools; ./maketools genmap
cd ~/Nek5000/run
cp -r ../examples/turbChannel .
cd turbChannel
genmap # run partioner, on input type ethier
makenek turbChannel # build case, edit script to change settings
nekbmpi turbChannel 2 # run Nek5000 on 2 ranks in the background
echo -2 >ioinfo # stop Nek5000 run and dump checkpoint
visnek turbChannel; visit -o turbChannel.nek5000 # requires a VisIt installation

CylRestart_P

There is a problem with projection in this case. I cannot reproduce it on my machine, but I've got other case that bows up with projection turned on

modification of CylRestart tests

I would like to reorganize CylRestart tests. Right now we have 4 test ca, cb, pa, pb running independently pn-pn and pn-pn-2. It is not what one could expect, as ca should prepare restart files for cb, and pn-pn restart files differ with pn-pn-2 ones. So I would like to merge ca with cb and pa with pb. Is it ok?

Fs2_St2, PnPn failure

The Fs2_St2, PnPn test fails with this error:

1:     FAILURE: gmres: Test value 520.0 exceeds target value 0.0 +/- 38.0
2:     FAILURE: amp: Test value 0.6981582 exceeds target value 0.6376171 +/- 1e-06

Nek5000/Nek5000@843a5f8 is the first bad commit

ExtCyl: Could not find label "gmres: "

I tried to run the automatic test for the ext_cyl case with the 'release/v16.0.0' branch. While it seems that the runs work fine, the tests fail and I get the following error message:

ValueError: Could not find label "gmres: " in logfile ".../NekExamples/ext_cyl/ext_cyl.log.1.pn_pn.parallel". The run may have failed.

After a short investigation, it turns out that the label "gmres: " has been replaced by "gmres " (without the colon) in the logfile produced with the branch 'release/v16.0.0'.
This is not a big issue but if I fix the problem for the branch 'release/v16.0.0', the test might not work any more with the branches 'master' or 'develop'. How would you recommend to solve this issue?