#38 use of subCycle.H not functional in foam-extend 4.0

[marc belleau]

Hi

This bug occurs in fe40 but not in OF1806

The relevant part of my code is just two consecutive lines:

......

    subCycle<volScalarField> neSubCycle(ne, 3);
    subCycle<volScalarField> npSubCycle(np, 3);

.....

I execute using the debugger gdb with one breakpoint.

gdb $FOAM_USER_APPBIN/plasmacoupleddbnsTurbFoam

Breakpoint 1, main (argc=1, argv=0x7fffffff87a8) at plasmacoupleddbnsTurbFoam.C:326
326 subCycle<volscalarfield> neSubCycle(ne, 3);
(gdb) n
327 subCycle<volscalarfield> npSubCycle(np, 3);
(gdb) n</volscalarfield></volscalarfield>

--> FOAM FATAL ERROR:

object already allocated
From function void autoPtr<t>::set(T*)
in file lnInclude/autoPtrI.H at line 83.</t>

FOAM aborting

Program received signal SIGABRT, Aborted.

0x00002aaaafbc0214 in raise () from /cvmfs/soft.computecanada.ca/nix/store/63pk88rnmkzjblpxydvrmskkc8ci7cx6-glibc-2.24/lib/libc.so.6
(gdb)

Maybe I will be forced to forget about fe40 and instead use OF1806 compressible but non dbns solver

[Hrvoje Jasak]

Can you provide a bug fix?

[marc belleau]

No, but on second thought it is perhaps not healthy to call npSubCycle(np,3) before executing neSubCycle.endSubCycle() in either OF1806 or fe40. It seems that if you do, you wind up with runTime.DeltaT divided by 6 instead of 3 for the field np. I could be wrong as I did not write the OF code and I could not find a tutorial that explains how to use subCycle.H. Anyhow, I thought this SourceForge web page was about reporting OF bugs and not fixing them. Regards, Marc

---

From: Hrvoje Jasak hjasak@users.sourceforge.net
Sent: January 14, 2020 5:37 AM
To: [foam-extend:tickets] 38@tickets.foam-extend.p.re.sourceforge.net
Subject: [foam-extend:tickets] #38 use of subCycle.H not functional in foam-extend 4.0

Can you provide a bug fix?

---

[tickets:#38]https://sourceforge.net/p/foam-extend/tickets/38/ use of subCycle.H not functional in foam-extend 4.0

Status: open
Milestone: 1.0
Created: Mon Jan 13, 2020 07:48 PM UTC by marc belleau
Last Updated: Mon Jan 13, 2020 07:48 PM UTC
Owner: nobody

Hi

This bug occurs in fe40 but not in OF1806

The relevant part of my code is just two consecutive lines:

......

subCycle<volScalarField> neSubCycle(ne, 3);
subCycle<volScalarField> npSubCycle(np, 3);

.....

I execute using the debugger gdb with one breakpoint.

gdb $FOAM_USER_APPBIN/plasmacoupleddbnsTurbFoam

Breakpoint 1, main (argc=1, argv=0x7fffffff87a8) at plasmacoupleddbnsTurbFoam.C:326
326 subCycle<volscalarfield> neSubCycle(ne, 3);
(gdb) n
327 subCycle<volscalarfield> npSubCycle(np, 3);
(gdb) n</volscalarfield></volscalarfield>

--> FOAM FATAL ERROR:

object already allocated
From function void autoPtr<t>::set(T*)
in file lnInclude/autoPtrI.H at line 83.</t>

FOAM aborting

Program received signal SIGABRT, Aborted.

0x00002aaaafbc0214 in raise () from /cvmfs/soft.computecanada.ca/nix/store/63pk88rnmkzjblpxydvrmskkc8ci7cx6-glibc-2.24/lib/libc.so.6
(gdb)

Maybe I will be forced to forget about fe40 and instead use OF1806 compressible but non dbns solver

---

Sent from sourceforge.net because you indicated interest in https://sourceforge.net/p/foam-extend/tickets/38/

To unsubscribe from further messages, please visit https://sourceforge.net/auth/subscriptions/