May 2015 - petsc-dev - mailman3.cels.anl.gov

Re: [petsc-dev] VecGetArrayReadF90 ???
by Mark Adams 31 May '15

31 May '15

Got it, its working. Thanks. On Sun, May 31, 2015 at 1:14 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > > I've added the manual pages for master > > > On May 30, 2015, at 2:01 PM, Mark Adams <mfadams(a)lbl.gov> wrote: > > > > Its not here: > > > > > http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/Vec/index.html > > > > On Sat, May 30, 2015 at 12:53 PM, Barry Smith <bsmith(a)mcs.anl.gov> > wrote: > > > > > On May 30, 2015, at 9:55 AM, Mark Adams <mfadams(a)lbl.gov> wrote: > > > > > > There does not appear to be such a beast. > > > > > > I am reading from this array. I guess the compiler will not check that > this is "const", so its not safe (Fortran after all). Should I just copy > this vector or should there be a VecGetArrayReadF90? > > > > There is a VecGetArrayReadF90(). > > > > > > > > Mark > > > > > >

1 0

Re: [petsc-dev] memory leak
by Mark Adams 31 May '15

31 May '15

On Sat, May 30, 2015 at 10:51 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > > Need more details. How can we reproduce this? > I was hoping this would be obvious. a missed free. What tests use Schur solvers? I could see if they have the same problem. > > Barry > > > On May 30, 2015, at 7:10 PM, Mark Adams <mfadams(a)lbl.gov> wrote: > > > > This look like there might be a memory leak in PETSc. Let me know if > this is not obvious and get more detail. > > Mark > > > > [0]Total space allocated 16 bytes > > [ 0]16 bytes PCFieldSplitGetSubKSP_FieldSplit_Schur() line 1263 in > /global/u2/m/madams/petsc/src/ksp/pc/impls/fieldsplit/fieldsplit.c > > [0] PCFieldSplitGetSubKSP_FieldSplit_Schur() line 1263 in > /global/u2/m/madams/petsc/src/ksp/pc/impls/fieldsplit/fieldsplit.c > > [0] PCFieldSplitGetSubKSP() line 1665 in > /global/u2/m/madams/petsc/src/ksp/pc/impls/fieldsplit/fieldsplit.c > > [0] KSPSetUp() line 247 in > /global/u2/m/madams/petsc/src/ksp/ksp/interface/itfunc.c > > > >

1 0

Re: [petsc-dev] VecGetArrayReadF90 ???
by Barry Smith 31 May '15

31 May '15

I've added the manual pages for master > On May 30, 2015, at 2:01 PM, Mark Adams <mfadams(a)lbl.gov> wrote: > > Its not here: > > http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/Vec/index.html > > On Sat, May 30, 2015 at 12:53 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > > > On May 30, 2015, at 9:55 AM, Mark Adams <mfadams(a)lbl.gov> wrote: > > > > There does not appear to be such a beast. > > > > I am reading from this array. I guess the compiler will not check that this is "const", so its not safe (Fortran after all). Should I just copy this vector or should there be a VecGetArrayReadF90? > > There is a VecGetArrayReadF90(). > > > > > Mark > >

1 0

Re: [petsc-dev] memory leak
by Barry Smith 30 May '15

30 May '15

Need more details. How can we reproduce this? Barry > On May 30, 2015, at 7:10 PM, Mark Adams <mfadams(a)lbl.gov> wrote: > > This look like there might be a memory leak in PETSc. Let me know if this is not obvious and get more detail. > Mark > > [0]Total space allocated 16 bytes > [ 0]16 bytes PCFieldSplitGetSubKSP_FieldSplit_Schur() line 1263 in /global/u2/m/madams/petsc/src/ksp/pc/impls/fieldsplit/fieldsplit.c > [0] PCFieldSplitGetSubKSP_FieldSplit_Schur() line 1263 in /global/u2/m/madams/petsc/src/ksp/pc/impls/fieldsplit/fieldsplit.c > [0] PCFieldSplitGetSubKSP() line 1665 in /global/u2/m/madams/petsc/src/ksp/pc/impls/fieldsplit/fieldsplit.c > [0] KSPSetUp() line 247 in /global/u2/m/madams/petsc/src/ksp/ksp/interface/itfunc.c >

1 0

memory leak
by Mark Adams 30 May '15

30 May '15

This look like there might be a memory leak in PETSc. Let me know if this is not obvious and get more detail. Mark [0]Total space allocated 16 bytes [ 0]16 bytes PCFieldSplitGetSubKSP_FieldSplit_Schur() line 1263 in /global/u2/m/madams/petsc/src/ksp/pc/impls/fieldsplit/fieldsplit.c [0] PCFieldSplitGetSubKSP_FieldSplit_Schur() line 1263 in /global/u2/m/madams/petsc/src/ksp/pc/impls/fieldsplit/fieldsplit.c [0] PCFieldSplitGetSubKSP() line 1665 in /global/u2/m/madams/petsc/src/ksp/pc/impls/fieldsplit/fieldsplit.c [0] KSPSetUp() line 247 in /global/u2/m/madams/petsc/src/ksp/ksp/interface/itfunc.c

1 0

Re: [petsc-dev] VecGetArrayReadF90 ???
by Barry Smith 30 May '15

30 May '15

It exists, at least in master, it just doesn't have a manual page. Barry > On May 30, 2015, at 2:01 PM, Mark Adams <mfadams(a)lbl.gov> wrote: > > Its not here: > > http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/Vec/index.html > > On Sat, May 30, 2015 at 12:53 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > > > On May 30, 2015, at 9:55 AM, Mark Adams <mfadams(a)lbl.gov> wrote: > > > > There does not appear to be such a beast. > > > > I am reading from this array. I guess the compiler will not check that this is "const", so its not safe (Fortran after all). Should I just copy this vector or should there be a VecGetArrayReadF90? > > There is a VecGetArrayReadF90(). > > > > > Mark > >

1 0

Re: [petsc-dev] VecGetArrayReadF90 ???
by Mark Adams 30 May '15

30 May '15

Its not here: http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/Vec/index.html On Sat, May 30, 2015 at 12:53 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > > > On May 30, 2015, at 9:55 AM, Mark Adams <mfadams(a)lbl.gov> wrote: > > > > There does not appear to be such a beast. > > > > I am reading from this array. I guess the compiler will not check that > this is "const", so its not safe (Fortran after all). Should I just copy > this vector or should there be a VecGetArrayReadF90? > > There is a VecGetArrayReadF90(). > > > > > Mark > >

1 0

Re: [petsc-dev] VecGetArrayReadF90 ???
by Barry Smith 30 May '15

30 May '15

> On May 30, 2015, at 9:55 AM, Mark Adams <mfadams(a)lbl.gov> wrote: > > There does not appear to be such a beast. > > I am reading from this array. I guess the compiler will not check that this is "const", so its not safe (Fortran after all). Should I just copy this vector or should there be a VecGetArrayReadF90? There is a VecGetArrayReadF90(). > > Mark

1 0

VecGetArrayReadF90 ???
by Mark Adams 30 May '15

30 May '15

There does not appear to be such a beast. I am reading from this array. I guess the compiler will not check that this is "const", so its not safe (Fortran after all). Should I just copy this vector or should there be a VecGetArrayReadF90? Mark

1 0

Re: [petsc-dev] FieldSplit error
by Barry Smith 29 May '15

29 May '15

fixed in master and next with the commit a9de710 Barry > On May 29, 2015, at 1:08 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > > > I will fix this. > > Barry > >> On May 29, 2015, at 10:40 AM, Mark Adams <mfadams(a)lbl.gov> wrote: >> >> I have a 2 x 2 MatComposite and PCSetup is hanging because this branch below does "NEED" to be checked. This is in matrix.c around line 7500. Barry added this a year ago (and the code has been working so I am puzzled). I commented this code out and it works. >> >> The 2nd block here is very small and is all on one processor, so all but one proc have (n == rend-rstart). >> >> This code seems to work in maint, so I am puzzled. Perhaps this code was merged recently? >> Mark >> >> if (!iscol || isrow == iscol) { >> >> PetscBool stride; >> >> ierr = PetscObjectTypeCompare((PetscObject)isrow,ISSTRIDE,&stride);CHKERRQ(ierr); >> >> if (stride) { >> >> PetscInt first,step,n,rstart,rend; >> >> ierr = ISStrideGetInfo(isrow,&first,&step);CHKERRQ(ierr); >> >> if (step == 1) { >> >> ierr = MatGetOwnershipRange(mat,&rstart,&rend);CHKERRQ(ierr); >> >> if (rstart == first) { >> >> ierr = ISGetLocalSize(isrow,&n);CHKERRQ(ierr); >> >> if (n == rend-rstart) { >> >> /* special case grabbing all rows; NEED to do a global reduction to make sure all processes are doing this */ >> >> if (cll == MAT_INITIAL_MATRIX) { >> >> *newmat = mat; >> >> ierr = PetscObjectReference((PetscObject)mat);CHKERRQ(ierr); >> >> } >> >> PetscFunctionReturn(0); >> >> } >> >> } >> >> } >> >> } >> >> } >> >

1 0