The OPAL Discussion Forum

["Calvo Portela, Pedro" <Pedro.Calvo AT ciemat.es>]

Dear Sherlp

I think I caught the bug.

When the predefined string values were introduced in the FSTYPE attribute, the options for the AMR solver were not considered. This bug had gone unnoticed until now, so I will fix it as soon as possible both in version 2021.1 and in the master branch.

I am sorry for any inconvenience this may cause, but until this bug is fixed, the AMR solver will not work. Then, you will be able to run it without any problem as long as you compile OPAL from source. Anyway, a new version OPAL-2022.1 is planned to be released at the beginning of April, with a new binary executable, where this bug will be fixed.

Cheers

Pedro

---

De: opal-request AT lists.psi.ch <opal-request AT lists.psi.ch> en nombre de Calvo Portela, Pedro <Pedro.Calvo AT ciemat.es>
Enviado: martes, 22 de marzo de 2022 8:49:10
Para: 'sherlp'; opal AT lists.psi.ch; Snuverink Jochem (PSI)
Asunto: Re: [Opal] simulation with amr fieldsolver

 

Dear Sherlp

I think that the error that OPAL throws may be due to a mismatch in the names of some Fieldsolver attributes. The regression test AmrRingCyclotron is not updated with the latest version of OPAL, nor with the master version, and moreover, it is not currently run with the other tests, so there is no maintenance tracking. Therefore, you should check that the fieldsolver attributes are correctly assigned with the accepted names according to the Manual.

I have detected some errors in your FIELDSOLVER command:

ITSOLVER options: "CG", "BICGSTAB", "GMRES"

AMR_MG_NORM options: "L1_NORM", "L2_NORM", "LINF_NORM"

Since OPAL 2021.1, all string variables are predefined internally in the code, so only some values are properly accepted. You must verify within the Manual the suitable names of those attributes

I'll try to update the regression tests related to AMR solver in order to avoid future mistakes

Cheers,

Pedro

------------------------------------------------

Dr. Pedro Calvo Portela
Unidad de Aceleradores - CIEMAT
Avenida Complutense 40, Edificio 7
28040 Madrid
Tel: +34 – 91 496 2554

---

De: opal-request AT lists.psi.ch <opal-request AT lists.psi.ch> en nombre de Snuverink Jochem (PSI) <jochem.snuverink AT psi.ch>
Enviado: martes, 22 de marzo de 2022 8:32
Para: 'sherlp'; opal AT lists.psi.ch
Asunto: AW: [Opal] simulation with amr fieldsolver

 

Dear Sherlp

Did you compile with ENABLE_AMR and ENABLE_AMR_MG_SOLVER equals TRUE? (https://gitlab.psi.ch/OPAL/src/-/wikis/For-Developers/Compile-OPAL#user-content-configure-opal).
If that is not the case then the AMR example won't work, though I admit the error message is not very helpful. Did you already try another non-AMR example?

Best regards,
Jochem

-----Ursprüngliche Nachricht-----
Von: opal-request AT lists.psi.ch <opal-request AT lists.psi.ch> Im Auftrag von sherlp
Gesendet: Dienstag, 22. März 2022 03:01
An: opal AT lists.psi.ch
Betreff: [Opal] simulation with amr fieldsolver

Dear all,

I tried to run the example named AmrRingCyclotron in RegressionTests, but fails. It shows that

Error>
Error> *** Parse error detected by function "Expressions::parseStringValue()"
Error> *** in line 82 of file "AmrRingCyclotron.in" before token ",":
Error>
FS1:FIELDSOLVER,FSTYPE=AMR_MG,MX=32,MY=32,MT=32,PARFFTX=TRUE,PARFFTY=TRUE,PARFFTT=TRUE,BCFFTX=OPEN,BCFFTY=OPEN,BCFFTT=OPEN,BBOXINCR=20,AMR_MAXLEVEL=2,AMR_MAXGRIDX=32,AMR_MAXGRIDY=32,AMR_MAXGRIDZ=32,AMR_BFX=16,AMR_BFY=16,AMR_BFZ=16,AMR_REFX=2,AMR_REFY=2,AMR_REFZ=2,AMR_DOMAIN_RATIO={1,0.75,0.75},AMR_TAGGING=CHARGE_DENSITY,AMR_DENSITY=1e09,
AMR_MG_REBALANCE=TRUE,AMR_MG_NSWEEPS=12,ITSOLVER=SA,AMR_MG_NORM=LINF;
Error>
Error>     String value expected

I looked up the manual but didn't find anything wrong with this .

Best regards,

Sherlp.