Re: [petsc-dev] The problem about running an example in PETSc
Please always use "reply-all" so that your messages go to the list. Lulu Liu <[email protected]> writes:
ierr = DMDACreate(comm, da);CHKERRQ(ierr); ierr = DMDASetDim(*da, 3);CHKERRQ(ierr); // <-- fix this line Why do we fix this line?
Just pass 'dim' so that it works for any dimension.
ierr = DMDASetSizes(*da, M, N, P);CHKERRQ(ierr); I think that we should fix this line...
The error shown like:
Partition in x direction is too fine! -6 1!
Hmm, that error should only be able to occur before the DM is set up. M should be positive after setup.
ierr = DMDASetNumProcs(*da, m, n, p);CHKERRQ(ierr); ierr = DMDASetBoundaryType(*da, bx, by, bz);CHKERRQ(ierr); ierr = DMDASetDof(*da, dof);CHKERRQ(ierr); ierr = DMDASetStencilType(*da, stencil_type);CHKERRQ(ierr); ierr = DMDASetStencilWidth(*da, s);CHKERRQ(ierr); ierr = DMDASetOwnershipRanges(*da, lx, ly, lz);CHKERRQ(ierr); /* This violates the behavior for other classes, but right now users expect negative dimensions to be handled this way */ ierr = DMSetFromOptions(*da);CHKERRQ(ierr); // <--- delete this ierr = DMSetUp(*da);CHKERRQ(ierr); ierr = DMViewFromOptions(*da,"-dm_view");CHKERRQ(ierr); // <--- delete or set a prefix
On Tue, Apr 30, 2013 at 9:20 PM, Jed Brown <[email protected]> wrote:
Lulu Liu <[email protected]> writes:
src/ts/examples/tutorials/ex29.c I want domain partition is 1x2
mpiexec -n 2 ./ex29 -da_refine 4 -ts_max_steps 10 -ts_rtol 1e-3 -ts_atol 1e-3 -ts_type arkimex -ts_monitor -snes_monitor -snes_type ngmres -npc_snes_type nasm -npc_snes_nasm_type restrict -da_overlap 4 -da_processors_x 1 -da_processors_y 2
Thanks for sending the command to reproduce.
0 TS dt 0.05 time 0 0 SNES Function norm 1.762942085583e+01 [0]PETSC ERROR: --------------------- Error Message ------------------------------------ [0]PETSC ERROR: Argument out of range! [0]PETSC ERROR: Too many processors in Y direction: 2 1! [0]PETSC ERROR: [1]PETSC ERROR: --------------------- Error Message ------------------------------------ [1]PETSC ERROR: Argument out of range! [1]PETSC ERROR: Too many processors in Y direction: 2 1! [1]PETSC ERROR: ------------------------------------------------------------------------ [1]PETSC ERROR: Petsc Development GIT revision: bc51b945dfb4f1d1228fb677a132a806066a0f32 GIT Date: 2013-04-29 10:22:29 -0500 ------------------------------------------------------------------------ [0]PETSC ERROR: Petsc Development GIT revision: bc51b945dfb4f1d1228fb677a132a806066a0f32 GIT Date: 2013-04-29 10:22:29 -0500 [0]PETSC ERROR: See docs/changes/index.html for recent updates. [0]PETSC ERROR: See docs/faq.html for hints about trouble shooting. [0]PETSC ERROR: See docs/index.html for manual pages. [0]PETSC ERROR: ------------------------------------------------------------------------ [0]PETSC ERROR: ./ex29 on a arch-basic named kl-12681.local by liul Tue Apr 30 21:07:37 2013 [0]PETSC ERROR: [1]PETSC ERROR: See docs/changes/index.html for recent updates. [1]PETSC ERROR: See docs/faq.html for hints about trouble shooting. [1]PETSC ERROR: See docs/index.html for manual pages. [1]PETSC ERROR: ------------------------------------------------------------------------ [1]PETSC ERROR: ./ex29 on a arch-basic named kl-12681.local by liul Tue Apr 30 21:07:37 2013 [1]PETSC ERROR: Libraries linked from /Users/liul/soft/petsc/arch-basic/lib [1]PETSC ERROR: Configure run at Tue Apr 30 15:46:16 2013 [1]PETSC ERROR: Configure options --download-cmake --download-f-blas-lapack --download-mpich --with-cc=gcc --with-fc=gfortran PETSC_ARCH=arch-basic [1]PETSC ERROR: ------------------------------------------------------------------------ [1]PETSC ERROR: Libraries linked from /Users/liul/soft/petsc/arch-basic/lib [0]PETSC ERROR: Configure run at Tue Apr 30 15:46:16 2013 [0]PETSC ERROR: Configure options --download-cmake --download-f-blas-lapack --download-mpich --with-cc=gcc --with-fc=gfortran PETSC_ARCH=arch-basic [0]PETSC ERROR: ------------------------------------------------------------------------ [0]PETSC ERROR: DMSetUp_DA_2D() line 266 in /Users/liul/soft/petsc/src/dm/impls/da/da2.c DMSetUp_DA_2D() line 266 in /Users/liul/soft/petsc/src/dm/impls/da/da2.c [1]PETSC ERROR: DMSetUp_DA() line 27 in /Users/liul/soft/petsc/src/dm/impls/da/dareg.c [1]PETSC ERROR: DMSetUp() line 469 in /Users/liul/soft/petsc/src/dm/interface/dm.c [1]PETSC ERROR: DMDACreate2d() line 888 in /Users/liul/soft/petsc/src/dm/impls/da/da2.c [1]PETSC ERROR: [0]PETSC ERROR: DMSetUp_DA() line 27 in /Users/liul/soft/petsc/src/dm/impls/da/dareg.c [0]PETSC ERROR: DMSetUp() line 469 in /Users/liul/soft/petsc/src/dm/interface/dm.c DMDAGetReducedDMDA() line 389 in /Users/liul/soft/petsc/src/dm/impls/da/dacorn.c [1]PETSC ERROR: CoefficientSubDomainRestrictHook() line 92 in src/ts/examples/tutorials/ex29.c [1]PETSC ERROR: DMCreateDomainDecomposition() line 1314 in /Users/liul/soft/petsc/src/dm/interface/dm.c [1]PETSC ERROR: SNESSetUp_NASM() line 112 in /Users/liul/soft/petsc/src/snes/impls/nasm/nasm.c [1]PETSC ERROR: SNESSetUp() line 2633 in /Users/liul/soft/petsc/src/snes/interface/snes.c [0]PETSC ERROR: DMDACreate2d() line 888 in /Users/liul/soft/petsc/src/dm/impls/da/da2.c [0]PETSC ERROR: DMDAGetReducedDMDA() line 389 in /Users/liul/soft/petsc/src/dm/impls/da/dacorn.c
if (dim == 1) { ierr = DMDACreate1d(PetscObjectComm((PetscObject)da),bx,M,nfields,s,dd->lx,nda);CHKERRQ(ierr); } else if (dim == 2) { ierr = DMDACreate2d(PetscObjectComm((PetscObject)da),bx,by,stencil_type,M,N,m,n,nfields,s,lx,ly,nda);CHKERRQ(ierr); } else if (dim == 3) { ierr = DMDACreate3d(PetscObjectComm((PetscObject)da),bx,by,bz,stencil_type,M,N,P,m,n,p,nfields,s,lx,ly,lz,nda);CHKERRQ(ierr); }
This code should be fixed to use the precise API instead of the convenience routines DMDACreate*d(), so that it does NOT call DMSetFromOptions (or if it does, call it after setting a prefix).
Lulu, would you be up for taking a crack at this and submitting a patch? Just take the sequence of calls found in DMDACreate3d:
ierr = DMDACreate(comm, da);CHKERRQ(ierr); ierr = DMDASetDim(*da, 3);CHKERRQ(ierr); // <-- fix this line ierr = DMDASetSizes(*da, M, N, P);CHKERRQ(ierr); ierr = DMDASetNumProcs(*da, m, n, p);CHKERRQ(ierr); ierr = DMDASetBoundaryType(*da, bx, by, bz);CHKERRQ(ierr); ierr = DMDASetDof(*da, dof);CHKERRQ(ierr); ierr = DMDASetStencilType(*da, stencil_type);CHKERRQ(ierr); ierr = DMDASetStencilWidth(*da, s);CHKERRQ(ierr); ierr = DMDASetOwnershipRanges(*da, lx, ly, lz);CHKERRQ(ierr); /* This violates the behavior for other classes, but right now users expect negative dimensions to be handled this way */ ierr = DMSetFromOptions(*da);CHKERRQ(ierr); // <--- delete this ierr = DMSetUp(*da);CHKERRQ(ierr); ierr = DMViewFromOptions(*da,"-dm_view");CHKERRQ(ierr); // <--- delete or set a prefix
I can help if you get stuck.
-- Best wishes, Lulu Liu Applied Mathematics and Computational Science King Abdullah University of Science and Technology Tel：＋966－0544701599
--
------------------------------ This message and its contents, including attachments are intended solely for the original recipient. If you are not the intended recipient or have received this message in error, please notify me immediately and delete this message from your computer system. Any unauthorized use or distribution is prohibited. Please consider the environment before printing this email.
participants (1)
-
Jed Brown