Hi all,
Regarding the failed Jenkins test related to IllinoisGRMHD, it appears to be due to a recent breakage in ML_BSSN (likely related to trac #1932):
ERROR from host 3486efd77091 process 0 while executing schedule bin (none), routine (no thorn)::(no routine) in thorn CarpetLib, file /home/jenkins/workspace/EinsteinToolkit/arrangements/Carpet/CarpetLib/src/ggf.cc:334: -> The variable "ML_BSSN::cS" has only 1 active time levels, which is not enough for boundary prolongation of order 2 cactus_sim: /home/jenkins/workspace/EinsteinToolkit/arrangements/Carpet/Carpet/src/helpers.cc:275: int Carpet::Abort(const cGH*, int): Assertion `0' failed.
Please let me know if there's something that needs to be fixed in the IllinoisGRMHD parfile.
-Zach
* * * Zachariah Etienne Assistant Professor of Mathematics West Virginia University
On Mon, Oct 3, 2016 at 11:07 AM, rhaas@illinois.edu wrote:
Present: Roland, Frank, Steve, Ian, Eloisa, Erik, Roberto, Vassili, Zach
Failing Jenkins tests:
- currently failing
https://build-test.barrywardell.net/job/EinsteinToolkit/ lastCompletedBuild/testReport/
- WENO test -> Roland will take care of it, NaNChecker -> Roland, Frank: will write C code to produce NaNs
- ML_BSSN constraint failures: push change to ML_BSSN_Helper -> push fix to ML_BSSN_Helper. Roland gets SEGFAULTs when applying this patch in ml-gw1d-small tests (in CarpetLib's copy_from routine CarpetLib/gdata.cc:188)
- perl @INC will drop "." in 5.26. Which will break compiles. Steve points out that one can adjust PERL5PATH env variable to fix this without a code change. A code change will still be applied for future versions.
ET website:
- gallery could use some language polish
- try and make gallery examples use only ET thornlists
- aim for this in the next release
Include Llama in ET:
- Erik tasked with starting the process of investigating whether it can be included
- needs to meet the guidelines for the ET
- needs documentation and basic examples, Vassili will talk with Erik about this at MW24
Update LORENE:
- have second thorn LORENE2 for new version
- try and make choice between the two of them a run-time choice
- Roland suggests to patch LORENE1 in ET to fix handling of piecewise polytropes since the fix was included in LORENE after the file format update. Roland will provide the patch (from the original authors).
- want to keep backwards compatible, list both in ThornList, eventually give preference to LORENE2 if only one can be compiled at the same time
- Roberto mentions that the code to regenerate the BNS data files is included in LORENE and can be used to regenerate the public data files in the new format
Yours, Roland -- My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from http://keys.gnupg.net.
Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users