Hi, I am taking a slightly different path in the development, so I'll answer later when I come back to that point. Thanks for your help so far
Cheers Timothée Le jeu. 16 janv. 2020 à 15:38, Smith, Barry F. <[email protected]> a écrit : > > > > On Jan 16, 2020, at 3:18 AM, Timothée Nicolas < > [email protected]> wrote: > > > > Actually, for the main solver it works. I'm thinking, could it be due to > the fact that the second SNES instance is defined in a routine that is > called somewhere inside the FormFunction of the main SNES? We are improving > our boundary condition, which becomes quite complex, and we have a small > problem to solve, so I'm trying to handle it with a SNES. So the two SNES > are nested, in a sense. > > This should be fine. We do this. > > Are you sure the inner SNES is actually being called? > > Run with -help | grep green does it print a help message for your > green options? > > Barry > > > > > Timothée > > > > Le mer. 15 janv. 2020 à 23:24, Timothée Nicolas < > [email protected]> a écrit : > > I can actually use some command line arguments. My line arguments > actually read > > > > -snes_mf -green_snes_monitor > > > > and the first -snes_mf argument (for the main solver snes) is correctly > taken into account. > > I will try what Barry suggested, I'll tell you if I find the reason. > > > > Best regards, thanks for your comments > > > > Timothée > > > > Le mer. 15 janv. 2020 à 18:56, Matthew Knepley <[email protected]> a > écrit : > > I think that Mark is suggesting that no command line arguments are > getting in. > > > > Timothee, > > > > Can you use any command line arguments? > > > > Thanks, > > > > Matt > > > > On Wed, Jan 15, 2020 at 12:04 PM Smith, Barry F. via petsc-users < > [email protected]> wrote: > > > > Should still work. Run in the debugger and put a break point in > snessetoptionsprefix_ and see what it is trying to do > > > > Barry > > > > > > > On Jan 15, 2020, at 8:58 AM, Timothée Nicolas < > [email protected]> wrote: > > > > > > Hi, thanks for your answer, > > > > > > I'm using Petsc version 3.10.4 > > > > > > Timothée > > > > > > Le mer. 15 janv. 2020 à 14:59, Mark Adams <[email protected]> a écrit : > > > I'm guessing a Fortran issue. What version of PETSc are you using? > > > > > > On Wed, Jan 15, 2020 at 8:36 AM Timothée Nicolas < > [email protected]> wrote: > > > Dear PETSc users, > > > > > > I am confused by the usage of SNESSetOptionsPrefix. I understand this > is required if you have for example different SNES in your program and want > to set different options for them. > > > So for my second snes I wrote > > > > > > call SNESCreate(MPI_COMM_SELF,snes,ierr) > > > call SNESSetOptionsPrefix(snes,'green_',ierr) > > > call SNESSetFromOptions(snes,ierr) > > > > > > etc. > > > > > > Then when launching the program I wanted to monitor that snes so I > launched it with the option -green_snes_monitor instead of -snes_monitor. > But I keep getting the message > > > > > > WARNING! There are options you set that were not used! > > > WARNING! could be spelling mistake, etc! > > > Option left: name:-green_snes_monitor (no value) > > > > > > What do I miss here? > > > > > > Best regards > > > > > > Timothée NICOLAS > > > > > > > > -- > > What most experimenters take for granted before they begin their > experiments is infinitely more interesting than any results to which their > experiments lead. > > -- Norbert Wiener > > > > https://www.cse.buffalo.edu/~knepley/ > >
