On Tue, 3 Apr 2018, Stefano Zampini wrote: > > > On Apr 3, 2018, at 4:58 PM, Satish Balay <[email protected]> wrote: > > > > On Tue, 3 Apr 2018, Kong, Fande wrote: > > > >> On Tue, Apr 3, 2018 at 1:17 AM, Smith, Barry F. <[email protected]> wrote: > >> > >>> > >>> Each external package definitely needs its own duplicated communicator; > >>> cannot share between packages. > >>> > >>> The only problem with the dups below is if they are in a loop and get > >>> called many times. > >>> > >> > >> > >> The "standard test" that has this issue actually has 1K fields. MOOSE > >> creates its own field-split preconditioner (not based on the PETSc > >> fieldsplit), and each filed is associated with one PC HYPRE. If PETSc > >> duplicates communicators, we should easily reach the limit 2048. > >> > >> I also want to confirm what extra communicators are introduced in the bad > >> commit. > > > > To me it looks like there is 1 extra comm created [for MATHYPRE] for each > > PCHYPRE that is created [which also creates one comm for this object]. > > > > You’re right; however, it was the same before the commit.
https://bitbucket.org/petsc/petsc/commits/49a781f5cee36db85e8d5b951eec29f10ac13593 Before the commit - PCHYPRE was not calling MatConvert(MATHYPRE) [this results in an additional call to MPI_Comm_dup() for hypre calls] PCHYPRE was calling MatHYPRE_IJMatrixCreate() directly [which I presume reusing the comm created by the call to MPI_Comm_dup() in PCHYPRE - for hypre calls] > I don’t understand how this specific commit is related with this issue, being > the error not in the MPI_Comm_Dup which is inside MatCreate_MATHYPRE. > Actually, the error comes from MPI_Comm_create > > frame #5: 0x00000001068defd4 libmpi.12.dylib`MPI_Comm_create + 3492 > frame #6: 0x00000001061345d9 > libpetsc.3.07.dylib`hypre_GenerateSubComm(comm=-1006627852, > participate=<unavailable>, new_comm_ptr=<unavailable>) + 409 at > gen_redcs_mat.c:531 [opt] > frame #7: 0x000000010618f8ba > libpetsc.3.07.dylib`hypre_GaussElimSetup(amg_data=0x00007fe7ff857a00, > level=<unavailable>, relax_type=9) + 74 at par_relax.c:4209 [opt] > frame #8: 0x0000000106140e93 > libpetsc.3.07.dylib`hypre_BoomerAMGSetup(amg_vdata=<unavailable>, > A=0x00007fe80842aff0, f=0x00007fe80842a980, u=0x00007fe80842a510) + 17699 at > par_amg_setup.c:2108 [opt] > frame #9: 0x0000000105ec773c > libpetsc.3.07.dylib`PCSetUp_HYPRE(pc=<unavailable>) + 2540 at hypre.c:226 [opt I thought this trace comes up after applying your patch - ierr = MatDestroy(&jac->hpmat);CHKERRQ(ierr); - ierr = MatConvert(pc->pmat,MATHYPRE,MAT_INITIAL_MATRIX,&jac->hpmat);CHKERRQ(ierr); + ierr = MatConvert(pc->pmat,MATHYPRE,jac->hpmat ? MAT_REUSE_MATRIX : MAT_INITIAL_MATRIX,&jac->hpmat);CHKERRQ(ierr); The stack before this patch was: [its a different format - so it was obtained in a different way than the above method?] preconditioners/pbp.lots_of_variables: Other MPI error, error stack: preconditioners/pbp.lots_of_variables: PMPI_Comm_dup(177)..................: MPI_Comm_dup(comm=0x84000001, new_comm=0x97d1068) failed preconditioners/pbp.lots_of_variables: PMPI_Comm_dup(162)..................: preconditioners/pbp.lots_of_variables: MPIR_Comm_dup_impl(57)..............: preconditioners/pbp.lots_of_variables: MPIR_Comm_copy(739).................: preconditioners/pbp.lots_of_variables: MPIR_Get_contextid_sparse_group(614): Too many communicators (0/2048 free on this process; ignore_id=0) Satish > > How did you perform the bisection? make clean + make all ? Which version of > HYPRE are you using? > > > But you might want to verify [by linking with mpi trace library?] > > > > > > There are some debugging hints at > > https://lists.mpich.org/pipermail/discuss/2012-December/000148.html [wrt > > mpich] - which I haven't checked.. > > > > Satish > > > >> > >> > >> Fande, > >> > >> > >> > >>> > >>> To debug the hypre/duplication issue in MOOSE I would run in the > >>> debugger with a break point in MPI_Comm_dup() and see > >>> who keeps calling it an unreasonable amount of times. (My guess is this is > >>> a new "feature" in hypre that they will need to fix but only debugging > >>> will > >>> tell) > >>> > >>> Barry > >>> > >>> > >>>> On Apr 2, 2018, at 7:44 PM, Balay, Satish <[email protected]> wrote: > >>>> > >>>> We do a MPI_Comm_dup() for objects related to externalpackages. > >>>> > >>>> Looks like we added a new mat type MATHYPRE - in 3.8 that PCHYPRE is > >>>> using. Previously there was one MPI_Comm_dup() PCHYPRE - now I think > >>>> is one more for MATHYPRE - so more calls to MPI_Comm_dup in 3.8 vs 3.7 > >>>> > >>>> src/dm/impls/da/hypre/mhyp.c: ierr = MPI_Comm_dup(PetscObjectComm(( > >>> PetscObject)B),&(ex->hcomm));CHKERRQ(ierr); > >>>> src/dm/impls/da/hypre/mhyp.c: ierr = MPI_Comm_dup(PetscObjectComm(( > >>> PetscObject)B),&(ex->hcomm));CHKERRQ(ierr); > >>>> src/dm/impls/swarm/data_ex.c: ierr = MPI_Comm_dup(comm,&d->comm); > >>> CHKERRQ(ierr); > >>>> src/ksp/pc/impls/hypre/hypre.c: ierr = MPI_Comm_dup(PetscObjectComm(( > >>> PetscObject)pc),&(jac->comm_hypre));CHKERRQ(ierr); > >>>> src/ksp/pc/impls/hypre/hypre.c: ierr = MPI_Comm_dup(PetscObjectComm(( > >>> PetscObject)pc),&(ex->hcomm));CHKERRQ(ierr); > >>>> src/ksp/pc/impls/hypre/hypre.c: ierr = MPI_Comm_dup(PetscObjectComm(( > >>> PetscObject)pc),&(ex->hcomm));CHKERRQ(ierr); > >>>> src/ksp/pc/impls/spai/ispai.c: ierr = > >>> MPI_Comm_dup(PetscObjectComm((PetscObject)pc),&(ispai->comm_ > >>> spai));CHKERRQ(ierr); > >>>> src/mat/examples/tests/ex152.c: ierr = MPI_Comm_dup(MPI_COMM_WORLD, > >>> &comm);CHKERRQ(ierr); > >>>> src/mat/impls/aij/mpi/mkl_cpardiso/mkl_cpardiso.c: ierr = > >>> MPI_Comm_dup(PetscObjectComm((PetscObject)A),&(mat_mkl_ > >>> cpardiso->comm_mkl_cpardiso));CHKERRQ(ierr); > >>>> src/mat/impls/aij/mpi/mumps/mumps.c: ierr = > >>> MPI_Comm_dup(PetscObjectComm((PetscObject)A),&(mumps->comm_ > >>> mumps));CHKERRQ(ierr); > >>>> src/mat/impls/aij/mpi/pastix/pastix.c: ierr = > >>> MPI_Comm_dup(PetscObjectComm((PetscObject)A),&(lu->pastix_ > >>> comm));CHKERRQ(ierr); > >>>> src/mat/impls/aij/mpi/superlu_dist/superlu_dist.c: ierr = > >>> MPI_Comm_dup(PetscObjectComm((PetscObject)A),&(lu->comm_ > >>> superlu));CHKERRQ(ierr); > >>>> src/mat/impls/hypre/mhypre.c: ierr = MPI_Comm_dup(PetscObjectComm(( > >>> PetscObject)B),&hB->comm);CHKERRQ(ierr); > >>>> src/mat/partition/impls/pmetis/pmetis.c: ierr = > >>> MPI_Comm_dup(pcomm,&comm);CHKERRQ(ierr); > >>>> src/sys/mpiuni/mpi.c: MPI_COMM_SELF, MPI_COMM_WORLD, and a > >>> MPI_Comm_dup() of each of these (duplicates of duplicates return the same > >>> communictor) > >>>> src/sys/mpiuni/mpi.c:int MPI_Comm_dup(MPI_Comm comm,MPI_Comm *out) > >>>> src/sys/objects/pinit.c: ierr = MPI_Comm_dup(MPI_COMM_WORLD,& > >>> local_comm);CHKERRQ(ierr); > >>>> src/sys/objects/pinit.c: ierr = MPI_Comm_dup(MPI_COMM_WORLD,& > >>> local_comm);CHKERRQ(ierr); > >>>> src/sys/objects/tagm.c: ierr = MPI_Comm_dup(comm_in,comm_out) > >>> ;CHKERRQ(ierr); > >>>> src/sys/utils/mpiu.c: ierr = MPI_Comm_dup(comm,&local_comm) > >>> ;CHKERRQ(ierr); > >>>> src/ts/impls/implicit/sundials/sundials.c: ierr = > >>> MPI_Comm_dup(PetscObjectComm((PetscObject)ts),&(cvode->comm_ > >>> sundials));CHKERRQ(ierr); > >>>> > >>>> Perhaps we need a PetscCommDuplicateExternalPkg() to somehow avoid > >>> these MPI_Comm_dup() calls? > >>>> > >>>> Satish > >>>> > >>>> On Tue, 3 Apr 2018, Smith, Barry F. wrote: > >>>> > >>>>> > >>>>> Are we sure this is a PETSc comm issue and not a hypre comm > >>> duplication issue > >>>>> > >>>>> frame #6: 0x00000001061345d9 libpetsc.3.07.dylib`hypre_ > >>> GenerateSubComm(comm=-1006627852, participate=<unavailable>, > >>> new_comm_ptr=<unavailable>) + 409 at gen_redcs_mat.c:531 [opt] > >>>>> > >>>>> Looks like hypre is needed to generate subcomms, perhaps it generates > >>> too many? > >>>>> > >>>>> Barry > >>>>> > >>>>> > >>>>>> On Apr 2, 2018, at 7:07 PM, Derek Gaston <[email protected]> wrote: > >>>>>> > >>>>>> I’m working with Fande on this and I would like to add a bit more. > >>> There are many circumstances where we aren’t working on COMM_WORLD at all > >>> (e.g. working on a sub-communicator) but PETSc was initialized using > >>> MPI_COMM_WORLD (think multi-level solves)… and we need to create > >>> arbitrarily many PETSc vecs/mats/solvers/preconditioners and solve. We > >>> definitely can’t rely on using PETSC_COMM_WORLD to avoid triggering > >>> duplication. > >>>>>> > >>>>>> Can you explain why PETSc needs to duplicate the communicator so much? > >>>>>> > >>>>>> Thanks for your help in tracking this down! > >>>>>> > >>>>>> Derek > >>>>>> > >>>>>> On Mon, Apr 2, 2018 at 5:44 PM Kong, Fande <[email protected]> wrote: > >>>>>> Why we do not use user-level MPI communicators directly? What are > >>> potential risks here? > >>>>>> > >>>>>> > >>>>>> Fande, > >>>>>> > >>>>>> On Mon, Apr 2, 2018 at 5:08 PM, Satish Balay <[email protected]> > >>> wrote: > >>>>>> PETSC_COMM_WORLD [via PetscCommDuplicate()] attempts to minimize calls > >>> to MPI_Comm_dup() - thus potentially avoiding such errors > >>>>>> > >>>>>> https://urldefense.proofpoint.com/v2/url?u=http-3A__www.mcs. > >>> anl.gov_petsc_petsc-2Dcurrent_docs_manualpages_Sys_ > >>> PetscCommDuplicate.html&d=DwIBAg&c=54IZrppPQZKX9mLzcGdPfFD1hxrcB_ > >>> _aEkJFOKJFd00&r=DUUt3SRGI0_JgtNaS3udV68GRkgV4ts7XKfj2opmi > >>> CY&m=jgv7gpZ3K52d_FWMgkK9yEScbLA7pkrWydFuJnYflsU&s=_ > >>> zpWRcyk3kHuEHoq02NDqYExnXIohXpNnjyabYnnDjU&e= > >>>>>> > >>>>>> > >>>>>> Satish > >>>>>> > >>>>>> On Mon, 2 Apr 2018, Kong, Fande wrote: > >>>>>> > >>>>>>> On Mon, Apr 2, 2018 at 4:23 PM, Satish Balay <[email protected]> > >>> wrote: > >>>>>>> > >>>>>>>> Does this 'standard test' use MPI_COMM_WORLD' to crate PETSc objects? > >>>>>>>> > >>>>>>>> If so - you could try changing to PETSC_COMM_WORLD > >>>>>>>> > >>>>>>> > >>>>>>> > >>>>>>> I do not think we are using PETSC_COMM_WORLD when creating PETSc > >>> objects. > >>>>>>> Why we can not use MPI_COMM_WORLD? > >>>>>>> > >>>>>>> > >>>>>>> Fande, > >>>>>>> > >>>>>>> > >>>>>>>> > >>>>>>>> Satish > >>>>>>>> > >>>>>>>> > >>>>>>>> On Mon, 2 Apr 2018, Kong, Fande wrote: > >>>>>>>> > >>>>>>>>> Hi All, > >>>>>>>>> > >>>>>>>>> I am trying to upgrade PETSc from 3.7.6 to 3.8.3 for MOOSE and its > >>>>>>>>> applications. I have a error message for a standard test: > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> *preconditioners/pbp.lots_of_variables: MPI had an > >>>>>>>>> errorpreconditioners/pbp.lots_of_variables: > >>>>>>>>> ------------------------------------------------ > >>>>>>>> preconditioners/pbp.lots_of_variables: > >>>>>>>>> Other MPI error, error stack:preconditioners/pbp.lots_of_variables: > >>>>>>>>> PMPI_Comm_dup(177)..................: MPI_Comm_dup(comm=0x84000001, > >>>>>>>>> new_comm=0x97d1068) failedpreconditioners/pbp.lots_of_variables: > >>>>>>>>> PMPI_Comm_dup(162)..................: > >>>>>>>>> preconditioners/pbp.lots_of_variables: > >>>>>>>>> MPIR_Comm_dup_impl(57)..............: > >>>>>>>>> preconditioners/pbp.lots_of_variables: > >>>>>>>>> MPIR_Comm_copy(739).................: > >>>>>>>>> preconditioners/pbp.lots_of_variables: > >>>>>>>>> MPIR_Get_contextid_sparse_group(614): Too many communicators > >>> (0/2048 > >>>>>>>> free > >>>>>>>>> on this process; ignore_id=0)* > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> I did "git bisect', and the following commit introduces this issue: > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> *commit 49a781f5cee36db85e8d5b951eec29f10ac13593Author: Stefano > >>> Zampini > >>>>>>>>> <[email protected] <[email protected]>>Date: Sat > >>> Nov 5 > >>>>>>>>> 20:15:19 2016 +0300 PCHYPRE: use internal Mat of type MatHYPRE > >>>>>>>>> hpmat already stores two HYPRE vectors* > >>>>>>>>> > >>>>>>>>> Before I debug line-by-line, anyone has a clue on this? > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> Fande, > >>>>>>>>> > >>>>>>>> > >>>>>>>> > >>>>>>> > >>>>>> > >>>>> > >>> > >>> > >
