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

changes to PETSc-dev bmake system and library locations
by Barry Smith 08 Jun '07

08 Jun '07

PETSc-dev users, After picking Satish's brain, I have made a set of changes to petsc-dev related to compiling and linking programs. Goal: Support the GNU; config/configure.py; make; make install model including all external packages PETSc builds for you. After "make install" PETSC_ARCH should not be needed. Constraints: * Allow skipping the "make install" step and yet having everything fully functional even with shared and dynamic libraries * Allow multiple builds in the non-"make install" approach which you can switch between by changing PETSC_ARCH * Not require any file links * A system that does not mix generated files and non-generated in the same directory in $PETSC_DIR * A system no more complicated then the previous version. Solution: In place, before "make install" petsc-dev/include same as now /bin same as now /conf basically the same as bmake/common was $PETSC_ARCH/include generated includes: petscconf.h petscfix.h .. lib generated libraries bin generated programs conf basically the same as bmake/$PETSC_ARCH/ except not the include files After "make install" prefix/include all includes /bin all programs, including mpiexec, mpicc if generated /conf the stuff previous in bmake/common and bmake/$PETSC_ARCH /lib the libraries, including from external packages The whole trick is that in the PETSc bmake files (now conf files :-)) the $PETSC_ARCH/ disappears in the "make install" version. I have fixed the external packages MPI.py, BlasLapack.py and Chaco.py but the others need to be modified to stick their libraries and includes in the new correct place. The only change you should need to your makefiles is to replace include ${PETSC_DIR}/bmake/common/base with include ${PETSC_DIR}/conf/base Bug reports to petsc-maint(a)mcs.anl.gov questions to petsc-dev(a)mcs.anl.gov Barry

1 0

Re: PETSI
by Barry Smith 06 Jun '07

06 Jun '07

Thank you for your interest in PETSc. The PETSc software is open source licensed under a FreeBSD style license, it may be used freely for commercial and non-commercial purposes, see http://www-unix.mcs.anl.gov/petsc/petsc-as/documentation/copyright.html PETSc provides a large collection of parallel algebraic solvers and tools to help use those solvers in application. It is used in a wide variety of applications http://www-unix.mcs.anl.gov/petsc/petsc-as/publications/index.html Our website http://www-unix.mcs.anl.gov/petsc/petsc-as/index.html contains all the information about PETSc and its use. Feel free to contact use with any specific questions or concerns. Barry Smith On Thu, 31 May 2007, Ian Foster wrote: > Dear Dr. Iarve: > > I have CCed Barry Smith who leads the PETSc effort. > > Regards -- Ian. > > Iarve, Endel V CTR USAF AFRL/MLBC wrote: > > > > Dear Dr. Foster > > > > I would like learn about the capabilities of the PETSI library and the > > conditions of its use in AF projects. > > > > Thank you > > > > Endel > > > > _____________________________________ > > > > Endel V. Iarve, Ph.D. > > > > Distinguished Research Engineer > > > > University of Dayton Research Institute > > > > 300 College Park Avenue > > > > Dayton, OH 45469-0168 > > > > Tel.: (937) 255 9075 > > > > Fax: (937) 258 8075 > > > > E-mail: Endel.Iarve(a)wpafb.af.mil > > > > _____________________________________ > > > >

1 0

Re: Complex matrices with UMFPACK
by Hong Zhang 06 Jun '07

06 Jun '07

Christophe, Thanks for contributing the interface. I'll test it, then include it as the new petsc-umfpack interface. We'll acknowledge your contribution in the source code. I'll let you know after it is included in petsc-dev. Best, Hong On Wed, 6 Jun 2007, Christophe Geuzaine wrote: > > Barry, > > Attached is a new version of the "umfpack.c" interface, which brings > support for complex arithmetic and 64 bit addressing to the UMFPACK LU > solvers: > > - complex arithmetic support is achieved by using a new UMFPACK feature > introduced in UMFACK 4.4, which allows to pass complex matrices directly > in packed format, with real and imaginary parts interleaved. This allows > to use complex matrices in UMFPACK without copying data around (this was > impossible with UMPACK <= 4.3, as these versions required passing real > and imaginary parts separately); > > - 64 bit addressing is achieved by switching to the "long" version of > the UMFPACK API, where all integers are replaced by "UF_long"s. This > does not change anything on 32 bit architectures, but allows to > manipulate >2Gb matrices on 64 bit machines. > > Notes: > > 1) This new version has only been tested with the latest version of > UMFPACK (version 5.1). Everything should work with UMFPACK >= 4.4, but > complex matrices will *not* work with UMFPACK <= 4.3. > > 2) To enable complex UMFPACK matrices one needs to remove the > "#requiresscalar real" directive in > "src/mat/impls/aij/seq/umfpack/makefile" and add "self.complex = 1" in > the "Configure" class in "python/PETSc/packages/UMFPACK.py". > > 3) If you integrate these changes in the official PETSc tree, you should > probably also upgrade "python/PETSc/packages/UMFPACK.py" to download > UMFPACK 5.1 instead of UMFPACK 4.3 (and fix any related configuration > magic...). > > > Cheers, > > Christophe > > > Barry Smith wrote: > > > > > > Christophe, > > > > We didn't plan to do it, but if UMFPACK supports complex > > numbers by compiling in C++ it should be trivial to add. > > Just edit src/mat/impls/aij/seq/makefile remove the > > rule at the top that requires real and see what happens > > when you compile in that directory. You may have to make some > > minor changes to the file to get it working. Properly completely > > the job also requires some changes to python/Petsc/packages/umfpack.py > > > > Good luck, > > > > If UMFPACK supports complex by some non-standard extension > > of C for the complex data type (like gnu gcc provides) it will > > not be so simple because we don't currently PETSc working this way. > > Actually a user was going to add this support but we haven't > > heard back. It may have been too difficult. > > > > Barry > > > > > > On Sun, 25 Jun 2006, Christophe Geuzaine wrote: > > > >> > >> Just a quick question: are there any plans to support complex matrices > >> with UMFPACK in the (relatively near) future? If not, how hard do you > >> think it would be to extend the current integration of UMFPACK to deal > >> with complex numbers? > >> > >> Many thanks for all your work on PETSc: it's proven a very useful and > >> reliable tool over the years! > >> > >> Best, > >> > >> Christophe > >> > >> > > > > > -- > Prof. Christophe Geuzaine > University of Liege, Electrical Engineering and Computer Science > http://www.montefiore.ulg.ac.be/~geuzaine > >

1 0

Re: Complex matrices with UMFPACK
by Christophe Geuzaine 06 Jun '07

06 Jun '07

Barry, Attached is a new version of the "umfpack.c" interface, which brings support for complex arithmetic and 64 bit addressing to the UMFPACK LU solvers: - complex arithmetic support is achieved by using a new UMFPACK feature introduced in UMFACK 4.4, which allows to pass complex matrices directly in packed format, with real and imaginary parts interleaved. This allows to use complex matrices in UMFPACK without copying data around (this was impossible with UMPACK <= 4.3, as these versions required passing real and imaginary parts separately); - 64 bit addressing is achieved by switching to the "long" version of the UMFPACK API, where all integers are replaced by "UF_long"s. This does not change anything on 32 bit architectures, but allows to manipulate >2Gb matrices on 64 bit machines. Notes: 1) This new version has only been tested with the latest version of UMFPACK (version 5.1). Everything should work with UMFPACK >= 4.4, but complex matrices will *not* work with UMFPACK <= 4.3. 2) To enable complex UMFPACK matrices one needs to remove the "#requiresscalar real" directive in "src/mat/impls/aij/seq/umfpack/makefile" and add "self.complex = 1" in the "Configure" class in "python/PETSc/packages/UMFPACK.py". 3) If you integrate these changes in the official PETSc tree, you should probably also upgrade "python/PETSc/packages/UMFPACK.py" to download UMFPACK 5.1 instead of UMFPACK 4.3 (and fix any related configuration magic...). Cheers, Christophe Barry Smith wrote: > > > Christophe, > > We didn't plan to do it, but if UMFPACK supports complex > numbers by compiling in C++ it should be trivial to add. > Just edit src/mat/impls/aij/seq/makefile remove the > rule at the top that requires real and see what happens > when you compile in that directory. You may have to make some > minor changes to the file to get it working. Properly completely > the job also requires some changes to python/Petsc/packages/umfpack.py > > Good luck, > > If UMFPACK supports complex by some non-standard extension > of C for the complex data type (like gnu gcc provides) it will > not be so simple because we don't currently PETSc working this way. > Actually a user was going to add this support but we haven't > heard back. It may have been too difficult. > > Barry > > > On Sun, 25 Jun 2006, Christophe Geuzaine wrote: > >> >> Just a quick question: are there any plans to support complex matrices >> with UMFPACK in the (relatively near) future? If not, how hard do you >> think it would be to extend the current integration of UMFPACK to deal >> with complex numbers? >> >> Many thanks for all your work on PETSc: it's proven a very useful and >> reliable tool over the years! >> >> Best, >> >> Christophe >> >> > -- Prof. Christophe Geuzaine University of Liege, Electrical Engineering and Computer Science http://www.montefiore.ulg.ac.be/~geuzaine #define PETSCMAT_DLL /* Provides an interface to the UMFPACKv5.1 sparse solver This interface uses the "UF_long version" of the UMFPACK API (*_dl_* and *_zl_* routines, instead of *_di_* and *_zi_* routines) so that UMFPACK can address more than 2Gb of memory on 64 bit machines. If sizeof(UF_long) == 32 the interface only allocates the memory necessary for UMFPACK's working arrays (W, Wi and possibly perm_c). If sizeof(UF_long) == 64, in addition to allocating the working arrays, the interface also has to re-allocate the matrix index arrays (ai and aj, which must be stored as UF_long). The interface is implemented for both real and complex arithmetic. Complex numbers are assumed to be in packed format, which requires UMFPACK >= 4.4. */ #include "src/mat/impls/aij/seq/aij.h" EXTERN_C_BEGIN #include "umfpack.h" EXTERN_C_END typedef struct { void *Symbolic, *Numeric; double Info[UMFPACK_INFO], Control[UMFPACK_CONTROL],*W; UF_long *Wi,*ai,*aj,*perm_c; PetscScalar *av; MatStructure flg; PetscTruth PetscMatOdering; /* A few function pointers for inheritance */ PetscErrorCode (*MatDuplicate)(Mat,MatDuplicateOption,Mat*); PetscErrorCode (*MatView)(Mat,PetscViewer); PetscErrorCode (*MatAssemblyEnd)(Mat,MatAssemblyType); PetscErrorCode (*MatLUFactorSymbolic)(Mat,IS,IS,MatFactorInfo*,Mat*); PetscErrorCode (*MatDestroy)(Mat); /* Flag to clean up UMFPACK objects during Destroy */ PetscTruth CleanUpUMFPACK; } Mat_UMFPACK; EXTERN PetscErrorCode MatDuplicate_UMFPACK(Mat,MatDuplicateOption,Mat*); EXTERN_C_BEGIN #undef __FUNCT__ #define __FUNCT__ "MatConvert_UMFPACK_SeqAIJ" PetscErrorCode PETSCMAT_DLLEXPORT MatConvert_UMFPACK_SeqAIJ(Mat A,MatType type,MatReuse reuse,Mat *newmat) { PetscErrorCode ierr; Mat B=*newmat; Mat_UMFPACK *lu=(Mat_UMFPACK*)A->spptr; PetscFunctionBegin; if (reuse == MAT_INITIAL_MATRIX) { ierr = MatDuplicate(A,MAT_COPY_VALUES,&B);CHKERRQ(ierr); } /* Reset the original function pointers */ B->ops->duplicate = lu->MatDuplicate; B->ops->view = lu->MatView; B->ops->assemblyend = lu->MatAssemblyEnd; B->ops->lufactorsymbolic = lu->MatLUFactorSymbolic; B->ops->destroy = lu->MatDestroy; ierr = PetscObjectComposeFunction((PetscObject)B,"MatConvert_seqaij_umfpack_C","",PETSC_NULL);CHKERRQ(ierr); ierr = PetscObjectComposeFunction((PetscObject)B,"MatConvert_umfpack_seqaij_C","",PETSC_NULL);CHKERRQ(ierr); ierr = PetscObjectChangeTypeName((PetscObject)B,MATSEQAIJ);CHKERRQ(ierr); *newmat = B; PetscFunctionReturn(0); } EXTERN_C_END #undef __FUNCT__ #define __FUNCT__ "MatDestroy_UMFPACK" PetscErrorCode MatDestroy_UMFPACK(Mat A) { PetscErrorCode ierr; Mat_UMFPACK *lu=(Mat_UMFPACK*)A->spptr; PetscFunctionBegin; if (lu->CleanUpUMFPACK) { #if defined(PETSC_USE_COMPLEX) umfpack_zl_free_symbolic(&lu->Symbolic); umfpack_zl_free_numeric(&lu->Numeric); #else umfpack_dl_free_symbolic(&lu->Symbolic); umfpack_dl_free_numeric(&lu->Numeric); #endif ierr = PetscFree(lu->Wi);CHKERRQ(ierr); ierr = PetscFree(lu->W);CHKERRQ(ierr); if(sizeof(UF_long) != sizeof(int)){ ierr = PetscFree(lu->ai);CHKERRQ(ierr); ierr = PetscFree(lu->aj);CHKERRQ(ierr); } if (lu->PetscMatOdering) { ierr = PetscFree(lu->perm_c);CHKERRQ(ierr); } } ierr = MatConvert_UMFPACK_SeqAIJ(A,MATSEQAIJ,MAT_REUSE_MATRIX,&A);CHKERRQ(ierr); ierr = (*A->ops->destroy)(A);CHKERRQ(ierr); PetscFunctionReturn(0); } #undef __FUNCT__ #define __FUNCT__ "MatSolve_UMFPACK" PetscErrorCode MatSolve_UMFPACK(Mat A,Vec b,Vec x) { Mat_UMFPACK *lu = (Mat_UMFPACK*)A->spptr; PetscScalar *av=lu->av,*ba,*xa; PetscErrorCode ierr; UF_long *ai=lu->ai,*aj=lu->aj,status; PetscFunctionBegin; /* solve Ax = b by umfpack_*_wsolve */ /* ----------------------------------*/ #if defined(PETSC_USE_COMPLEX) ierr = VecConjugate(b); #endif ierr = VecGetArray(b,&ba); ierr = VecGetArray(x,&xa); #if defined(PETSC_USE_COMPLEX) status = umfpack_zl_wsolve(UMFPACK_At,ai,aj,(double*)av,NULL,(double*)xa,NULL,(double*)ba,NULL, lu->Numeric,lu->Control,lu->Info,lu->Wi,lu->W); umfpack_zl_report_info(lu->Control, lu->Info); if (status < 0){ umfpack_zl_report_status(lu->Control, status); SETERRQ(PETSC_ERR_LIB,"umfpack_zl_wsolve failed"); } #else status = umfpack_dl_wsolve(UMFPACK_At,ai,aj,av,xa,ba, lu->Numeric,lu->Control,lu->Info,lu->Wi,lu->W); umfpack_dl_report_info(lu->Control, lu->Info); if (status < 0){ umfpack_dl_report_status(lu->Control, status); SETERRQ(PETSC_ERR_LIB,"umfpack_dl_wsolve failed"); } #endif ierr = VecRestoreArray(b,&ba); ierr = VecRestoreArray(x,&xa); #if defined(PETSC_USE_COMPLEX) ierr = VecConjugate(b); #endif PetscFunctionReturn(0); } #undef __FUNCT__ #define __FUNCT__ "MatLUFactorNumeric_UMFPACK" PetscErrorCode MatLUFactorNumeric_UMFPACK(Mat A,MatFactorInfo *info,Mat *F) { Mat_UMFPACK *lu=(Mat_UMFPACK*)(*F)->spptr; PetscErrorCode ierr; UF_long *ai=lu->ai,*aj=lu->aj,m=A->rmap.n,status; PetscScalar *av=lu->av; PetscFunctionBegin; /* numeric factorization of A' */ /* ----------------------------*/ #if defined(PETSC_USE_COMPLEX) if (lu->flg == SAME_NONZERO_PATTERN && lu->Numeric){ umfpack_zl_free_numeric(&lu->Numeric); } status = umfpack_zl_numeric(ai,aj,(double*)av,NULL,lu->Symbolic,&lu->Numeric,lu->Control,lu->Info); if (status < 0) SETERRQ(PETSC_ERR_LIB,"umfpack_zl_numeric failed"); /* report numeric factorization of A' when Control[PRL] > 3 */ (void) umfpack_zl_report_numeric(lu->Numeric, lu->Control); #else if (lu->flg == SAME_NONZERO_PATTERN && lu->Numeric){ umfpack_dl_free_numeric(&lu->Numeric); } status = umfpack_dl_numeric(ai,aj,av,lu->Symbolic,&lu->Numeric,lu->Control,lu->Info); if (status < 0) SETERRQ(PETSC_ERR_LIB,"umfpack_dl_numeric failed"); /* report numeric factorization of A' when Control[PRL] > 3 */ (void) umfpack_dl_report_numeric(lu->Numeric, lu->Control); #endif if (lu->flg == DIFFERENT_NONZERO_PATTERN){ /* first numeric factorization */ /* allocate working space to be used by Solve */ ierr = PetscMalloc(m * sizeof(UF_long), &lu->Wi);CHKERRQ(ierr); #if defined(PETSC_USE_COMPLEX) ierr = PetscMalloc(2*5*m * sizeof(double), &lu->W);CHKERRQ(ierr); #else ierr = PetscMalloc(5*m * sizeof(double), &lu->W);CHKERRQ(ierr); #endif } lu->flg = SAME_NONZERO_PATTERN; lu->CleanUpUMFPACK = PETSC_TRUE; PetscFunctionReturn(0); } /* Note the r permutation is ignored */ #undef __FUNCT__ #define __FUNCT__ "MatLUFactorSymbolic_UMFPACK" PetscErrorCode MatLUFactorSymbolic_UMFPACK(Mat A,IS r,IS c,MatFactorInfo *info,Mat *F) { Mat B; Mat_SeqAIJ *mat=(Mat_SeqAIJ*)A->data; Mat_UMFPACK *lu; PetscErrorCode ierr; int i,m=A->rmap.n,n=A->cmap.n,*ra,idx; UF_long status; PetscScalar *av=mat->a; const char *strategy[]={"AUTO","UNSYMMETRIC","SYMMETRIC","2BY2"}, *scale[]={"NONE","SUM","MAX"}; PetscTruth flg; PetscFunctionBegin; /* Create the factorization matrix F */ ierr = MatCreate(A->comm,&B);CHKERRQ(ierr); ierr = MatSetSizes(B,PETSC_DECIDE,PETSC_DECIDE,m,n);CHKERRQ(ierr); ierr = MatSetType(B,A->type_name);CHKERRQ(ierr); ierr = MatSeqAIJSetPreallocation(B,0,PETSC_NULL);CHKERRQ(ierr); B->ops->lufactornumeric = MatLUFactorNumeric_UMFPACK; B->ops->solve = MatSolve_UMFPACK; B->factor = FACTOR_LU; B->assembled = PETSC_TRUE; /* required by -ksp_view */ lu = (Mat_UMFPACK*)(B->spptr); /* initializations */ /* ------------------------------------------------*/ /* get the default control parameters */ #if defined(PETSC_USE_COMPLEX) umfpack_zl_defaults(lu->Control); #else umfpack_dl_defaults(lu->Control); #endif lu->perm_c = PETSC_NULL; /* use defaul UMFPACK col permutation */ lu->Control[UMFPACK_IRSTEP] = 0; /* max num of iterative refinement steps to attempt */ ierr = PetscOptionsBegin(A->comm,A->prefix,"UMFPACK Options","Mat");CHKERRQ(ierr); /* Control parameters used by reporting routiones */ ierr = PetscOptionsReal("-mat_umfpack_prl","Control[UMFPACK_PRL]","None",lu->Control[UMFPACK_PRL],&lu->Control[UMFPACK_PRL],PETSC_NULL);CHKERRQ(ierr); /* Control parameters for symbolic factorization */ ierr = PetscOptionsEList("-mat_umfpack_strategy","ordering and pivoting strategy","None",strategy,4,strategy[0],&idx,&flg);CHKERRQ(ierr); if (flg) { switch (idx){ case 0: lu->Control[UMFPACK_STRATEGY] = UMFPACK_STRATEGY_AUTO; break; case 1: lu->Control[UMFPACK_STRATEGY] = UMFPACK_STRATEGY_UNSYMMETRIC; break; case 2: lu->Control[UMFPACK_STRATEGY] = UMFPACK_STRATEGY_SYMMETRIC; break; case 3: lu->Control[UMFPACK_STRATEGY] = UMFPACK_STRATEGY_2BY2; break; } } ierr = PetscOptionsReal("-mat_umfpack_dense_col","Control[UMFPACK_DENSE_COL]","None",lu->Control[UMFPACK_DENSE_COL],&lu->Control[UMFPACK_DENSE_COL],PETSC_NULL);CHKERRQ(ierr); ierr = PetscOptionsReal("-mat_umfpack_dense_row","Control[UMFPACK_DENSE_ROW]","None",lu->Control[UMFPACK_DENSE_ROW],&lu->Control[UMFPACK_DENSE_ROW],PETSC_NULL);CHKERRQ(ierr); ierr = PetscOptionsReal("-mat_umfpack_amd_dense","Control[UMFPACK_AMD_DENSE]","None",lu->Control[UMFPACK_AMD_DENSE],&lu->Control[UMFPACK_AMD_DENSE],PETSC_NULL);CHKERRQ(ierr); ierr = PetscOptionsReal("-mat_umfpack_block_size","Control[UMFPACK_BLOCK_SIZE]","None",lu->Control[UMFPACK_BLOCK_SIZE],&lu->Control[UMFPACK_BLOCK_SIZE],PETSC_NULL);CHKERRQ(ierr); ierr = PetscOptionsReal("-mat_umfpack_2by2_tolerance","Control[UMFPACK_2BY2_TOLERANCE]","None",lu->Control[UMFPACK_2BY2_TOLERANCE],&lu->Control[UMFPACK_2BY2_TOLERANCE],PETSC_NULL);CHKERRQ(ierr); ierr = PetscOptionsReal("-mat_umfpack_fixq","Control[UMFPACK_FIXQ]","None",lu->Control[UMFPACK_FIXQ],&lu->Control[UMFPACK_FIXQ],PETSC_NULL);CHKERRQ(ierr); ierr = PetscOptionsReal("-mat_umfpack_aggressive","Control[UMFPACK_AGGRESSIVE]","None",lu->Control[UMFPACK_AGGRESSIVE],&lu->Control[UMFPACK_AGGRESSIVE],PETSC_NULL);CHKERRQ(ierr); /* Control parameters used by numeric factorization */ ierr = PetscOptionsReal("-mat_umfpack_pivot_tolerance","Control[UMFPACK_PIVOT_TOLERANCE]","None",lu->Control[UMFPACK_PIVOT_TOLERANCE],&lu->Control[UMFPACK_PIVOT_TOLERANCE],PETSC_NULL);CHKERRQ(ierr); ierr = PetscOptionsReal("-mat_umfpack_sym_pivot_tolerance","Control[UMFPACK_SYM_PIVOT_TOLERANCE]","None",lu->Control[UMFPACK_SYM_PIVOT_TOLERANCE],&lu->Control[UMFPACK_SYM_PIVOT_TOLERANCE],PETSC_NULL);CHKERRQ(ierr); ierr = PetscOptionsEList("-mat_umfpack_scale","Control[UMFPACK_SCALE]","None",scale,3,scale[0],&idx,&flg);CHKERRQ(ierr); if (flg) { switch (idx){ case 0: lu->Control[UMFPACK_SCALE] = UMFPACK_SCALE_NONE; break; case 1: lu->Control[UMFPACK_SCALE] = UMFPACK_SCALE_SUM; break; case 2: lu->Control[UMFPACK_SCALE] = UMFPACK_SCALE_MAX; break; } } ierr = PetscOptionsReal("-mat_umfpack_alloc_init","Control[UMFPACK_ALLOC_INIT]","None",lu->Control[UMFPACK_ALLOC_INIT],&lu->Control[UMFPACK_ALLOC_INIT],PETSC_NULL);CHKERRQ(ierr); ierr = PetscOptionsReal("-mat_umfpack_front_alloc_init","Control[UMFPACK_FRONT_ALLOC_INIT]","None",lu->Control[UMFPACK_FRONT_ALLOC_INIT],&lu->Control[UMFPACK_ALLOC_INIT],PETSC_NULL);CHKERRQ(ierr); ierr = PetscOptionsReal("-mat_umfpack_droptol","Control[UMFPACK_DROPTOL]","None",lu->Control[UMFPACK_DROPTOL],&lu->Control[UMFPACK_DROPTOL],PETSC_NULL);CHKERRQ(ierr); /* Control parameters used by solve */ ierr = PetscOptionsReal("-mat_umfpack_irstep","Control[UMFPACK_IRSTEP]","None",lu->Control[UMFPACK_IRSTEP],&lu->Control[UMFPACK_IRSTEP],PETSC_NULL);CHKERRQ(ierr); /* use Petsc mat ordering (note: size is for the transpose, and PETSc r = Umfpack perm_c) */ ierr = PetscOptionsHasName(PETSC_NULL,"-pc_factor_mat_ordering_type",&lu->PetscMatOdering);CHKERRQ(ierr); if (lu->PetscMatOdering) { ierr = ISGetIndices(r,&ra);CHKERRQ(ierr); ierr = PetscMalloc(m*sizeof(UF_long),&lu->perm_c);CHKERRQ(ierr); /* we cannot simply memcpy on 64 bit archs */ for(i = 0; i < m; i++) lu->perm_c[i] = ra[i]; ierr = ISRestoreIndices(r,&ra);CHKERRQ(ierr); } PetscOptionsEnd(); if(sizeof(UF_long) != sizeof(int)){ /* we cannot directly use mat->i and mat->j on 64 bit archs */ ierr = PetscMalloc((m+1)*sizeof(UF_long),&lu->ai);CHKERRQ(ierr); ierr = PetscMalloc(mat->nz*sizeof(UF_long),&lu->aj);CHKERRQ(ierr); for(i = 0; i < m + 1; i++) lu->ai[i] = mat->i[i]; for(i = 0; i < mat->nz; i++) lu->aj[i] = mat->j[i]; } else{ lu->ai = (UF_long*)mat->i; lu->aj = (UF_long*)mat->j; } /* print the control parameters */ #if defined(PETSC_USE_COMPLEX) if(lu->Control[UMFPACK_PRL] > 1) umfpack_zl_report_control(lu->Control); #else if(lu->Control[UMFPACK_PRL] > 1) umfpack_dl_report_control(lu->Control); #endif /* symbolic factorization of A' */ /* ---------------------------------------------------------------------- */ #if defined(PETSC_USE_COMPLEX) if (lu->PetscMatOdering) { /* use Petsc row ordering */ status = umfpack_zl_qsymbolic(n,m,lu->ai,lu->aj,(double*)av,NULL, lu->perm_c,&lu->Symbolic,lu->Control,lu->Info); } else { /* use Umfpack col ordering */ status = umfpack_zl_symbolic(n,m,lu->ai,lu->aj,(double*)av,NULL, &lu->Symbolic,lu->Control,lu->Info); } if (status < 0){ umfpack_zl_report_info(lu->Control, lu->Info); umfpack_zl_report_status(lu->Control, status); SETERRQ(PETSC_ERR_LIB,"umfpack_dl_symbolic failed"); } /* report sumbolic factorization of A' when Control[PRL] > 3 */ (void) umfpack_zl_report_symbolic(lu->Symbolic, lu->Control); #else if (lu->PetscMatOdering) { /* use Petsc row ordering */ status = umfpack_dl_qsymbolic(n,m,lu->ai,lu->aj,av, lu->perm_c,&lu->Symbolic,lu->Control,lu->Info); } else { /* use Umfpack col ordering */ status = umfpack_dl_symbolic(n,m,lu->ai,lu->aj,av, &lu->Symbolic,lu->Control,lu->Info); } if (status < 0){ umfpack_dl_report_info(lu->Control, lu->Info); umfpack_dl_report_status(lu->Control, status); SETERRQ(PETSC_ERR_LIB,"umfpack_dl_symbolic failed"); } /* report sumbolic factorization of A' when Control[PRL] > 3 */ (void) umfpack_dl_report_symbolic(lu->Symbolic, lu->Control); #endif lu->flg = DIFFERENT_NONZERO_PATTERN; lu->av = av; lu->CleanUpUMFPACK = PETSC_TRUE; *F = B; PetscFunctionReturn(0); } #undef __FUNCT__ #define __FUNCT__ "MatAssemblyEnd_UMFPACK" PetscErrorCode MatAssemblyEnd_UMFPACK(Mat A,MatAssemblyType mode) { PetscErrorCode ierr; Mat_UMFPACK *lu=(Mat_UMFPACK*)(A->spptr); PetscFunctionBegin; ierr = (*lu->MatAssemblyEnd)(A,mode);CHKERRQ(ierr); lu->MatLUFactorSymbolic = A->ops->lufactorsymbolic; A->ops->lufactorsymbolic = MatLUFactorSymbolic_UMFPACK; PetscFunctionReturn(0); } /* used by -ksp_view */ #undef __FUNCT__ #define __FUNCT__ "MatFactorInfo_UMFPACK" PetscErrorCode MatFactorInfo_UMFPACK(Mat A,PetscViewer viewer) { Mat_UMFPACK *lu= (Mat_UMFPACK*)A->spptr; PetscErrorCode ierr; PetscFunctionBegin; /* check if matrix is UMFPACK type */ if (A->ops->solve != MatSolve_UMFPACK) PetscFunctionReturn(0); ierr = PetscViewerASCIIPrintf(viewer,"UMFPACK run parameters:\n");CHKERRQ(ierr); /* Control parameters used by reporting routiones */ ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_PRL]: %g\n",lu->Control[UMFPACK_PRL]);CHKERRQ(ierr); /* Control parameters used by symbolic factorization */ ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_STRATEGY]: %g\n",lu->Control[UMFPACK_STRATEGY]);CHKERRQ(ierr); ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_DENSE_COL]: %g\n",lu->Control[UMFPACK_DENSE_COL]);CHKERRQ(ierr); ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_DENSE_ROW]: %g\n",lu->Control[UMFPACK_DENSE_ROW]);CHKERRQ(ierr); ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_AMD_DENSE]: %g\n",lu->Control[UMFPACK_AMD_DENSE]);CHKERRQ(ierr); ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_BLOCK_SIZE]: %g\n",lu->Control[UMFPACK_BLOCK_SIZE]);CHKERRQ(ierr); ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_2BY2_TOLERANCE]: %g\n",lu->Control[UMFPACK_2BY2_TOLERANCE]);CHKERRQ(ierr); ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_FIXQ]: %g\n",lu->Control[UMFPACK_FIXQ]);CHKERRQ(ierr); ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_AGGRESSIVE]: %g\n",lu->Control[UMFPACK_AGGRESSIVE]);CHKERRQ(ierr); /* Control parameters used by numeric factorization */ ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_PIVOT_TOLERANCE]: %g\n",lu->Control[UMFPACK_PIVOT_TOLERANCE]);CHKERRQ(ierr); ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_SYM_PIVOT_TOLERANCE]: %g\n",lu->Control[UMFPACK_SYM_PIVOT_TOLERANCE]);CHKERRQ(ierr); ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_SCALE]: %g\n",lu->Control[UMFPACK_SCALE]);CHKERRQ(ierr); ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_ALLOC_INIT]: %g\n",lu->Control[UMFPACK_ALLOC_INIT]);CHKERRQ(ierr); ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_DROPTOL]: %g\n",lu->Control[UMFPACK_DROPTOL]);CHKERRQ(ierr); /* Control parameters used by solve */ ierr = PetscViewerASCIIPrintf(viewer," Control[UMFPACK_IRSTEP]: %g\n",lu->Control[UMFPACK_IRSTEP]);CHKERRQ(ierr); /* mat ordering */ if(!lu->PetscMatOdering) ierr = PetscViewerASCIIPrintf(viewer," UMFPACK default matrix ordering is used (not the PETSc matrix ordering) \n");CHKERRQ(ierr); PetscFunctionReturn(0); } #undef __FUNCT__ #define __FUNCT__ "MatView_UMFPACK" PetscErrorCode MatView_UMFPACK(Mat A,PetscViewer viewer) { PetscErrorCode ierr; PetscTruth iascii; PetscViewerFormat format; Mat_UMFPACK *lu=(Mat_UMFPACK*)(A->spptr); PetscFunctionBegin; ierr = (*lu->MatView)(A,viewer);CHKERRQ(ierr); ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_ASCII,&iascii);CHKERRQ(ierr); if (iascii) { ierr = PetscViewerGetFormat(viewer,&format);CHKERRQ(ierr); if (format == PETSC_VIEWER_ASCII_INFO) { ierr = MatFactorInfo_UMFPACK(A,viewer);CHKERRQ(ierr); } } PetscFunctionReturn(0); } EXTERN_C_BEGIN #undef __FUNCT__ #define __FUNCT__ "MatConvert_SeqAIJ_UMFPACK" PetscErrorCode PETSCMAT_DLLEXPORT MatConvert_SeqAIJ_UMFPACK(Mat A,MatType type,MatReuse reuse,Mat *newmat) { PetscErrorCode ierr; Mat B=*newmat; Mat_UMFPACK *lu; PetscFunctionBegin; if (reuse == MAT_INITIAL_MATRIX) { ierr = MatDuplicate(A,MAT_COPY_VALUES,&B);CHKERRQ(ierr); } ierr = PetscNew(Mat_UMFPACK,&lu);CHKERRQ(ierr); lu->MatDuplicate = A->ops->duplicate; lu->MatView = A->ops->view; lu->MatAssemblyEnd = A->ops->assemblyend; lu->MatLUFactorSymbolic = A->ops->lufactorsymbolic; lu->MatDestroy = A->ops->destroy; lu->CleanUpUMFPACK = PETSC_FALSE; B->spptr = (void*)lu; B->ops->duplicate = MatDuplicate_UMFPACK; B->ops->view = MatView_UMFPACK; B->ops->assemblyend = MatAssemblyEnd_UMFPACK; B->ops->lufactorsymbolic = MatLUFactorSymbolic_UMFPACK; B->ops->destroy = MatDestroy_UMFPACK; ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatConvert_seqaij_umfpack_C", "MatConvert_SeqAIJ_UMFPACK",MatConvert_SeqAIJ_UMFPACK);CHKERRQ(ierr); ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatConvert_umfpack_seqaij_C", "MatConvert_UMFPACK_SeqAIJ",MatConvert_UMFPACK_SeqAIJ);CHKERRQ(ierr); ierr = PetscInfo(0,"Using UMFPACK for SeqAIJ LU factorization and solves.\n");CHKERRQ(ierr); ierr = PetscObjectChangeTypeName((PetscObject)B,MATUMFPACK);CHKERRQ(ierr); *newmat = B; PetscFunctionReturn(0); } EXTERN_C_END #undef __FUNCT__ #define __FUNCT__ "MatDuplicate_UMFPACK" PetscErrorCode MatDuplicate_UMFPACK(Mat A, MatDuplicateOption op, Mat *M) { PetscErrorCode ierr; Mat_UMFPACK *lu=(Mat_UMFPACK*)A->spptr; PetscFunctionBegin; ierr = (*lu->MatDuplicate)(A,op,M);CHKERRQ(ierr); ierr = PetscMemcpy((*M)->spptr,lu,sizeof(Mat_UMFPACK));CHKERRQ(ierr); PetscFunctionReturn(0); } /*MC MATUMFPACK - MATUMFPACK = "umfpack" - A matrix type providing direct solvers (LU) for sequential matrices via the external package UMFPACK. If UMFPACK is installed (see the manual for instructions on how to declare the existence of external packages), a matrix type can be constructed which invokes UMFPACK solvers. After calling MatCreate(...,A), simply call MatSetType(A,UMFPACK). This matrix inherits from MATSEQAIJ. As a result, MatSeqAIJSetPreallocation is supported for this matrix type. One can also call MatConvert for an inplace conversion to or from the MATSEQAIJ type without data copy. Consult UMFPACK documentation for more information about the Control parameters which correspond to the options database keys below. Options Database Keys: + -mat_type umfpack - sets the matrix type to "umfpack" during a call to MatSetFromOptions() . -mat_umfpack_prl - UMFPACK print level: Control[UMFPACK_PRL] . -mat_umfpack_dense_col <alpha_c> - UMFPACK dense column threshold: Control[UMFPACK_DENSE_COL] . -mat_umfpack_block_size <bs> - UMFPACK block size for BLAS-Level 3 calls: Control[UMFPACK_BLOCK_SIZE] . -mat_umfpack_pivot_tolerance <delta> - UMFPACK partial pivot tolerance: Control[UMFPACK_PIVOT_TOLERANCE] . -mat_umfpack_alloc_init <delta> - UMFPACK factorized matrix allocation modifier: Control[UMFPACK_ALLOC_INIT] - -mat_umfpack_irstep <maxit> - UMFPACK maximum number of iterative refinement steps: Control[UMFPACK_IRSTEP] Level: beginner .seealso: PCLU M*/ EXTERN_C_BEGIN #undef __FUNCT__ #define __FUNCT__ "MatCreate_UMFPACK" PetscErrorCode PETSCMAT_DLLEXPORT MatCreate_UMFPACK(Mat A) { PetscErrorCode ierr; PetscFunctionBegin; ierr = MatSetType(A,MATSEQAIJ);CHKERRQ(ierr); ierr = MatConvert_SeqAIJ_UMFPACK(A,MATUMFPACK,MAT_REUSE_MATRIX,&A);CHKERRQ(ierr); PetscFunctionReturn(0); } EXTERN_C_END

1 0

New PETSc release 2.3.3
by Barry Smith 24 May '07

24 May '07

---------------------------------------------------------------------------- Portable, Extensible Toolkit for Scientific Computation (PETSc) ---------------------------------------------------------------------------- http://www.mcs.anl.gov/petsc We are pleased to announce the release of the PETSc 2.3.3 parallel software libraries for the implicit solution of PDEs and related problems. There are a couple of MAJOR interface changes, listed at http://www-unix.mcs.anl.gov/petsc/petsc-as/documentation/changes/233.html please check these changes before compiling your codes. In particular the calling sequence for VecScatterBegin/End() now takes the scatter context as the first argument not the last. You can download from http://www-unix.mcs.anl.gov/petsc/petsc-as/download/index.html and and the installataion instructions are at http://www-unix.mcs.anl.gov/petsc/petsc-as/documentation/installation.html As always, please send bug reports, questions, and requests for new features to petsc-maint(a)mcs.anl.gov Thanks for your continued support. The PETSc developers, Satish, Lisandro, Matt, Victor, Barry and Hong

1 0

Re: change to VecScatterBegin/End
by Mark Adams 18 May '07

18 May '07

Satish, this is done, I compiled the library, but did not test it, and did NOT change the version number (the current version is already associated with the dev version as I recall). Mark On May 8, 2007, at 11:19 PM, Satish Balay wrote: > > The prototypes for VecScatterBegin() and VecScatterEnd() are now > changed. This change is made to have these routines consistant with > PETSc calling convention [i.e the first argument should be the > appropriate object for that functionality]. i.e: > > PetscErrorCode VecScatterBegin > (Vec,Vec,InsertMode,ScatterMode,VecScatter); > to > PetscErrorCode VecScatterBegin > (VecScatter,Vec,Vec,InsertMode,ScatterMode); > > Satish > > ********************************************************************** Mark Adams Ph.D. Columbia University 289 Engineering Terrace MC 4701 New York NY 10027 adams(a)pppl.gov www.columbia.edu/~ma2325 voice: 212.854.4485 fax: 212.854.8257 **********************************************************************

1 0

Re: Sun Studio Performance Library
by Yi-Feng Adam Zhang 17 May '07

17 May '07

Hi Satish, Thanks for your input. I will try to debug it and discuss with the petsc-maint alias. Regards, Adam Satish Balay wrote: > We have it working on a solaris [sparc] box - so I'm not sure what the > problem here is.. > > You can run this in a debugger to chek where it crashing.. If you have > -lsunperf, configure shoul automatically detect it [so no extra option > is necessary] > > Also - looks like -lsunperf requires -lsunmath -lm - so perhaps these > need specifying? > > [petsc-dev(a)mcs.anl.gov is not an appropriate list for such > discussions. You can send these issues to petsc-maint(a)mcs.anl.gov. If > you wish to discuss thse issues in a public list - then you can use > petsc-users(a)mcs.anl.gov list] > > Satish > > On Wed, 16 May 2007, Yi-Feng Adam Zhang wrote: > > >> Hi all, >> >> Does anyone use the latest Sun Studio Performance Library(libsunperf.a) with >> PETSc? I just make a compilation but failed in the test stage. >> >> The configure.py is : >> #!/usr/bin/env python >> if __name__ == '__main__': >> import sys >> sys.path.insert(0, '/work/adam/petsc/perf.mpi.suncc.blas.c/config') >> import configure >> configure_options = ['--with-cc-cc', '--with-shared=0', '--download-mpich=1', >> '-PETSC_ARCH=solaris2.10-c-debug', >> '--with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so', '--with-fc=0'] >> configure.petsc_configure(configure_options) >> >> >> The error message I get after issue "make test": >> -bash-3.00# make test >> Running test examples to verify correct installation >> Possible error running C/C++ src/snes/examples/tutorials/ex19 with 1 MPI >> processSee >> http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html >> [cli_0]: aborting job: >> application called MPI_Abort(MPI_COMM_WORLD, 59) - process 0 >> lid velocity = 0.0016, prandtl # = 1, grashof # = 1 >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably >> memory access out of range >> [0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger >> [0]PETSC ERROR: or see >> http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… >> ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find >> memory corruption errors >> [0]PETSC ERROR: likely location of problem given in stack below >> [0]PETSC ERROR: --------------------- Stack Frames >> ------------------------------------ >> [0]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, >> [0]PETSC ERROR: INSTEAD the line number of the start of the function >> [0]PETSC ERROR: is given. >> [0]PETSC ERROR: [0] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c >> [0]PETSC ERROR: [0] VecNorm line 148 src/vec/vec/interface/rvector.c >> [0]PETSC ERROR: [0] SNESSolve line 1811 src/snes/interface/snes.c >> [0]PETSC ERROR: [0] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c >> [0]PETSC ERROR: [0] DMMGSolve line 250 src/snes/utils/damg.c >> [0]PETSC ERROR: --------------------- Error Message >> ------------------------------------ >> [0]PETSC ERROR: Signal received! >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT >> 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 >> [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 >> 14:07:31 2007 >> [0]PETSC ERROR: Libraries linked from >> /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug >> [0]PETSC ERROR: Configure run at Tue May 15 11:05:42 2007 >> [0]PETSC ERROR: Configure options --with-cc-cc >> --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 >> --with-fc=0 --with-shared=0 >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: User provided function() line 0 in unknown directory unknown >> file >> Possible error running C/C++ src/snes/examples/tutorials/ex19 with 2 MPI >> processes >> See http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html >> [cli_0]: aborting job: >> application called MPI_Abort(MPI_COMM_WORLD, 59) - process 0 >> [cli_1]: aborting job: >> application called MPI_Abort(MPI_COMM_WORLD, 59) - process 1 >> lid velocity = 0.0016, prandtl # = 1, grashof # = 1 >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably >> memory access out of range >> [0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger >> [0]PETSC ERROR: or see >> http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… >> ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find >> memory corruption errors >> [0]PETSC ERROR: likely location of problem given in stack below >> [0]PETSC ERROR: --------------------- Stack Frames >> ------------------------------------ >> [0]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, >> [0]PETSC ERROR: INSTEAD the line number of the start of the function >> [0]PETSC ERROR: is given. >> [0]PETSC ERROR: [0] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c >> [0]PETSC ERROR: [0] VecNorm line 148 src/vec/vec/interface/rvector.c >> [0]PETSC ERR[1]PETSC ERROR: >> ------------------------------------------------------------------------ >> [1]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably >> memory access out of range >> [1]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger >> [1]PETSC ERROR: or see >> http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… >> ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find >> memory corruption errors >> [1]PETSC ERROR: likely location of problem given in stack below >> [1]PETSC ERROR: --------------------- Stack Frames >> ------------------------------------ >> [1]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, >> [1]PETSC ERROR: INSTEAD the line number of the start of the function >> [1]PETSC ERROR: is given. >> [1]PETSC ERROR: [1] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c >> [1]PETSC ERROR: [1] VecNorm line 148 src/vec/vec/interface/rvector.c >> [1]PETSC ERROR: [1] SNESSolve line 1811 src/snes/interface/snes.OR: [0] >> SNESSolve line 1811 src/snes/interface/snes.c >> [0]PETSC ERROR: [0] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c >> [0]PETSC ERROR: [0] DMMGSolve line 250 src/snes/utils/damg.c >> [0]PETSC ERROR: --------------------- Error Message >> ------------------------------------ >> [0]PETSC ERROR: Signal received! >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT >> 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 >> [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 >> 14:07:32 2007 >> [0]PETSC ERROR: Libraries linked from >> /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug >> [0]PETSC ERROR: Configure ruc >> [1]PETSC ERROR: [1] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c >> [1]PETSC ERROR: [1] DMMGSolve line 250 src/snes/utils/damg.c >> [1]PETSC ERROR: --------------------- Error Message >> ------------------------------------ >> [1]PETSC ERROR: Signal received! >> [1]PETSC ERROR: >> ------------------------------------------------------------------------ >> [1]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT >> 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 >> [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 >> 14:07:32 2007 >> [1]PETSC ERROR: Libraries linked from >> /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug >> [1]PETSC ERROR: Configure run at Tue May 15 11:05:42 2007 >> [1]PETSC ERROR: Confign at Tue May 15 11:05:42 2007 >> [0]PETSC ERROR: Configure options --with-cc-cc >> --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 >> --with-fc=0 --with-shared=0 >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: User provided function() line 0 in unknown directory unknown >> file >> ure options --with-cc-cc >> --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 >> --with-fc=0 --with-shared=0 >> [1]PETSC ERROR: >> ------------------------------------------------------------------------ >> [1]PETSC ERROR: User provided function() line 0 in unknown directory unknown >> file >> Possible error running Graphics examples src/snes/examples/tutorials/ex19 1 >> MPI process >> See http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html >> [cli_0]: aborting job: >> application called MPI_Abort(MPI_COMM_WORLD, 59) - process 0 >> lid velocity = 0.0016, prandtl # = 1, grashof # = 1 >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably >> memory access out of range >> [0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger >> [0]PETSC ERROR: or see >> http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… >> ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find >> memory corruption errors >> [0]PETSC ERROR: likely location of problem given in stack below >> [0]PETSC ERROR: --------------------- Stack Frames >> ------------------------------------ >> [0]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, >> [0]PETSC ERROR: INSTEAD the line number of the start of the function >> [0]PETSC ERROR: is given. >> [0]PETSC ERROR: [0] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c >> [0]PETSC ERROR: [0] VecNorm line 148 src/vec/vec/interface/rvector.c >> [0]PETSC ERROR: [0] SNESSolve line 1811 src/snes/interface/snes.c >> [0]PETSC ERROR: [0] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c >> [0]PETSC ERROR: [0] DMMGSolve line 250 src/snes/utils/damg.c >> [0]PETSC ERROR: --------------------- Error Message >> ------------------------------------ >> [0]PETSC ERROR: Signal received! >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT >> 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 >> [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 >> 14:07:34 2007 >> [0]PETSC ERROR: Libraries linked from >> /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug >> [0]PETSC ERROR: Configure run at Tue May 15 11:05:42 2007 >> [0]PETSC ERROR: Configure options --with-cc-cc >> --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 >> --with-fc=0 --with-shared=0 >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: User provided function() line 0 in unknown directory unknown >> file >> Completed test examples >> >> It looks like there is a fatal error in memory operations at the >> PetscFunctionBegin macro. >> >> Regards, >> Adam >> >> >> > >

1 0

Re: Sun Studio Performance Library
by Satish Balay 16 May '07

16 May '07

We have it working on a solaris [sparc] box - so I'm not sure what the problem here is.. You can run this in a debugger to chek where it crashing.. If you have -lsunperf, configure shoul automatically detect it [so no extra option is necessary] Also - looks like -lsunperf requires -lsunmath -lm - so perhaps these need specifying? [petsc-dev(a)mcs.anl.gov is not an appropriate list for such discussions. You can send these issues to petsc-maint(a)mcs.anl.gov. If you wish to discuss thse issues in a public list - then you can use petsc-users(a)mcs.anl.gov list] Satish On Wed, 16 May 2007, Yi-Feng Adam Zhang wrote: > Hi all, > > Does anyone use the latest Sun Studio Performance Library(libsunperf.a) with > PETSc? I just make a compilation but failed in the test stage. > > The configure.py is : > #!/usr/bin/env python > if __name__ == '__main__': > import sys > sys.path.insert(0, '/work/adam/petsc/perf.mpi.suncc.blas.c/config') > import configure > configure_options = ['--with-cc-cc', '--with-shared=0', '--download-mpich=1', > '-PETSC_ARCH=solaris2.10-c-debug', > '--with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so', '--with-fc=0'] > configure.petsc_configure(configure_options) > > > The error message I get after issue "make test": > -bash-3.00# make test > Running test examples to verify correct installation > Possible error running C/C++ src/snes/examples/tutorials/ex19 with 1 MPI > processSee > http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html > [cli_0]: aborting job: > application called MPI_Abort(MPI_COMM_WORLD, 59) - process 0 > lid velocity = 0.0016, prandtl # = 1, grashof # = 1 > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably > memory access out of range > [0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger > [0]PETSC ERROR: or see > http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… > ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find > memory corruption errors > [0]PETSC ERROR: likely location of problem given in stack below > [0]PETSC ERROR: --------------------- Stack Frames > ------------------------------------ > [0]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, > [0]PETSC ERROR: INSTEAD the line number of the start of the function > [0]PETSC ERROR: is given. > [0]PETSC ERROR: [0] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c > [0]PETSC ERROR: [0] VecNorm line 148 src/vec/vec/interface/rvector.c > [0]PETSC ERROR: [0] SNESSolve line 1811 src/snes/interface/snes.c > [0]PETSC ERROR: [0] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c > [0]PETSC ERROR: [0] DMMGSolve line 250 src/snes/utils/damg.c > [0]PETSC ERROR: --------------------- Error Message > ------------------------------------ > [0]PETSC ERROR: Signal received! > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT > 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 > [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 > 14:07:31 2007 > [0]PETSC ERROR: Libraries linked from > /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug > [0]PETSC ERROR: Configure run at Tue May 15 11:05:42 2007 > [0]PETSC ERROR: Configure options --with-cc-cc > --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 > --with-fc=0 --with-shared=0 > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: User provided function() line 0 in unknown directory unknown > file > Possible error running C/C++ src/snes/examples/tutorials/ex19 with 2 MPI > processes > See http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html > [cli_0]: aborting job: > application called MPI_Abort(MPI_COMM_WORLD, 59) - process 0 > [cli_1]: aborting job: > application called MPI_Abort(MPI_COMM_WORLD, 59) - process 1 > lid velocity = 0.0016, prandtl # = 1, grashof # = 1 > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably > memory access out of range > [0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger > [0]PETSC ERROR: or see > http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… > ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find > memory corruption errors > [0]PETSC ERROR: likely location of problem given in stack below > [0]PETSC ERROR: --------------------- Stack Frames > ------------------------------------ > [0]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, > [0]PETSC ERROR: INSTEAD the line number of the start of the function > [0]PETSC ERROR: is given. > [0]PETSC ERROR: [0] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c > [0]PETSC ERROR: [0] VecNorm line 148 src/vec/vec/interface/rvector.c > [0]PETSC ERR[1]PETSC ERROR: > ------------------------------------------------------------------------ > [1]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably > memory access out of range > [1]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger > [1]PETSC ERROR: or see > http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… > ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find > memory corruption errors > [1]PETSC ERROR: likely location of problem given in stack below > [1]PETSC ERROR: --------------------- Stack Frames > ------------------------------------ > [1]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, > [1]PETSC ERROR: INSTEAD the line number of the start of the function > [1]PETSC ERROR: is given. > [1]PETSC ERROR: [1] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c > [1]PETSC ERROR: [1] VecNorm line 148 src/vec/vec/interface/rvector.c > [1]PETSC ERROR: [1] SNESSolve line 1811 src/snes/interface/snes.OR: [0] > SNESSolve line 1811 src/snes/interface/snes.c > [0]PETSC ERROR: [0] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c > [0]PETSC ERROR: [0] DMMGSolve line 250 src/snes/utils/damg.c > [0]PETSC ERROR: --------------------- Error Message > ------------------------------------ > [0]PETSC ERROR: Signal received! > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT > 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 > [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 > 14:07:32 2007 > [0]PETSC ERROR: Libraries linked from > /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug > [0]PETSC ERROR: Configure ruc > [1]PETSC ERROR: [1] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c > [1]PETSC ERROR: [1] DMMGSolve line 250 src/snes/utils/damg.c > [1]PETSC ERROR: --------------------- Error Message > ------------------------------------ > [1]PETSC ERROR: Signal received! > [1]PETSC ERROR: > ------------------------------------------------------------------------ > [1]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT > 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 > [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 > 14:07:32 2007 > [1]PETSC ERROR: Libraries linked from > /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug > [1]PETSC ERROR: Configure run at Tue May 15 11:05:42 2007 > [1]PETSC ERROR: Confign at Tue May 15 11:05:42 2007 > [0]PETSC ERROR: Configure options --with-cc-cc > --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 > --with-fc=0 --with-shared=0 > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: User provided function() line 0 in unknown directory unknown > file > ure options --with-cc-cc > --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 > --with-fc=0 --with-shared=0 > [1]PETSC ERROR: > ------------------------------------------------------------------------ > [1]PETSC ERROR: User provided function() line 0 in unknown directory unknown > file > Possible error running Graphics examples src/snes/examples/tutorials/ex19 1 > MPI process > See http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html > [cli_0]: aborting job: > application called MPI_Abort(MPI_COMM_WORLD, 59) - process 0 > lid velocity = 0.0016, prandtl # = 1, grashof # = 1 > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably > memory access out of range > [0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger > [0]PETSC ERROR: or see > http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… > ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find > memory corruption errors > [0]PETSC ERROR: likely location of problem given in stack below > [0]PETSC ERROR: --------------------- Stack Frames > ------------------------------------ > [0]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, > [0]PETSC ERROR: INSTEAD the line number of the start of the function > [0]PETSC ERROR: is given. > [0]PETSC ERROR: [0] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c > [0]PETSC ERROR: [0] VecNorm line 148 src/vec/vec/interface/rvector.c > [0]PETSC ERROR: [0] SNESSolve line 1811 src/snes/interface/snes.c > [0]PETSC ERROR: [0] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c > [0]PETSC ERROR: [0] DMMGSolve line 250 src/snes/utils/damg.c > [0]PETSC ERROR: --------------------- Error Message > ------------------------------------ > [0]PETSC ERROR: Signal received! > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT > 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 > [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 > 14:07:34 2007 > [0]PETSC ERROR: Libraries linked from > /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug > [0]PETSC ERROR: Configure run at Tue May 15 11:05:42 2007 > [0]PETSC ERROR: Configure options --with-cc-cc > --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 > --with-fc=0 --with-shared=0 > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: User provided function() line 0 in unknown directory unknown > file > Completed test examples > > It looks like there is a fatal error in memory operations at the > PetscFunctionBegin macro. > > Regards, > Adam > >

1 0

Sun Studio Performance Library
by Yi-Feng Adam Zhang 16 May '07

16 May '07

Hi all, Does anyone use the latest Sun Studio Performance Library(libsunperf.a) with PETSc? I just make a compilation but failed in the test stage. The configure.py is : #!/usr/bin/env python if __name__ == '__main__': import sys sys.path.insert(0, '/work/adam/petsc/perf.mpi.suncc.blas.c/config') import configure configure_options = ['--with-cc-cc', '--with-shared=0', '--download-mpich=1', '-PETSC_ARCH=solaris2.10-c-debug', '--with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so', '--with-fc=0'] configure.petsc_configure(configure_options) The error message I get after issue "make test": -bash-3.00# make test Running test examples to verify correct installation Possible error running C/C++ src/snes/examples/tutorials/ex19 with 1 MPI processSee http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html [cli_0]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 59) - process 0 lid velocity = 0.0016, prandtl # = 1, grashof # = 1 [0]PETSC ERROR: ------------------------------------------------------------------------ [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably memory access out of range [0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger [0]PETSC ERROR: or see http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find memory corruption errors [0]PETSC ERROR: likely location of problem given in stack below [0]PETSC ERROR: --------------------- Stack Frames ------------------------------------ [0]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, [0]PETSC ERROR: INSTEAD the line number of the start of the function [0]PETSC ERROR: is given. [0]PETSC ERROR: [0] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c [0]PETSC ERROR: [0] VecNorm line 148 src/vec/vec/interface/rvector.c [0]PETSC ERROR: [0] SNESSolve line 1811 src/snes/interface/snes.c [0]PETSC ERROR: [0] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c [0]PETSC ERROR: [0] DMMGSolve line 250 src/snes/utils/damg.c [0]PETSC ERROR: --------------------- Error Message ------------------------------------ [0]PETSC ERROR: Signal received! [0]PETSC ERROR: ------------------------------------------------------------------------ [0]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 14:07:31 2007 [0]PETSC ERROR: Libraries linked from /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug [0]PETSC ERROR: Configure run at Tue May 15 11:05:42 2007 [0]PETSC ERROR: Configure options --with-cc-cc --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 --with-fc=0 --with-shared=0 [0]PETSC ERROR: ------------------------------------------------------------------------ [0]PETSC ERROR: User provided function() line 0 in unknown directory unknown file Possible error running C/C++ src/snes/examples/tutorials/ex19 with 2 MPI processes See http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html [cli_0]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 59) - process 0 [cli_1]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 59) - process 1 lid velocity = 0.0016, prandtl # = 1, grashof # = 1 [0]PETSC ERROR: ------------------------------------------------------------------------ [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably memory access out of range [0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger [0]PETSC ERROR: or see http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find memory corruption errors [0]PETSC ERROR: likely location of problem given in stack below [0]PETSC ERROR: --------------------- Stack Frames ------------------------------------ [0]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, [0]PETSC ERROR: INSTEAD the line number of the start of the function [0]PETSC ERROR: is given. [0]PETSC ERROR: [0] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c [0]PETSC ERROR: [0] VecNorm line 148 src/vec/vec/interface/rvector.c [0]PETSC ERR[1]PETSC ERROR: ------------------------------------------------------------------------ [1]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably memory access out of range [1]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger [1]PETSC ERROR: or see http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find memory corruption errors [1]PETSC ERROR: likely location of problem given in stack below [1]PETSC ERROR: --------------------- Stack Frames ------------------------------------ [1]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, [1]PETSC ERROR: INSTEAD the line number of the start of the function [1]PETSC ERROR: is given. [1]PETSC ERROR: [1] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c [1]PETSC ERROR: [1] VecNorm line 148 src/vec/vec/interface/rvector.c [1]PETSC ERROR: [1] SNESSolve line 1811 src/snes/interface/snes.OR: [0] SNESSolve line 1811 src/snes/interface/snes.c [0]PETSC ERROR: [0] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c [0]PETSC ERROR: [0] DMMGSolve line 250 src/snes/utils/damg.c [0]PETSC ERROR: --------------------- Error Message ------------------------------------ [0]PETSC ERROR: Signal received! [0]PETSC ERROR: ------------------------------------------------------------------------ [0]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 14:07:32 2007 [0]PETSC ERROR: Libraries linked from /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug [0]PETSC ERROR: Configure ruc [1]PETSC ERROR: [1] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c [1]PETSC ERROR: [1] DMMGSolve line 250 src/snes/utils/damg.c [1]PETSC ERROR: --------------------- Error Message ------------------------------------ [1]PETSC ERROR: Signal received! [1]PETSC ERROR: ------------------------------------------------------------------------ [1]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 14:07:32 2007 [1]PETSC ERROR: Libraries linked from /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug [1]PETSC ERROR: Configure run at Tue May 15 11:05:42 2007 [1]PETSC ERROR: Confign at Tue May 15 11:05:42 2007 [0]PETSC ERROR: Configure options --with-cc-cc --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 --with-fc=0 --with-shared=0 [0]PETSC ERROR: ------------------------------------------------------------------------ [0]PETSC ERROR: User provided function() line 0 in unknown directory unknown file ure options --with-cc-cc --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 --with-fc=0 --with-shared=0 [1]PETSC ERROR: ------------------------------------------------------------------------ [1]PETSC ERROR: User provided function() line 0 in unknown directory unknown file Possible error running Graphics examples src/snes/examples/tutorials/ex19 1 MPI process See http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html [cli_0]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 59) - process 0 lid velocity = 0.0016, prandtl # = 1, grashof # = 1 [0]PETSC ERROR: ------------------------------------------------------------------------ [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably memory access out of range [0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger [0]PETSC ERROR: or see http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html#Si… ERROR: or try http://valgrind.org on linux or man libgmalloc on Apple to find memory corruption errors [0]PETSC ERROR: likely location of problem given in stack below [0]PETSC ERROR: --------------------- Stack Frames ------------------------------------ [0]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, [0]PETSC ERROR: INSTEAD the line number of the start of the function [0]PETSC ERROR: is given. [0]PETSC ERROR: [0] VecNorm_MPI line 57 src/vec/vec/impls/mpi/pvec2.c [0]PETSC ERROR: [0] VecNorm line 148 src/vec/vec/interface/rvector.c [0]PETSC ERROR: [0] SNESSolve line 1811 src/snes/interface/snes.c [0]PETSC ERROR: [0] DMMGSolveSNES line 402 src/snes/utils/damgsnes.c [0]PETSC ERROR: [0] DMMGSolve line 250 src/snes/utils/damg.c [0]PETSC ERROR: --------------------- Error Message ------------------------------------ [0]PETSC ERROR: Signal received! [0]PETSC ERROR: ------------------------------------------------------------------------ [0]PETSC ERROR: Petsc Release Version 2.3.2, Patch 10, Wed Mar 28 19:13:22 CDT 2007 HG revision: d7298c71db7f5e767f359ae35d33cab3bed44428 [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: ./ex19 on a solaris2. named newjersey by root Wed May 16 14:07:34 2007 [0]PETSC ERROR: Libraries linked from /work/adam/petsc/perf.mpi.suncc.blas.c/lib/solaris2.10-c-debug [0]PETSC ERROR: Configure run at Tue May 15 11:05:42 2007 [0]PETSC ERROR: Configure options --with-cc-cc --with-blas-lapack-lib=/opt/SUNWspro/lib/libsunperf.so --download-mpich=1 --with-fc=0 --with-shared=0 [0]PETSC ERROR: ------------------------------------------------------------------------ [0]PETSC ERROR: User provided function() line 0 in unknown directory unknown file Completed test examples It looks like there is a fatal error in memory operations at the PetscFunctionBegin macro. Regards, Adam

1 0

Re: Compiling on Solaris 10 for X86
by Yi-Feng Adam Zhang 10 May '07

10 May '07

Hi Satish, Yes. I just finished the compilation, including BLAS, LAPACK and MPICH and the test stage (bundled in the makefile) with GCC and Sun Studio. PETSc performs great. The code is very wonderful, cross multi-platform indeed. I hope I can maintain a stable Solaris release for PETSc. However my team is very new. Lots of things are still under determination. I will update you via this petsc-dev alias later if I have a firm plan. Thank you! Regards, Adam Satish Balay wrote: > On Fri, 11 May 2007, Yi-Feng Adam Zhang wrote: > > >> I am a Software Engineer, working for Sun Microsystems. My team is >> called Open Source Team. Our intend or goal is to make some >> significant invest on Open Source Community to support them to make >> their software or solution available on Solaris and Sun Grid. Both >> of them are open sourced now: www.opensolaris.org, >> https://sungrid.dev.java.net/ and http://gridengine.sunsource.net/ >> PETSc is a well known with high-profile community in Scientific >> Computing area. I'd like to port PETSc to Solaris OS 10 for X86 >> Platform and make it available on Sun Grid Compute Utility which is >> available from http://www.network.com . If everything is OK both on >> technical side and license side. >> > > http://www-unix.mcs.anl.gov/petsc/petsc-as/documentation/copyright.html > I don't think there is an issue with redistribution from the license > side. > > And PETSc compiles on may platforms with different types of compilers > [including sparc-solaris, sun compilers on linux etc..], so I don't > see any technical issues with here. > > > >> I'd like to release a Solaris Package on the Internet, Maybe PETSc >> website, https://sungrid.dev.java.net/ or other 3rd party >> website. So the developers including Solaris developers or Sun Grid >> Developers can use or adopt the PETSc with a already-built Solaris >> Package. >> > > PETSc keeps evolving with patch updates and new releases every so > often. Do you see the work of providing "PETSc for opensolaris > binaries" as a continues package maintainance effort? [or just a one > time port..] > > Its just that one time ports have been attempted before - and they get > out-of-date before they were even released [so they were not very > useful] > > >> BTW, I find I can't subscribe or use another alias: >> petsc-maint(a)mcs.anl.gov which is mainly for installation & >> compilation questioning. Is this alias still working? >> > > This address is still valid and used for all correspondence to petsc > developers - thats not appropriate/relavent to public mailing lists. > > Satish > >

1 0