[petsc-users] infinite loop with NEWTONTR?

[Barry Smith]

   Run with -info and send ALL the output. -info triggers the printing of all the messages with from the calls to PetscInfo() thus we’ll be able to see what is happening to the values. In theory it should always eventually get out but there code be either a bug in our code or an error in your function evaluation that prevents it from ever getting out. 

   Barry

On Apr 25, 2014, at 7:59 AM, Norihiro Watanabe <norihiro.w at gmail.com> wrote:

> I mean "it's keep running but not print anything". I guess the program is running in while(1) loop after linear solve in SNESSolve_NEWTONTR() in src/snes/impls/tr/tr.c. But I'm not sure if the breaking condition (if (rho > neP->sigma) break; )  can be always satisfied at the end.
> 
> 
> 
> On Fri, Apr 25, 2014 at 2:49 PM, Barry Smith <bsmith at mcs.anl.gov> wrote:
> 
> On Apr 25, 2014, at 7:31 AM, Norihiro Watanabe <norihiro.w at gmail.com> wrote:
> 
> > Hi,
> >
> > In my simulation, nonlinear solve with the trust regtion method got stagnent after linear solve (see output below).
> 
>    What do you mean, get stagnant? Does the code just hang, that is keep running but not print anything.
> 
> > Is it possible that the method goes to inifite loop?
> 
>    This is not suppose to be possible.
> 
> > Is there any parameter to avoid this situation?
> 
>   You need to determine what it is “hanging” on. Try running with -start_in_debugger and when it “hangs” hit control C and type where to determine where it is.
> 
>    Barry
> 
> >
> >   0 SNES Function norm 1.828728087153e+03
> >     0 KSP Residual norm 91.2735
> >     Linear solve converged due to CONVERGED_ITS iterations 1
> >     Linear solve converged due to CONVERGED_RTOL iterations 3
> >     1 KSP Residual norm 3.42223
> >   Linear solve converged due to CONVERGED_STEP_LENGTH iterations 1
> >
> >
> > Thank you in advance,
> > Nori
> 
> 
> 
> 
> -- 
> Norihiro Watanabe