> On Jan 20, 2020, at 6:32 PM, Sam Guo <[email protected]> wrote: > > Hi Barry, > I understand ierr != 0 means something catastrophic. I just want to > release all memory before I exit PETSc.
In general not possible. If you run with the debug version and -malloc_debug it is possible but because of the unknown error it could be that the releasing of the memory causes a real crash. Is your main concern when you use PETSc for a large problem and it errors because it is "out of memory"? Barry > > Thanks, > Sam > > On Mon, Jan 20, 2020 at 4:06 PM Smith, Barry F. <[email protected]> wrote: > > Sam, > > I am not sure what your goal is but PETSc error return codes are error > return codes not exceptions. They mean that something catastrophic happened > and there is no recovery. > > Note that PETSc solvers do not return nonzero error codes on failure to > converge etc. You call, for example, KPSGetConvergedReason() after a KSP > solve to see if it has failed, this is not a catastrophic failure. If a > MatCreate() or any other call returns a nonzero ierr the game is up, you > cannot continue running PETSc. > > Barry > > > > On Jan 20, 2020, at 5:41 PM, Matthew Knepley <[email protected]> wrote: > > > > Not if you initialize the pointers to zero: Mat A = NULL. > > > > Matt > > > > On Mon, Jan 20, 2020 at 6:31 PM Sam Guo <[email protected]> wrote: > > I mean MatDestroy. > > > > On Mon, Jan 20, 2020 at 3:28 PM Sam Guo <[email protected]> wrote: > > Does it hurt to call Destroy function without calling CreateFunction? For > > example > > Mat A, B; > > PetscErrorCode ierr1, ierr2; > > ierr1 = MatCreate(PETSC_COMM_WORLD,&A); > > if(ierr1 == 0) > > { > > ierr2 = MatCreate(PETSC_COMM_WORLD > > ,&B); > > > > } > > if(ierr1 !=0 || ierr2 != 0) > > { > > Destroy(&A); > > Destroy(&B); // if ierr1 !=0, MatCreat is not called on B. Does it hurt > > to call Destroy B here? > > } > > > > > > > > On Mon, Jan 20, 2020 at 11:11 AM Dave May <[email protected]> wrote: > > > > > > On Mon 20. Jan 2020 at 19:47, Sam Guo <[email protected]> wrote: > > Can I assume if there is MatCreat or VecCreate, I should clean up the > > memory myself? > > > > Yes. You will need to call the matching Destroy function. > > > > > > > > On Mon, Jan 20, 2020 at 10:45 AM Sam Guo <[email protected]> wrote: > > I only include the first few lines of SLEPc example. What about following > > ierr = MatCreate(PETSC_COMM_WORLD,&A);CHKERRQ(ierr); > > ierr = MatSetSizes(A,PETSC_DECIDE,PETSC_DECIDE,n,n);CHKERRQ(ierr); > > Is there any memory lost? > > > > On Mon, Jan 20, 2020 at 10:41 AM Dave May <[email protected]> wrote: > > > > > > On Mon 20. Jan 2020 at 19:39, Sam Guo <[email protected]> wrote: > > I don't have a specific case yet. Currently every call of PETSc is checked. > > If ierr is not zero, print the error and return. For example, > > Mat A; /* problem matrix */ > > EPS eps; /* eigenproblem solver context */ > > EPSType type; > > PetscReal error,tol,re,im; > > PetscScalar kr,ki; Vec xr,xi; 25 > > PetscInt n=30,i,Istart,Iend,nev,maxit,its,nconv; > > PetscErrorCode ierr; > > ierr = SlepcInitialize(&argc,&argv,(char*)0,help);CHKERRQ(ierr); > > ierr = PetscOptionsGetInt(NULL,NULL,"-n",&n,NULL);CHKERRQ(ierr); > > ierr = PetscPrintf(PETSC_COMM_WORLD,"\n1-D Laplacian Eigenproblem, > > n=%D\n\n",n);CHKERRQ(ierr); > > > > I am wondering if the memory is lost by calling CHKERRQ. > > > > No. > > > > > > > > On Mon, Jan 20, 2020 at 10:14 AM Dave May <[email protected]> wrote: > > > > > > On Mon 20. Jan 2020 at 19:11, Sam Guo <[email protected]> wrote: > > Dear PETSc dev team, > > If PETSc function returns an error, what's the correct way to clean > > PETSc? > > > > The answer depends on the error message reported. Send the complete error > > message and a better answer can be provided. > > > > Particularly how to clean up the memory? > > > > Totally depends on the objects which aren’t being freed. You need to > > provide more information > > > > Thanks > > Dave > > > > > > Thanks, > > Sam > > > > > > -- > > 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/ >
