<div dir="auto">I’m using TSSSP steppers. </div><div dir="auto">I know the RHS is not zero because when I output the solution from the Fortran code in VTK format, it looks as expected and indeed moves forward with time. I won’t put the code as an attachement there is a little too much but the MR is up to date on the VTK output (at least). </div><div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">Le lun. 28 déc. 2020 à 18:42, Jed Brown <<a href="mailto:jed@jedbrown.org">jed@jedbrown.org</a>> a écrit :<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-style:solid;padding-left:1ex;border-left-color:rgb(204,204,204)">I think I'm not following something. The time stepper implementation (TSStep_XYZ) is where ts->vec_sol should be updated. The controller doesn't have anything to do with it. What TS are you using and how do you know your RHS is nonzero?<br>
<br>
Thibault Bridel-Bertomeu <<a href="mailto:thibault.bridelbertomeu@gmail.com" target="_blank">thibault.bridelbertomeu@gmail.com</a>> writes:<br>
<br>
> Hi Jed,<br>
><br>
> Thanks for your message.<br>
> I implemented everything in C as you suggested and it works fine except for<br>
> one thing : the ts->vec_sol does not seem to get updated when seen from the<br>
> C code (it is on the Fortran side though the solution is correct).<br>
> As a result, the time step (that uses among other things the max velocity<br>
> in the domain) is always at the value it gets from the initial solution.<br>
> Any idea why ts->vec_sol does not seem to be updated ? (I checked the<br>
> stepnum and time is updated though , when accessed with TSGetTime for<br>
> instance).<br>
><br>
> Cheers,<br>
> Thibault<br>
><br>
> Le lun. 28 déc. 2020 à 15:30, Jed Brown <<a href="mailto:jed@jedbrown.org" target="_blank">jed@jedbrown.org</a>> a écrit :<br>
><br>
>> Thibault Bridel-Bertomeu <<a href="mailto:thibault.bridelbertomeu@gmail.com" target="_blank">thibault.bridelbertomeu@gmail.com</a>> writes:<br>
>><br>
>> > Good morning everyone,<br>
>> ><br>
>> > Thank you Barry for the answer, it works now !<br>
>> ><br>
>> > I am facing (yet) another situation: the TSAdaptRegister function.<br>
>> > In the MR on gitlab, Jed mentioned that sometimes, when function pointers<br>
>> > are not stored in PETSc objects, one can use stack memory to pass that<br>
>> > pointer from fortran to C.<br>
>><br>
>> The issue with stack memory is that when it returns, that memory is<br>
>> invalid. You can't use it in this instance.<br>
>><br>
>> I think you're going to have problems implementing a TSAdaptCreate_XYZ in<br>
>> Fortran (because the body of that function will need to access private<br>
>> struct members; see below).<br>
>><br>
>> I would implement what you need in C and you can call out to Fortran if<br>
>> you want from inside TSAdaptChoose_YourMethod().<br>
>><br>
>> PETSC_EXTERN PetscErrorCode TSAdaptCreate_DSP(TSAdapt adapt)<br>
>> {<br>
>> TSAdapt_DSP *dsp;<br>
>> PetscErrorCode ierr;<br>
>><br>
>> PetscFunctionBegin;<br>
>> ierr = PetscNewLog(adapt,&dsp);CHKERRQ(ierr);<br>
>> adapt->reject_safety = 1.0; /* unused */<br>
>><br>
>> adapt->data = (void*)dsp;<br>
>> adapt->ops->choose = TSAdaptChoose_DSP;<br>
>> adapt->ops->setfromoptions = TSAdaptSetFromOptions_DSP;<br>
>> adapt->ops->destroy = TSAdaptDestroy_DSP;<br>
>> adapt->ops->view = TSAdaptView_DSP;<br>
>><br>
>> ierr =<br>
>> PetscObjectComposeFunction((PetscObject)adapt,"TSAdaptDSPSetFilter_C",TSAdaptDSPSetFilter_DSP);CHKERRQ(ierr);<br>
>> ierr =<br>
>> PetscObjectComposeFunction((PetscObject)adapt,"TSAdaptDSPSetPID_C",TSAdaptDSPSetPID_DSP);CHKERRQ(ierr);<br>
>><br>
>> ierr = TSAdaptDSPSetFilter_DSP(adapt,"PI42");CHKERRQ(ierr);<br>
>> ierr = TSAdaptRestart_DSP(adapt);CHKERRQ(ierr);<br>
>> PetscFunctionReturn(0);<br>
>> }<br>
>><br>
>> > Can anyone develop that idea ? Because for TSAdaptRegister, i guess the<br>
>> > wrapper would start like :<br>
>> ><br>
>> > PETSC_EXTERN void tsadaptregister_(char *sname,<br>
>> > void<br>
>> (*func)(TSAdapt*,PetscErrorCode*),<br>
>> > PetscErrorCode *ierr,<br>
>> > PETSC_FORTRAN_CHARLEN_T snamelen)<br>
>> ><br>
>> > but then the C TSAdaptRegister function takes a PetscErrorCode<br>
>> > (*func)(TSAdapt) function pointer as argument ... I cannot use any<br>
>> > FORTRAN_CALLBACK here since I do not have any object to hook it to, and I<br>
>> > could not find a similar situation among the pre-existing wrappers. Does<br>
>> > anyone have an idea on how to proceed ?<br>
>> ><br>
>> > Thanks !!<br>
>> ><br>
>> > Thibault<br>
>> ><br>
>> > Le mar. 22 déc. 2020 à 21:20, Barry Smith <<a href="mailto:bsmith@petsc.dev" target="_blank">bsmith@petsc.dev</a>> a écrit :<br>
>> ><br>
>> >><br>
>> >> PetscObjectUseFortranCallback((PetscDS)ctx,<br>
>> >><br>
>> >><br>
>> >> *ierr = PetscObjectSetFortranCallback((PetscObject)*prob<br>
>> >><br>
>> >><br>
>> >> It looks like the problem is that these user provided functions do not<br>
>> >> take a PetscDS directly as an argument so the Fortran callback<br>
>> information<br>
>> >> cannot be obtained from them.<br>
>> >><br>
>> >> The manual page for PetscDSAddBoundary() says<br>
>> >><br>
>> >> - ctx - An optional user context for bcFunc<br>
>> >><br>
>> >> but then when it lists the calling sequence for bcFunc it does not list<br>
>> >> the ctx as an argument, so either the manual page or code is wrong.<br>
>> >><br>
>> >> It looks like you make the ctx be the PetscDS prob argument when you<br>
>> >> call PetscDSAddBoundary<br>
>> >><br>
>> >> In principle this sounds like it might work. I think you need to track<br>
>> >> through the debugger to see if the ctx passed to ourbocofunc() is<br>
>> >> actually the PetscDS prob variable and if not why it is not.<br>
>> >><br>
>> >> Barry<br>
>> >><br>
>> >><br>
>> >> On Dec 22, 2020, at 5:49 AM, Thibault Bridel-Bertomeu <<br>
>> >> <a href="mailto:thibault.bridelbertomeu@gmail.com" target="_blank">thibault.bridelbertomeu@gmail.com</a>> wrote:<br>
>> >><br>
>> >> Dear all,<br>
>> >><br>
>> >> I have hit two snags while implementing the missing wrappers necessary<br>
>> to<br>
>> >> transcribe ex11 to Fortran.<br>
>> >><br>
>> >> First is about the PetscDSAddBoundary wrapper, that I have done so :<br>
>> >><br>
>> >> static PetscErrorCode ourbocofunc(PetscReal time, const PetscReal *c,<br>
>> >> const PetscReal *n, const PetscScalar *a_xI, const PetscScalar *a_xG,<br>
>> void<br>
>> >> *ctx)<br>
>> >> {<br>
>> >> PetscObjectUseFortranCallback((PetscDS)ctx, bocofunc,<br>
>> >> (PetscReal*,const PetscReal*,const<br>
>> >> PetscReal*,const PetscScalar*,const PetscScalar*,void*,PetscErrorCode*),<br>
>> >> (&time,c,n,a_xI,a_xG,ctx,&ierr));<br>
>> >> }<br>
>> >> static PetscErrorCode ourbocofunc_time(PetscReal time, const PetscReal<br>
>> *c,<br>
>> >> const PetscReal *n, const PetscScalar *a_xI, const PetscScalar *a_xG,<br>
>> void<br>
>> >> *ctx)<br>
>> >> {<br>
>> >> PetscObjectUseFortranCallback((PetscDS)ctx, bocofunc_time,<br>
>> >> (PetscReal*,const PetscReal*,const<br>
>> >> PetscReal*,const PetscScalar*,const PetscScalar*,void*,PetscErrorCode*),<br>
>> >> (&time,c,n,a_xI,a_xG,ctx,&ierr));<br>
>> >> }<br>
>> >> PETSC_EXTERN void petscdsaddboundary_(PetscDS *prob,<br>
>> >> DMBoundaryConditionType *type, char *name, char *labelname, PetscInt<br>
>> >> *field, PetscInt *numcomps, PetscInt *comps,<br>
>> >> void (*bcFunc)(void),<br>
>> >> void (*bcFunc_t)(void),<br>
>> >> PetscInt *numids, const PetscInt<br>
>> >> *ids, void *ctx, PetscErrorCode *ierr,<br>
>> >> PETSC_FORTRAN_CHARLEN_T namelen,<br>
>> >> PETSC_FORTRAN_CHARLEN_T labelnamelen)<br>
>> >> {<br>
>> >> char *newname, *newlabelname;<br>
>> >> FIXCHAR(name, namelen, newname);<br>
>> >> FIXCHAR(labelname, labelnamelen, newlabelname);<br>
>> >> *ierr = PetscObjectSetFortranCallback((PetscObject)*prob,<br>
>> >> PETSC_FORTRAN_CALLBACK_CLASS, &bocofunc, (PetscVoidFunction)bcFunc,<br>
>> >> ctx);<br>
>> >> *ierr = PetscObjectSetFortranCallback((PetscObject)*prob,<br>
>> >> PETSC_FORTRAN_CALLBACK_CLASS, &bocofunc_time,<br>
>> (PetscVoidFunction)bcFunc_t,<br>
>> >> ctx);<br>
>> >> *ierr = PetscDSAddBoundary(*prob, *type, newname, newlabelname,<br>
>> >> *field, *numcomps, comps,<br>
>> >> (void (*)(void))ourbocofunc,<br>
>> >> (void (*)(void))ourbocofunc_time,<br>
>> >> *numids, ids, *prob);<br>
>> >> FREECHAR(name, newname);<br>
>> >> FREECHAR(labelname, newlabelname);<br>
>> >> }<br>
>> >><br>
>> >><br>
>> >><br>
>> >> but when I call it in the program, with adequate routines, I obtain the<br>
>> >> following error :<br>
>> >><br>
>> >> [0]PETSC ERROR: --------------------- Error Message<br>
>> --------------------------------------------------------------[0]PETSC<br>
>> ERROR: Corrupt argument:<br>
>> <a href="https://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind[0]PETSC" rel="noreferrer" target="_blank">https://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind[0]PETSC</a><br>
>> ERROR: Fortran callback not set on this object[0]PETSC ERROR: See<br>
>> <a href="https://www.mcs.anl.gov/petsc/documentation/faq.html" rel="noreferrer" target="_blank">https://www.mcs.anl.gov/petsc/documentation/faq.html</a> for trouble<br>
>> shooting.[0]PETSC ERROR: Petsc Development GIT revision:<br>
>> v3.14.2-297-gf36a7edeb8 GIT Date: 2020-12-18 04:42:53 +0000[0]PETSC ERROR:<br>
>> ../../../bin/eulerian3D on a named macbook-pro-de-thibault.home by tbridel<br>
>> Sun Dec 20 15:05:15 2020[0]PETSC ERROR: Configure options --with-clean=0<br>
>> --prefix=/Users/tbridel/Documents/1-CODES/04-PETSC/build --with-make-np=2<br>
>> --with-windows-graphics=0 --with-debugging=0 --download-fblaslapack<br>
>> --download-mpich-shared=0 --with-x=0 --with-pthread=0 --with-valgrind=0<br>
>> --PETSC_ARCH=macosx<br>
>> --with-fc=/usr/local/Cellar/open-mpi/4.0.1_2/bin/mpifort<br>
>> --with-cc=/usr/local/Cellar/open-mpi/4.0.1_2/bin/mpicc<br>
>> --with-cxx=/usr/local/Cellar/open-mpi/4.0.1_2/bin/mpic++ --with-openmp=0<br>
>> --download-hypre=yes --download-sowing=yes --download-metis=yes<br>
>> --download-parmetis=yes --download-triangle=yes --download-tetgen=yes<br>
>> --download-ctetgen=yes --download-p4est=yes --download-zlib=yes<br>
>> --download-c2html=yes --download-eigen=yes --download-pragmatic=yes<br>
>> --with-hdf5-dir=/usr/local/Cellar/hdf5/1.10.5_1<br>
>> --with-cmake-dir=/usr/local/Cellar/cmake/3.15.3[0]PETSC ERROR: #1<br>
>> PetscObjectGetFortranCallback() line 258 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/sys/objects/inherit.c[0]PETSC<br>
>> ERROR: #2 ourbocofunc() line 141 in<br>
>> /Users/tbridel/Documents/1-CODES/59-EULERIAN3D/sources/petsc_wrapping/wrapper_petsc.c[0]PETSC<br>
>> ERROR: #3 DMPlexInsertBoundaryValuesRiemann() line 989 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/dm/impls/plex/plexfem.c[0]PETSC<br>
>> ERROR: #4 DMPlexInsertBoundaryValues_Plex() line 1052 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/dm/impls/plex/plexfem.c[0]PETSC<br>
>> ERROR: #5 DMPlexInsertBoundaryValues() line 1142 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/dm/impls/plex/plexfem.c[0]PETSC<br>
>> ERROR: #6 DMPlexComputeResidual_Internal() line 4524 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/dm/impls/plex/plexfem.c[0]PETSC<br>
>> ERROR: #7 DMPlexTSComputeRHSFunctionFVM() line 74 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/ts/utils/dmplexts.c[0]PETSC<br>
>> ERROR: #8 ourdmtsrhsfunc() line 186 in<br>
>> /Users/tbridel/Documents/1-CODES/59-EULERIAN3D/sources/petsc_wrapping/wrapper_petsc.c[0]PETSC<br>
>> ERROR: #9 TSComputeRHSFunction_DMLocal() line 105 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/ts/utils/dmlocalts.c[0]PETSC<br>
>> ERROR: #10 TSComputeRHSFunction() line 653 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/ts/interface/ts.c[0]PETSC<br>
>> ERROR: #11 TSSSPStep_RK_3() line 120 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/ts/impls/explicit/ssp/ssp.c[0]PETSC<br>
>> ERROR: #12 TSStep_SSP() line 208 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/ts/impls/explicit/ssp/ssp.c[0]PETSC<br>
>> ERROR: #13 TSStep() line 3757 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/ts/interface/ts.c[0]PETSC<br>
>> ERROR: #14 TSSolve() line 4154 in<br>
>> /Users/tbridel/Documents/1-CODES/04-PETSC/src/ts/interface/ts.c[0]PETSC<br>
>> ERROR: #15 User provided function() line 0 in User file<br>
>> >><br>
>> >><br>
>> >> Second is about the DMProjectFunction wrapper, that I have done so :<br>
>> >><br>
>> >> static PetscErrorCode ourdmprojfunc(PetscInt dim, PetscReal time,<br>
>> >> PetscReal x[], PetscInt Nf, PetscScalar u[], void *ctx)<br>
>> >> {<br>
>> >> PetscObjectUseFortranCallback((DM)ctx, dmprojfunc,<br>
>> >><br>
>> >><br>
>> (PetscInt*,PetscReal*,PetscReal*,PetscInt*,PetscScalar*,void*,PetscErrorCode*),<br>
>> >> (&dim,&time,x,&Nf,u,_ctx,&ierr))<br>
>> >> }<br>
>> >> PETSC_EXTERN void dmprojectfunction_(DM *dm, PetscReal *time,<br>
>> >> void<br>
>> >><br>
>> (*func)(PetscInt*,PetscReal*,PetscReal*,PetscInt*,PetscScalar*,void*,PetscErrorCode*),<br>
>> >> void *ctx, InsertMode *mode, Vec X,<br>
>> >> PetscErrorCode *ierr)<br>
>> >> {<br>
>> >> PetscErrorCode (*funcarr[1]) (PetscInt dim, PetscReal time,<br>
>> PetscReal<br>
>> >> x[], PetscInt Nf, PetscScalar *u, void *ctx);<br>
>> >> *ierr = PetscObjectSetFortranCallback((PetscObject)*dm,<br>
>> >> PETSC_FORTRAN_CALLBACK_CLASS, &dmprojfunc, (PetscVoidFunction)func,<br>
>> ctx);<br>
>> >> funcarr[0] = ourdmprojfunc;<br>
>> >> *ierr = DMProjectFunction(*dm, *time, funcarr, &ctx, *mode, X);<br>
>> >> }<br>
>> >><br>
>> >><br>
>> >> This time there is no error because I cannot reach this point in the<br>
>> >> program, but I am not sure anyways how to write this wrapper, especially<br>
>> >> because of the double pointers that DMProjectFunction takes as<br>
>> arguments.<br>
>> >><br>
>> >> Does anyone have any idea what could be going wrong with those two<br>
>> >> wrappers ?<br>
>> >><br>
>> >> Thank you very much in advance !!<br>
>> >><br>
>> >> Thibault<br>
>> >><br>
>> >> Le ven. 18 déc. 2020 à 11:02, Thibault Bridel-Bertomeu <<br>
>> >> <a href="mailto:thibault.bridelbertomeu@gmail.com" target="_blank">thibault.bridelbertomeu@gmail.com</a>> a écrit :<br>
>> >><br>
>> >>> Aah that is a nice trick, I was getting ready to fork, clone the fork<br>
>> and<br>
>> >>> redo the work, but that worked fine ! Thank you Barry !<br>
>> >>><br>
>> >>> The MR will appear in a little while !<br>
>> >>><br>
>> >>> Thibault<br>
>> >>><br>
>> >>><br>
>> >>> Le ven. 18 déc. 2020 à 10:16, Barry Smith <<a href="mailto:bsmith@petsc.dev" target="_blank">bsmith@petsc.dev</a>> a écrit :<br>
>> >>><br>
>> >>>><br>
>> >>>> Good question. There is a trick to limit the amount of work you<br>
>> need<br>
>> >>>> to do with a new fork after you have already made changes with a<br>
>> PETSc<br>
>> >>>> clone, but it looks like we do not document this clearly in the<br>
>> webpages.<br>
>> >>>> (I couldn't find it).<br>
>> >>>><br>
>> >>>> Yes, you do need to make a fork, but after you have made the fork on<br>
>> >>>> the GitLab website (and have done nothing on your machine) edit the<br>
>> file<br>
>> >>>> $PETSC_DIR/.git/config for your clone on your machine<br>
>> >>>><br>
>> >>>> Locate the line that has url = git@gitlab.com:petsc/petsc.git<br>
>> (this<br>
>> >>>> may have an https at the beginning of the line)<br>
>> >>>><br>
>> >>>> Change this line to point to the fork url instead with git@ not<br>
>> >>>> https, which will be pretty much the same URL but with your user id<br>
>> instead<br>
>> >>>> of petsc in the address. Then git push and it will push to your fork.<br>
>> >>>><br>
>> >>>> Now you changes will be in your fork and you can make the MR from<br>
>> your<br>
>> >>>> fork URL on Gitlab. (In other words this editing trick converts your<br>
>> PETSc<br>
>> >>>> clone on your machine to a PETSc fork).<br>
>> >>>><br>
>> >>>> I hope I have explained this clearly enough it goes smoothly.<br>
>> >>>><br>
>> >>>> Barry<br>
>> >>>><br>
>> >>>><br>
>> >>>><br>
>> >>>> On Dec 18, 2020, at 3:00 AM, Thibault Bridel-Bertomeu <<br>
>> >>>> <a href="mailto:thibault.bridelbertomeu@gmail.com" target="_blank">thibault.bridelbertomeu@gmail.com</a>> wrote:<br>
>> >>>><br>
>> >>>> Hello Barry,<br>
>> >>>><br>
>> >>>> I'll start the MR as soon as possible then so that specialists can<br>
>> >>>> indeed have a look. Do I have to fork PETSc to start a MR or are<br>
>> PETSc repo<br>
>> >>>> settings such that can I push a branch from the PETSc clone I got ?<br>
>> >>>><br>
>> >>>> Thibault<br>
>> >>>><br>
>> >>>><br>
>> >>>> Le mer. 16 déc. 2020 à 07:47, Barry Smith <<a href="mailto:bsmith@petsc.dev" target="_blank">bsmith@petsc.dev</a>> a écrit<br>
>> :<br>
>> >>>><br>
>> >>>>><br>
>> >>>>> Thibault,<br>
>> >>>>><br>
>> >>>>> A subdirectory for the example is fine; we have other examples that<br>
>> >>>>> use subdirectories and multiple files.<br>
>> >>>>><br>
>> >>>>> Note: even if you don't have something completely working you can<br>
>> >>>>> still make MR and list it as DRAFT request for comments, some other<br>
>> PETSc<br>
>> >>>>> members who understand the packages you are using and Fortran better<br>
>> than I<br>
>> >>>>> may be able to help as you develop the code.<br>
>> >>>>><br>
>> >>>>> Barry<br>
>> >>>>><br>
>> >>>>><br>
>> >>>>><br>
>> >>>>><br>
>> >>>>> On Dec 16, 2020, at 12:35 AM, Thibault Bridel-Bertomeu <<br>
>> >>>>> <a href="mailto:thibault.bridelbertomeu@gmail.com" target="_blank">thibault.bridelbertomeu@gmail.com</a>> wrote:<br>
>> >>>>><br>
>> >>>>> Hello everyone,<br>
>> >>>>><br>
>> >>>>> Thank you Barry for the feedback.<br>
>> >>>>> OK, yes I'll work up an MR as soon as I have got something working.<br>
>> By<br>
>> >>>>> the way, does the fortran-version of the example have to be a single<br>
>> file ?<br>
>> >>>>> If my push contains a directory with several files (different<br>
>> modules and<br>
>> >>>>> the main), and the Makefile that goes with it, is that ok ?<br>
>> >>>>><br>
>> >>>>> Thibault Bridel-Bertomeu<br>
>> >>>>><br>
>> >>>>><br>
>> >>>>> Le mer. 16 déc. 2020 à 04:46, Barry Smith <<a href="mailto:bsmith@petsc.dev" target="_blank">bsmith@petsc.dev</a>> a<br>
>> écrit :<br>
>> >>>>><br>
>> >>>>>><br>
>> >>>>>> This is great. If you make a branch off of the PETSc git<br>
>> repository<br>
>> >>>>>> with these additions and work on ex11 you can make a merge request<br>
>> and we<br>
>> >>>>>> can run the code easily on all our test systems (for security<br>
>> reasons one<br>
>> >>>>>> of use needs to launch the tests from your MR).<br>
>> >>>>>> <a href="https://docs.petsc.org/en/latest/developers/integration/" rel="noreferrer" target="_blank">https://docs.petsc.org/en/latest/developers/integration/</a><br>
>> >>>>>><br>
>> >>>>>> Barry<br>
>> >>>>>><br>
>> >>>>>><br>
>> >>>>>> On Dec 15, 2020, at 5:35 AM, Thibault Bridel-Bertomeu <<br>
>> >>>>>> <a href="mailto:thibault.bridelbertomeu@gmail.com" target="_blank">thibault.bridelbertomeu@gmail.com</a>> wrote:<br>
>> >>>>>><br>
>> >>>>>> Hello everyone,<br>
>> >>>>>><br>
>> >>>>>> So far, I have the wrappers in the files attached to this e-mail. I<br>
>> >>>>>> still do not know if they work properly - at least the code<br>
>> compiles and<br>
>> >>>>>> the calls to the wrapped-subroutine do not fail - but I wanted to<br>
>> put this<br>
>> >>>>>> here in case someone sees something really wrong with it already.<br>
>> >>>>>><br>
>> >>>>>> Thank you again for your help, I'll try to post updates of the F90<br>
>> >>>>>> version of ex11 regularly in this thread.<br>
>> >>>>>><br>
>> >>>>>> Stay safe,<br>
>> >>>>>><br>
>> >>>>>> Thibault Bridel-Bertomeu<br>
>> >>>>>><br>
>> >>>>>> Le dim. 13 déc. 2020 à 16:39, Jed Brown <<a href="mailto:jed@jedbrown.org" target="_blank">jed@jedbrown.org</a>> a écrit<br>
>> :<br>
>> >>>>>><br>
>> >>>>>>> Thibault Bridel-Bertomeu <<a href="mailto:thibault.bridelbertomeu@gmail.com" target="_blank">thibault.bridelbertomeu@gmail.com</a>><br>
>> writes:<br>
>> >>>>>>><br>
>> >>>>>>> > Thank you Mark for your answer.<br>
>> >>>>>>> ><br>
>> >>>>>>> > I am not sure what you think could be in the setBC1 routine ? How<br>
>> >>>>>>> to make<br>
>> >>>>>>> > the connection with the PetscDS ?<br>
>> >>>>>>> ><br>
>> >>>>>>> > On the other hand, I actually found after a while TSMonitorSet<br>
>> has a<br>
>> >>>>>>> > fortran wrapper, and it does take as arguments two function<br>
>> >>>>>>> pointers, so I<br>
>> >>>>>>> > guess it is possible ? Although I am not sure exactly how to play<br>
>> >>>>>>> with the<br>
>> >>>>>>> > PetscObjectSetFortranCallback & PetscObjectUseFortranCallback<br>
>> >>>>>>> macros -<br>
>> >>>>>>> > could anybody advise please ?<br>
>> >>>>>>><br>
>> >>>>>>> tsmonitorset_ is a good example to follow. In your file, create one<br>
>> >>>>>>> of these static structs with a member for each callback. These are<br>
>> IDs that<br>
>> >>>>>>> will be used as keys for Fortran callbacks and their contexts. The<br>
>> salient<br>
>> >>>>>>> parts of the file are below.<br>
>> >>>>>>><br>
>> >>>>>>> static struct {<br>
>> >>>>>>> PetscFortranCallbackId prestep;<br>
>> >>>>>>> PetscFortranCallbackId poststep;<br>
>> >>>>>>> PetscFortranCallbackId rhsfunction;<br>
>> >>>>>>> PetscFortranCallbackId rhsjacobian;<br>
>> >>>>>>> PetscFortranCallbackId ifunction;<br>
>> >>>>>>> PetscFortranCallbackId ijacobian;<br>
>> >>>>>>> PetscFortranCallbackId monitor;<br>
>> >>>>>>> PetscFortranCallbackId mondestroy;<br>
>> >>>>>>> PetscFortranCallbackId transform;<br>
>> >>>>>>> #if defined(PETSC_HAVE_F90_2PTR_ARG)<br>
>> >>>>>>> PetscFortranCallbackId function_pgiptr;<br>
>> >>>>>>> #endif<br>
>> >>>>>>> } _cb;<br>
>> >>>>>>><br>
>> >>>>>>> /*<br>
>> >>>>>>> Note ctx is the same as ts so we need to get the Fortran context<br>
>> >>>>>>> out of the TS; this gets put in _ctx using the callback ID<br>
>> >>>>>>> */<br>
>> >>>>>>> static PetscErrorCode ourmonitor(TS ts,PetscInt i,PetscReal d,Vec<br>
>> >>>>>>> v,void *ctx)<br>
>> >>>>>>> {<br>
>> >>>>>>><br>
>> >>>>>>><br>
>> PetscObjectUseFortranCallback(ts,_cb.monitor,(TS*,PetscInt*,PetscReal*,Vec<br>
>> >>>>>>> *,void*,PetscErrorCode*),(&ts,&i,&d,&v,_ctx,&ierr));<br>
>> >>>>>>> }<br>
>> >>>>>>><br>
>> >>>>>>> Then follow as in tsmonitorset_, which sets two callbacks.<br>
>> >>>>>>><br>
>> >>>>>>> PETSC_EXTERN void tsmonitorset_(TS *ts,void<br>
>> >>>>>>> (*func)(TS*,PetscInt*,PetscReal*,Vec*,void*,PetscErrorCode*),void<br>
>> >>>>>>> *mctx,void (*d)(void*,PetscErrorCode*),PetscErrorCode *ierr)<br>
>> >>>>>>> {<br>
>> >>>>>>> CHKFORTRANNULLFUNCTION(d);<br>
>> >>>>>>> if ((PetscVoidFunction)func == (PetscVoidFunction)<br>
>> >>>>>>> tsmonitordefault_) {<br>
>> >>>>>>> *ierr = TSMonitorSet(*ts,(PetscErrorCode<br>
>> >>>>>>><br>
>> (*)(TS,PetscInt,PetscReal,Vec,void*))TSMonitorDefault,*(PetscViewerAndFormat**)mctx,(PetscErrorCode<br>
>> >>>>>>> (*)(void **))PetscViewerAndFormatDestroy);<br>
>> >>>>>>> } else {<br>
>> >>>>>>> *ierr =<br>
>> >>>>>>><br>
>> PetscObjectSetFortranCallback((PetscObject)*ts,PETSC_FORTRAN_CALLBACK_CLASS,&_cb.monitor,(PetscVoidFunction)func,mctx);<br>
>> >>>>>>> *ierr =<br>
>> >>>>>>><br>
>> PetscObjectSetFortranCallback((PetscObject)*ts,PETSC_FORTRAN_CALLBACK_CLASS,&_cb.mondestroy,(PetscVoidFunction)d,mctx);<br>
>> >>>>>>> *ierr = TSMonitorSet(*ts,ourmonitor,*ts,ourmonitordestroy);<br>
>> >>>>>>> }<br>
>> >>>>>>> }<br>
>> >>>>>>><br>
>> >>>>>> <wrapper_petsc.h90><wrapper_petsc.c><br>
>> >>>>>><br>
>> >>>>>><br>
>> >>>>>><br>
>> >>>>><br>
>> >>>><br>
>> >><br>
>><br>
> -- <br>
> Thibault Bridel-Bertomeu<br>
> —<br>
> Eng, MSc, PhD<br>
> Research Engineer<br>
> CEA/CESTA<br>
> 33114 LE BARP<br>
> Tel.: (+33)557046924<br>
> Mob.: (+33)611025322<br>
> Mail: <a href="mailto:thibault.bridelbertomeu@gmail.com" target="_blank">thibault.bridelbertomeu@gmail.com</a><br>
</blockquote></div></div>-- <br><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div dir="ltr"><div><div><div><div><div>Thibault Bridel-Bertomeu<br>—<br></div></div></div></div>Eng, MSc, PhD</div><div>Research Engineer</div><div>CEA/CESTA</div><div>33114 LE BARP</div><div>Tel.: (+33)557046924</div><div>Mob.: (+33)611025322<br></div><div>Mail: <a href="mailto:thibault.bridelbertomeu@gmail.com" target="_blank">thibault.bridelbertomeu@gmail.com</a><br></div></div></div></div></div></div>