[petsc-dev] Something changed with SBAIJ on april 17?

Mon Apr 22 11:42:20 CDT 2019

Eric,

I have update (force pushed) the branch. It should be fixed now, Can you
please confirm?

Thanks

Il giorno lun 22 apr 2019 alle ore 09:00 Stefano Zampini <
stefano.zampini at gmail.com> ha scritto:

> Eric
>
> I'll try to reproduce your setting.
> Few questions.
>
> Is your AIJ matrix lA symmetric?
> Does it have any block size set?
> Do you get the error the first time you do mat copy? Or in subsequent ones?
>
>
> Il Lun 22 Apr 2019, 08:41 Éric Chamberland via petsc-dev <
> petsc-dev at mcs.anl.gov> ha scritto:
>
>> Hi,
>>
>> the patch fixed the 1st kind of error I got ("No support for this
>> operation for this object type"), but I still have all the same faulty
>> tests now returning the 2nd error I reported:
>>
>> [0]PETSC ERROR: --------------------- Error Message --------------------------------------------------------------
>> [0]PETSC ERROR: Argument out of range
>> [0]PETSC ERROR: New nonzero at (0,1) caused a malloc
>> Use MatSetOption(A, MAT_NEW_NONZERO_ALLOCATION_ERR, PETSC_FALSE) to turn off this check
>> [0]PETSC ERROR: See http://www.mcs.anl.gov/petsc/documentation/faq.html for trouble shooting.
>> [0]PETSC ERROR: Petsc Release Version 3.11.1, unknown
>> [0]PETSC ERROR: /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/BIB/bin/Test.deliMEFPP.dev on a  named compile by Unknown Sun Apr 21 18:14:41 2019
>> [0]PETSC ERROR: Configure options --prefix=/opt/petsc-master_debug --with-mpi-compilers=1 --with-mpi-dir=/opt/openmpi-1.10.2 --with-cxx-dialect=C++11 --with-make-np=12 --with-shared-libraries=1 --with-debugging=yes --with-memalign=64 --with-visibility=0 --with-64-bit-indices=0 --download-ml=yes --download-mumps=yes --download-superlu=yes --download-superlu_dist=yes --download-parmetis=yes --download-ptscotch=yes --download-metis=yes --download-suitesparse=yes --download-hypre=yes --with-blaslapack-dir=/opt/intel/composer_xe_2015.2.164/mkl/lib/intel64 --with-mkl_pardiso-dir=/opt/intel/composer_xe_2015.2.164/mkl --with-mkl_cpardiso-dir=/opt/intel/composer_xe_2015.2.164/mkl --with-scalapack=1 --with-scalapack-include=/opt/intel/composer_xe_2015.2.164/mkl/include --with-scalapack-lib="-L/opt/intel/composer_xe_2015.2.164/mkl/lib/intel64 -lmkl_scalapack_lp64 -lmkl_blacs_openmpi_lp64"
>> [0]PETSC ERROR: #1 MatSetValues_SeqSBAIJ() line 992 in /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/impls/sbaij/seq/sbaij.c
>> [0]PETSC ERROR: #2 MatSetValues() line 1407 in /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/interface/matrix.c
>> [0]PETSC ERROR: #3 MatAXPY_BasicWithPreallocation() line 171 in /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/utils/axpy.c
>> [0]PETSC ERROR: #4 MatAXPY() line 52 in /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/utils/axpy.c
>> [0]PETSC ERROR: #5 MatAYPX() line 348 in /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/utils/axpy.c
>> [0]PETSC ERROR: #6 MatCopy_Basic() line 4044 in /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/interface/matrix.c
>> [0]PETSC ERROR: #7 MatCopy_SeqAIJ() line 2844 in /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/impls/aij/seq/aij.c
>> [0]PETSC ERROR: #8 MatCopy() line 4098 in /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/interface/matrix.c
>>
>>
>> We are calling a MatCopy from an aij (lA) to sbaij (lB) matrix with
>> DIFFERENT_NONZERO_PATTERN option.
>>
>> But on first call, the SBAIJ matrix has first been created then filled
>> with MatConvert call:
>>
>> MatConvert(lA,MATSBAIJ, MAT_INITIAL_MATRIX, &lB);
>>
>> I just noticed we then do a:
>> MatSetOption(lA, MAT_SYMMETRIC, PETSC_FALSE);
>> (I have to find why???....)
>>
>> and on other calls, we call the "faulty" MatCopy with previously created
>> matrices:
>>
>> MatCopy(lA,lB,DIFFERENT_NONZERO_PATTERN);
>>
>> which gives the "Argument out of range" error...
>>
>> Thanks,
>>
>> Eric
>>
>> Le 19-04-21 à 22:52, Éric Chamberland via petsc-dev a écrit :
>>
>> Hi Stefano,
>>
>> sure!
>>
>> configuration is running with your branch and tests will begin later.  I
>> will see and share the results tomorrow.
>>
>> thanks! :)
>>
>> Eric
>>
>>
>> Le 19-04-21 à 17:21, Stefano Zampini a écrit :
>>
>> Eric
>>
>> I have opened a PRmay fix the issues you reported
>> https://bitbucket.org/petsc/petsc/pull-requests/1571/fix-sbaij/diff
>> Could you please try
>> https://bitbucket.org/petsc/petsc/branch/stefano_zampini/fix-matcopy-sbaij?
>> (off maint)
>>
>> Thanks
>>
>> Il giorno ven 19 apr 2019 alle ore 23:26 Éric Chamberland via petsc-dev <
>> petsc-dev at mcs.anl.gov> ha scritto:
>>
>>> Hi,
>>>
>>> ok, I will try do to one, but since I am on vacation right now, I will
>>> try to do it at my return in a week.
>>>
>>> I wanted to report the bug asap, so you may figure out the problem
>>> quickly.
>>>
>>> Thanks,
>>>
>>> Eric
>>>
>>>
>>> Le 19-04-19 à 19:10, Stefano Zampini via petsc-dev a écrit :
>>>
>>> Can you please send a MWE to reproduce the errors so that I can provide
>>> a fix?
>>>
>>> Il Ven 19 Apr 2019, 17:00 Matthew Knepley via petsc-dev <
>>> petsc-dev at mcs.anl.gov> ha scritto:
>>>
>>>> On Fri, Apr 19, 2019 at 9:40 AM Balay, Satish via petsc-dev <
>>>> petsc-dev at mcs.anl.gov> wrote:
>>>>
>>>>> There were a few merges in the the past few days.
>>>>>
>>>>> For ex:
>>>>> https://bitbucket.org/petsc/petsc/pull-requests/1547/add-support-for-zeroing-rows-and-columns/diff
>>>>>
>>>>> Perhaps you can narrow down to the commit that triggered this change
>>>>> with git bisect.
>>>>>
>>>>
>>>> I think its this commit:
>>>>
>>>>
>>>> https://bitbucket.org/petsc/petsc/commits/2b4bbc58b197e97e85869bd03594884bbca73c67
>>>>
>>>>     Matt
>>>>
>>>>
>>>>> cc: Stefano
>>>>>
>>>>> Satish
>>>>>
>>>>> On Fri, 19 Apr 2019, Éric Chamberland via petsc-dev wrote:
>>>>>
>>>>> > Hi,
>>>>> >
>>>>> > Looks like something pushed on master change or broke some sbaij
>>>>> usages.  I
>>>>> > have two kind off errors with MatCopy:
>>>>> >
>>>>> > 1) MatCopy error saying: No support for this operation for this
>>>>> object type
>>>>> >
>>>>> > [0]PETSC ERROR: --------------------- Error Message
>>>>> > --------------------------------------------------------------
>>>>> > [0]PETSC ERROR: No support for this operation for this object type
>>>>> > [0]PETSC ERROR: MatGetRow is not supported for SBAIJ matrix format.
>>>>> Getting
>>>>> > the upper triangular part of row, run with
>>>>> -mat_getrow_uppertriangular, call
>>>>> > MatSetOption(mat,MAT_GETROW_UPPERTRIANGULAR,PETSC_TRUE) or
>>>>> > MatGetRowUpperTriangular()
>>>>> > [0]PETSC ERROR: See
>>>>> http://www.mcs.anl.gov/petsc/documentation/faq.html for
>>>>> > trouble shooting.
>>>>> > [0]PETSC ERROR: Petsc Development GIT revision:
>>>>> v3.11.1-277-g63b3610  GIT
>>>>> > Date: 2019-04-18 08:33:56 -0500
>>>>> > [0]PETSC ERROR:
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/BIB/bin/
>>>>> Test.deliMEFPP.dev
>>>>> > on a  named compile by Unknown Fri Apr 19 03:19:09 2019
>>>>> > [0]PETSC ERROR: Configure options --prefix=/opt/petsc-master_debug
>>>>> > --with-mpi-compilers=1 --with-mpi-dir=/opt/openmpi-1.10.2
>>>>> > --with-cxx-dialect=C++11 --with-make-np=12 --with-shared-libraries=1
>>>>> > --with-debugging=yes --with-memalign=64 --with-visibility=0
>>>>> > --with-64-bit-indices=0 --download-ml=yes --download-mumps=yes
>>>>> > --download-superlu=yes --download-superlu_dist=yes
>>>>> --download-parmetis=yes
>>>>> > --download-ptscotch=yes --download-metis=yes
>>>>> --download-suitesparse=yes
>>>>> > --download-hypre=yes
>>>>> >
>>>>> --with-blaslapack-dir=/opt/intel/composer_xe_2015.2.164/mkl/lib/intel64
>>>>> > --with-mkl_pardiso-dir=/opt/intel/composer_xe_2015.2.164/mkl
>>>>> > --with-mkl_cpardiso-dir=/opt/intel/composer_xe_2015.2.164/mkl
>>>>> > --with-scalapack=1
>>>>> >
>>>>> --with-scalapack-include=/opt/intel/composer_xe_2015.2.164/mkl/include
>>>>> >
>>>>> --with-scalapack-lib="-L/opt/intel/composer_xe_2015.2.164/mkl/lib/intel64
>>>>> > -lmkl_scalapack_lp64 -lmkl_blacs_openmpi_lp64"
>>>>> > [0]PETSC ERROR: #1 MatGetRow_SeqSBAIJ() line 281 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/impls/sbaij/seq/sbaij.c
>>>>> > [0]PETSC ERROR: #2 MatGetRow() line 574 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/interface/matrix.c
>>>>> > [0]PETSC ERROR: #3 MatAXPY_Basic_Preallocate() line 138 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/utils/axpy.c
>>>>> > [0]PETSC ERROR: #4 MatAXPY() line 95 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/utils/axpy.c
>>>>> > [0]PETSC ERROR: #5 MatAYPX() line 384 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/utils/axpy.c
>>>>> > [0]PETSC ERROR: #6 MatCopy_Basic() line 4056 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/interface/matrix.c
>>>>> > [0]PETSC ERROR: #7 MatCopy_SeqAIJ() line 2844 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/impls/aij/seq/aij.c
>>>>> > [0]PETSC ERROR: #8 MatCopy() line 4110 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/interface/matrix.c
>>>>> >
>>>>> > 2) MatCopy error saying:
>>>>> >
>>>>> > Argument out of range: New nonzero at (6,7) caused a malloc
>>>>> >
>>>>> > [0]PETSC ERROR: --------------------- Error Message
>>>>> > --------------------------------------------------------------
>>>>> > [0]PETSC ERROR: Argument out of range
>>>>> > [0]PETSC ERROR: New nonzero at (6,7) caused a malloc
>>>>> > Use MatSetOption(A, MAT_NEW_NONZERO_ALLOCATION_ERR, PETSC_FALSE) to
>>>>> turn off
>>>>> > this check
>>>>> > [0]PETSC ERROR: See
>>>>> http://www.mcs.anl.gov/petsc/documentation/faq.html for
>>>>> > trouble shooting.
>>>>> > [0]PETSC ERROR: Petsc Development GIT revision:
>>>>> v3.11.1-277-g63b3610  GIT
>>>>> > Date: 2019-04-18 08:33:56 -0500
>>>>> > [0]PETSC ERROR:
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/BIB/bin/
>>>>> Test.BIBProblemeGD.dev
>>>>> > on a  named compile by Unknown Fri Apr 19 03:22:25 2019
>>>>> > [0]PETSC ERROR: Configure options --prefix=/opt/petsc-master_debug
>>>>> > --with-mpi-compilers=1 --with-mpi-dir=/opt/openmpi-1.10.2
>>>>> > --with-cxx-dialect=C++11 --with-make-np=12 --with-shared-libraries=1
>>>>> > --with-debugging=yes --with-memalign=64 --with-visibility=0
>>>>> > --with-64-bit-indices=0 --download-ml=yes --download-mumps=yes
>>>>> > --download-superlu=yes --download-superlu_dist=yes
>>>>> --download-parmetis=yes
>>>>> > --download-ptscotch=yes --download-metis=yes
>>>>> --download-suitesparse=yes
>>>>> > --download-hypre=yes
>>>>> >
>>>>> --with-blaslapack-dir=/opt/intel/composer_xe_2015.2.164/mkl/lib/intel64
>>>>> > --with-mkl_pardiso-dir=/opt/intel/composer_xe_2015.2.164/mkl
>>>>> > --with-mkl_cpardiso-dir=/opt/intel/composer_xe_2015.2.164/mkl
>>>>> > --with-scalapack=1
>>>>> >
>>>>> --with-scalapack-include=/opt/intel/composer_xe_2015.2.164/mkl/include
>>>>> >
>>>>> --with-scalapack-lib="-L/opt/intel/composer_xe_2015.2.164/mkl/lib/intel64
>>>>> > -lmkl_scalapack_lp64 -lmkl_blacs_openmpi_lp64"
>>>>> > [0]PETSC ERROR: #1 MatSetValues_SeqSBAIJ() line 990 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/impls/sbaij/seq/sbaij.c
>>>>> > [0]PETSC ERROR: #2 MatSetValues() line 1404 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/interface/matrix.c
>>>>> > [0]PETSC ERROR: #3 MatAXPY_BasicWithPreallocation() line 209 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/utils/axpy.c
>>>>> > [0]PETSC ERROR: #4 MatAXPY() line 96 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/utils/axpy.c
>>>>> > [0]PETSC ERROR: #5 MatAYPX() line 384 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/utils/axpy.c
>>>>> > [0]PETSC ERROR: #6 MatCopy_Basic() line 4056 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/interface/matrix.c
>>>>> > [0]PETSC ERROR: #7 MatCopy_SeqAIJ() line 2844 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/impls/aij/seq/aij.c
>>>>> > [0]PETSC ERROR: #8 MatCopy() line 4110 in
>>>>> >
>>>>> /pmi/cmpbib/compilation_BIB_gcc_redhat_petsc-master_debug/COMPILE_AUTO/petsc-master-debug/src/mat/interface/matrix.c
>>>>> >
>>>>> > Here are my installation logs as of april 18, at 02h00:
>>>>> >
>>>>> >
>>>>> http://www.giref.ulaval.ca/~cmpgiref/petsc-master-debug/2019.04.18.02h00m01s_configure.log
>>>>> >
>>>>> http://www.giref.ulaval.ca/~cmpgiref/petsc-master-debug/2019.04.18.02h00m01s_make.log
>>>>> >
>>>>> http://www.giref.ulaval.ca/~cmpgiref/petsc-master-debug/2019.04.18.02h00m01s_default.log
>>>>> >
>>>>> http://www.giref.ulaval.ca/~cmpgiref/petsc-master-debug/2019.04.18.02h00m01s_RDict.log
>>>>> >
>>>>> http://www.giref.ulaval.ca/~cmpgiref/petsc-master-debug/2019.04.18.02h00m01s_CMakeLists.txt
>>>>> >
>>>>> http://www.giref.ulaval.ca/~cmpgiref/petsc-master-debug/2019.04.18.02h00m01s_make_test.log
>>>>> >
>>>>> http://www.giref.ulaval.ca/~cmpgiref/petsc-master-debug/2019.04.18.02h00m01s_make_streams.log
>>>>> >
>>>>> > Thanks,
>>>>> >
>>>>> > Eric
>>>>> >
>>>>> >
>>>>> >
>>>>>
>>>>
>>>>
>>>> --
>>>> What most experimenters take for granted before they begin their
>>>> experiments is infinitely more interesting than any results to which their
>>>> experiments lead.
>>>> -- Norbert Wiener
>>>>
>>>> https://www.cse.buffalo.edu/~knepley/
>>>> <http://www.cse.buffalo.edu/~knepley/>
>>>>
>>>
>>
>> --
>> Stefano
>>
>>

-- 
Stefano
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20190422/0e26c96a/attachment-0001.html>