April 2011 - petsc-dev - mailman3.cels.anl.gov

Re: [petsc-dev] Floating point exception in ex47cu.cu
by Satish Balay 25 Apr '11

25 Apr '11

That would be wierd. Currently configure defaults to sm_10 for single precision, and sm_13 for double precision. [I've verified this again] can you send configure.log [without the cuda-arch option] - to petsc-maint? Satish On Mon, 25 Apr 2011, Евгений Козлов wrote: > Yes, this was the problem. Now it works. > > Thank you. > > 2011/4/25 Victor Minden <victorminden(a)gmail.com>: > > Eugene, > > Based off of > > Configure options --prefix=/home/kukushkinav > > --with-blas-lapack-dir=/opt/intel/composerxe-2011.0.084/mkl > > --with-mpi-dir=/opt/intel/impi/4.0.1.007/intel64/bin --with-cuda=1 > > --with-cusp=1 --with-thrust=1 > > --with-thrust-dir=/home/kukushkinav/include > > --with-cusp-dir=/home/kukushkinav/include > > It looks like maybe you're not setting the configure flag > > --with-cuda-arch=XXX. Without this, PETSc uses the nvcc default, which when > > I last looked was sm_10. The problem with this is that sm_10 doesn't > > support double-precision. Maybe this is your problem? For example, I use > > -with-cuda-arch=sm_13, which corresponds to NVIDIA compute capability 1.3 > > Cheers, > > Victor > > --- > > Victor L. Minden > > > > Tufts University > > School of Engineering > > Class of 2012 > > > > > > On Fri, Apr 22, 2011 at 12:22 PM, Евгений Козлов <neoveneficus(a)gmail.com> > > wrote: > >> > >> Hello, > >> > >> I am interesting in using PETSc for iterative solving sparse linear > >> system on multi-GPU systems. > >> > >> First of all, I compiled PETSc-dev and tried to run some examples, > >> which will run on GPU. > >> > >> I found src/snes/examples/tutorials/ex47cu.cu. It was compiled and > >> executed. > >> > >> Out of original program src/snes/examples/tutorials/ex47cu.cu: > >> > >> [0]PETSC ERROR: --------------------- Error Message > >> ------------------------------------ > >> [0]PETSC ERROR: Floating point exception! > >> [0]PETSC ERROR: User provided compute function generated a Not-a-Number! > >> [0]PETSC ERROR: > >> ------------------------------------------------------------------------ > >> [0]PETSC ERROR: Petsc Development HG revision: > >> d3e10315d68b1dd5481adb2889c7d354880da362 HG Date: Wed Apr 20 21:03:56 > >> 2011 -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: ex47cu on a arch-linu named cn03 by kukushkinav Fri > >> Apr 22 18:33:32 2011 > >> [0]PETSC ERROR: Libraries linked from /home/kukushkinav/lib > >> [0]PETSC ERROR: Configure run at Thu Apr 21 19:18:22 2011 > >> [0]PETSC ERROR: Configure options --prefix=/home/kukushkinav > >> --with-blas-lapack-dir=/opt/intel/composerxe-2011.0.084/mkl > >> --with-mpi-dir=/opt/intel/impi/4.0.1.007/intel64/bin --with-cuda=1 > >> --with-cusp=1 --with-thrust=1 > >> --with-thrust-dir=/home/kukushkinav/include > >> --with-cusp-dir=/home/kukushkinav/include > >> [0]PETSC ERROR: > >> ------------------------------------------------------------------------ > >> [0]PETSC ERROR: SNESSolve_LS() line 167 in src/snes/impls/ls/ls.c > >> [0]PETSC ERROR: SNESSolve() line 2407 in src/snes/interface/snes.c > >> [0]PETSC ERROR: main() line 38 in src/snes/examples/tutorials/ex47cu.cu > >> application called MPI_Abort(MPI_COMM_WORLD, 72) - process 0 > >> > >> Then I tried to find the place in source with the problem, I changed > >> the function in struct ApplyStencil to > >> > >> void operator()(Tuple t) { thrust::get<0>(t) = 1; } > >> > >> Result: > >> [0]PETSC ERROR: --------------------- Error Message > >> ------------------------------------ > >> [0]PETSC ERROR: Floating point exception! > >> [0]PETSC ERROR: Infinite or not-a-number generated in mdot, entry 0! > >> [0]PETSC ERROR: > >> ------------------------------------------------------------------------ > >> [0]PETSC ERROR: Petsc Development HG revision: > >> d3e10315d68b1dd5481adb2889c7d354880da362 HG Date: Wed Apr 20 21:03:56 > >> 2011 -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: ex47cu on a arch-linu named cn11 by kukushkinav Fri > >> Apr 22 18:58:04 2011 > >> [0]PETSC ERROR: Libraries linked from /home/kukushkinav/lib > >> [0]PETSC ERROR: Configure run at Thu Apr 21 19:18:22 2011 > >> [0]PETSC ERROR: Configure options --prefix=/home/kukushkinav > >> --with-blas-lapack-dir=/opt/intel/composerxe-2011.0.084/mkl > >> --with-mpi-dir=/opt/intel/impi/4.0.1.007/intel64/bin --with-cuda=1 > >> --with-cusp=1 --with-thrust=1 > >> --with-thrust-dir=/home/kukushkinav/include > >> --with-cusp-dir=/home/kukushkinav/include > >> [0]PETSC ERROR: > >> ------------------------------------------------------------------------ > >> [0]PETSC ERROR: VecMDot() line 1146 in src/vec/vec/interface/rvector.c > >> [0]PETSC ERROR: KSPGMRESClassicalGramSchmidtOrthogonalization() line > >> 66 in src/ksp/ksp/impls/gmres/borthog2.c > >> [0]PETSC ERROR: GMREScycle() line 161 in src/ksp/ksp/impls/gmres/gmres.c > >> [0]PETSC ERROR: KSPSolve_GMRES() line 244 in > >> src/ksp/ksp/impls/gmres/gmres.c > >> [0]PETSC ERROR: KSPSolve() line 426 in src/ksp/ksp/interface/itfunc.c > >> [0]PETSC ERROR: SNES_KSPSolve() line 3107 in src/snes/interface/snes.c > >> [0]PETSC ERROR: SNESSolve_LS() line 190 in src/snes/impls/ls/ls.c > >> [0]PETSC ERROR: SNESSolve() line 2407 in src/snes/interface/snes.cС > >> уважением, > >> Евгений > >> [0]PETSC ERROR: main() line 38 in src/snes/examples/tutorials/ex47cu.cu > >> > >> RedHat 5.5, Cuda 3.2 > >> > >> Question1: Is it my problem or a bug of the algorithm? > >> > >> Question2: Where can I find simple doc or example, which describe how > >> to solve sparse linear systems on multi-GPU systems using PETSc? > >> > >> > >> -- > >> Best regards, > >> Eugene > > > > > > > >

1 0

Re: [petsc-dev] Floating point exception in ex47cu.cu
by Евгений Козлов 25 Apr '11

25 Apr '11

Yes, this was the problem. Now it works. Thank you. 2011/4/25 Victor Minden <victorminden(a)gmail.com>: > Eugene, > Based off of > Configure options --prefix=/home/kukushkinav > --with-blas-lapack-dir=/opt/intel/composerxe-2011.0.084/mkl > --with-mpi-dir=/opt/intel/impi/4.0.1.007/intel64/bin --with-cuda=1 > --with-cusp=1 --with-thrust=1 > --with-thrust-dir=/home/kukushkinav/include > --with-cusp-dir=/home/kukushkinav/include > It looks like maybe you're not setting the configure flag > --with-cuda-arch=XXX. Without this, PETSc uses the nvcc default, which when > I last looked was sm_10. The problem with this is that sm_10 doesn't > support double-precision. Maybe this is your problem? For example, I use > -with-cuda-arch=sm_13, which corresponds to NVIDIA compute capability 1.3 > Cheers, > Victor > --- > Victor L. Minden > > Tufts University > School of Engineering > Class of 2012 > > > On Fri, Apr 22, 2011 at 12:22 PM, Евгений Козлов <neoveneficus(a)gmail.com> > wrote: >> >> Hello, >> >> I am interesting in using PETSc for iterative solving sparse linear >> system on multi-GPU systems. >> >> First of all, I compiled PETSc-dev and tried to run some examples, >> which will run on GPU. >> >> I found src/snes/examples/tutorials/ex47cu.cu. It was compiled and >> executed. >> >> Out of original program src/snes/examples/tutorials/ex47cu.cu: >> >> [0]PETSC ERROR: --------------------- Error Message >> ------------------------------------ >> [0]PETSC ERROR: Floating point exception! >> [0]PETSC ERROR: User provided compute function generated a Not-a-Number! >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: Petsc Development HG revision: >> d3e10315d68b1dd5481adb2889c7d354880da362 HG Date: Wed Apr 20 21:03:56 >> 2011 -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: ex47cu on a arch-linu named cn03 by kukushkinav Fri >> Apr 22 18:33:32 2011 >> [0]PETSC ERROR: Libraries linked from /home/kukushkinav/lib >> [0]PETSC ERROR: Configure run at Thu Apr 21 19:18:22 2011 >> [0]PETSC ERROR: Configure options --prefix=/home/kukushkinav >> --with-blas-lapack-dir=/opt/intel/composerxe-2011.0.084/mkl >> --with-mpi-dir=/opt/intel/impi/4.0.1.007/intel64/bin --with-cuda=1 >> --with-cusp=1 --with-thrust=1 >> --with-thrust-dir=/home/kukushkinav/include >> --with-cusp-dir=/home/kukushkinav/include >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: SNESSolve_LS() line 167 in src/snes/impls/ls/ls.c >> [0]PETSC ERROR: SNESSolve() line 2407 in src/snes/interface/snes.c >> [0]PETSC ERROR: main() line 38 in src/snes/examples/tutorials/ex47cu.cu >> application called MPI_Abort(MPI_COMM_WORLD, 72) - process 0 >> >> Then I tried to find the place in source with the problem, I changed >> the function in struct ApplyStencil to >> >> void operator()(Tuple t) { thrust::get<0>(t) = 1; } >> >> Result: >> [0]PETSC ERROR: --------------------- Error Message >> ------------------------------------ >> [0]PETSC ERROR: Floating point exception! >> [0]PETSC ERROR: Infinite or not-a-number generated in mdot, entry 0! >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: Petsc Development HG revision: >> d3e10315d68b1dd5481adb2889c7d354880da362 HG Date: Wed Apr 20 21:03:56 >> 2011 -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: ex47cu on a arch-linu named cn11 by kukushkinav Fri >> Apr 22 18:58:04 2011 >> [0]PETSC ERROR: Libraries linked from /home/kukushkinav/lib >> [0]PETSC ERROR: Configure run at Thu Apr 21 19:18:22 2011 >> [0]PETSC ERROR: Configure options --prefix=/home/kukushkinav >> --with-blas-lapack-dir=/opt/intel/composerxe-2011.0.084/mkl >> --with-mpi-dir=/opt/intel/impi/4.0.1.007/intel64/bin --with-cuda=1 >> --with-cusp=1 --with-thrust=1 >> --with-thrust-dir=/home/kukushkinav/include >> --with-cusp-dir=/home/kukushkinav/include >> [0]PETSC ERROR: >> ------------------------------------------------------------------------ >> [0]PETSC ERROR: VecMDot() line 1146 in src/vec/vec/interface/rvector.c >> [0]PETSC ERROR: KSPGMRESClassicalGramSchmidtOrthogonalization() line >> 66 in src/ksp/ksp/impls/gmres/borthog2.c >> [0]PETSC ERROR: GMREScycle() line 161 in src/ksp/ksp/impls/gmres/gmres.c >> [0]PETSC ERROR: KSPSolve_GMRES() line 244 in >> src/ksp/ksp/impls/gmres/gmres.c >> [0]PETSC ERROR: KSPSolve() line 426 in src/ksp/ksp/interface/itfunc.c >> [0]PETSC ERROR: SNES_KSPSolve() line 3107 in src/snes/interface/snes.c >> [0]PETSC ERROR: SNESSolve_LS() line 190 in src/snes/impls/ls/ls.c >> [0]PETSC ERROR: SNESSolve() line 2407 in src/snes/interface/snes.cС >> уважением, >> Евгений >> [0]PETSC ERROR: main() line 38 in src/snes/examples/tutorials/ex47cu.cu >> >> RedHat 5.5, Cuda 3.2 >> >> Question1: Is it my problem or a bug of the algorithm? >> >> Question2: Where can I find simple doc or example, which describe how >> to solve sparse linear systems on multi-GPU systems using PETSc? >> >> >> -- >> Best regards, >> Eugene > > -- С уважением, Евгений

1 0

Re: [petsc-dev] Floating point exception in ex47cu.cu
by Victor Minden 24 Apr '11

24 Apr '11

Eugene, Based off of Configure options --prefix=/home/kukushkinav --with-blas-lapack-dir=/opt/intel/composerxe-2011.0.084/mkl --with-mpi-dir=/opt/intel/impi/4.0.1.007/intel64/bin<http://4.0.1.7/intel64/bin> --with-cuda=1 --with-cusp=1 --with-thrust=1 --with-thrust-dir=/home/kukushkinav/include --with-cusp-dir=/home/kukushkinav/include It looks like maybe you're not setting the configure flag --with-cuda-arch=XXX. Without this, PETSc uses the nvcc default, which when I last looked was sm_10. The problem with this is that sm_10 doesn't support double-precision. Maybe this is your problem? For example, I use -with-cuda-arch=sm_13, which corresponds to NVIDIA compute capability 1.3 Cheers, Victor --- Victor L. Minden Tufts University School of Engineering Class of 2012 On Fri, Apr 22, 2011 at 12:22 PM, Евгений Козлов <neoveneficus(a)gmail.com>wrote: > Hello, > > I am interesting in using PETSc for iterative solving sparse linear > system on multi-GPU systems. > > First of all, I compiled PETSc-dev and tried to run some examples, > which will run on GPU. > > I found src/snes/examples/tutorials/ex47cu.cu. It was compiled and > executed. > > Out of original program src/snes/examples/tutorials/ex47cu.cu: > > [0]PETSC ERROR: --------------------- Error Message > ------------------------------------ > [0]PETSC ERROR: Floating point exception! > [0]PETSC ERROR: User provided compute function generated a Not-a-Number! > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: Petsc Development HG revision: > d3e10315d68b1dd5481adb2889c7d354880da362 HG Date: Wed Apr 20 21:03:56 > 2011 -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: ex47cu on a arch-linu named cn03 by kukushkinav Fri > Apr 22 18:33:32 2011 > [0]PETSC ERROR: Libraries linked from /home/kukushkinav/lib > [0]PETSC ERROR: Configure run at Thu Apr 21 19:18:22 2011 > [0]PETSC ERROR: Configure options --prefix=/home/kukushkinav > --with-blas-lapack-dir=/opt/intel/composerxe-2011.0.084/mkl > --with-mpi-dir=/opt/intel/impi/4.0.1.007/intel64/bin --with-cuda=1 > --with-cusp=1 --with-thrust=1 > --with-thrust-dir=/home/kukushkinav/include > --with-cusp-dir=/home/kukushkinav/include > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: SNESSolve_LS() line 167 in src/snes/impls/ls/ls.c > [0]PETSC ERROR: SNESSolve() line 2407 in src/snes/interface/snes.c > [0]PETSC ERROR: main() line 38 in src/snes/examples/tutorials/ex47cu.cu > application called MPI_Abort(MPI_COMM_WORLD, 72) - process 0 > > Then I tried to find the place in source with the problem, I changed > the function in struct ApplyStencil to > > void operator()(Tuple t) { thrust::get<0>(t) = 1; } > > Result: > [0]PETSC ERROR: --------------------- Error Message > ------------------------------------ > [0]PETSC ERROR: Floating point exception! > [0]PETSC ERROR: Infinite or not-a-number generated in mdot, entry 0! > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: Petsc Development HG revision: > d3e10315d68b1dd5481adb2889c7d354880da362 HG Date: Wed Apr 20 21:03:56 > 2011 -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: ex47cu on a arch-linu named cn11 by kukushkinav Fri > Apr 22 18:58:04 2011 > [0]PETSC ERROR: Libraries linked from /home/kukushkinav/lib > [0]PETSC ERROR: Configure run at Thu Apr 21 19:18:22 2011 > [0]PETSC ERROR: Configure options --prefix=/home/kukushkinav > --with-blas-lapack-dir=/opt/intel/composerxe-2011.0.084/mkl > --with-mpi-dir=/opt/intel/impi/4.0.1.007/intel64/bin --with-cuda=1 > --with-cusp=1 --with-thrust=1 > --with-thrust-dir=/home/kukushkinav/include > --with-cusp-dir=/home/kukushkinav/include > [0]PETSC ERROR: > ------------------------------------------------------------------------ > [0]PETSC ERROR: VecMDot() line 1146 in src/vec/vec/interface/rvector.c > [0]PETSC ERROR: KSPGMRESClassicalGramSchmidtOrthogonalization() line > 66 in src/ksp/ksp/impls/gmres/borthog2.c > [0]PETSC ERROR: GMREScycle() line 161 in src/ksp/ksp/impls/gmres/gmres.c > [0]PETSC ERROR: KSPSolve_GMRES() line 244 in > src/ksp/ksp/impls/gmres/gmres.c > [0]PETSC ERROR: KSPSolve() line 426 in src/ksp/ksp/interface/itfunc.c > [0]PETSC ERROR: SNES_KSPSolve() line 3107 in src/snes/interface/snes.c > [0]PETSC ERROR: SNESSolve_LS() line 190 in src/snes/impls/ls/ls.c > [0]PETSC ERROR: SNESSolve() line 2407 in src/snes/interface/snes.cС > уважением, > Евгений > [0]PETSC ERROR: main() line 38 in src/snes/examples/tutorials/ex47cu.cu > > RedHat 5.5, Cuda 3.2 > > Question1: Is it my problem or a bug of the algorithm? > > Question2: Where can I find simple doc or example, which describe how > to solve sparse linear systems on multi-GPU systems using PETSc? > > > -- > Best regards, > Eugene >

1 0

Re: [petsc-dev] [petsc-maint #71297] Problem installing PETSc
by Jed Brown 24 Apr '11

24 Apr '11

On Sun, Apr 24, 2011 at 05:28, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > It could be that this fix just needs to be backported to petsc 3.1 Done.

1 0

Re: [petsc-dev] [petsc-maint #71297] Problem installing PETSc
by Barry Smith 23 Apr '11

23 Apr '11

On Apr 23, 2011, at 9:28 PM, Jed Brown wrote: > On Sun, Apr 24, 2011 at 03:46, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > >> Looks like we need a patch to deal with the "illegal" placement of >> XOPEN_SOURCE and BSD_SOURCE in the source code so this issue doesn't keep >> coming up. > > > What exactly is the error? In file included from /usr/include/stdio.h:22, from /home/rgill/PETSc/petsc-3.1-p8/include/petscsys.h:147, from options.c:14: /usr/include/sys/feature_tests.h:336:2: error: #error "Compiler or options invalid; UNIX 03 and POSIX.1-2001 applications require the use of c99" > Note that we already changed _XOPEN_SOURCE to > _POSIX_C_SOURCE (in dev) because Open Solaris did not like the former. It could be that this fix just needs to be backported to petsc 3.1 Barry > One > possible fix could be the ugly > > #ifndef _BSD_SOURCE > # define _BSD_SOURCE > #endif > > > I really do not want to put feature test macros in public headers (our usual > model with finding which header declares certain functions) because they > change which symbols are available in headers and introduce order > dependence. That is, a user writing > > #include <stdlib.h> > #include <petsc.h> > > would not bring in atoll(), but including petsc.h first would have it (and > lots of other symbols). I don't think a library should ever change the > language "dialect" that a user writes their code in. > > > I don't really understand why Solaris is having a problem with our defining > feature test macros. As far as I know, we are using them as recommended. > Searching for solaris feature test macros brings up results like the > following man page that explains how to set them. > > http://www.cims.nyu.edu/cgi-systems/man.cgi?section=5&topic=standards > > On Sun, Apr 24, 2011 at 03:46, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > Looks like we need a patch to deal with the "illegal" placement of XOPEN_SOURCE and BSD_SOURCE in the source code so this issue doesn't keep coming up. > > What exactly is the error? Note that we already changed _XOPEN_SOURCE to _POSIX_C_SOURCE (in dev) because Open Solaris did not like the former. One possible fix could be the ugly > > #ifndef _BSD_SOURCE > # define _BSD_SOURCE > #endif > > > I really do not want to put feature test macros in public headers (our usual model with finding which header declares certain functions) because they change which symbols are available in headers and introduce order dependence. That is, a user writing > > #include <stdlib.h> > #include <petsc.h> > > would not bring in atoll(), but including petsc.h first would have it (and lots of other symbols). I don't think a library should ever change the language "dialect" that a user writes their code in. > > > I don't really understand why Solaris is having a problem with our defining feature test macros. As far as I know, we are using them as recommended. Searching for solaris feature test macros brings up results like the following man page that explains how to set them. > > http://www.cims.nyu.edu/cgi-systems/man.cgi?section=5&topic=standards

1 0

Re: [petsc-dev] [petsc-maint #71297] Problem installing PETSc
by Jed Brown 23 Apr '11

23 Apr '11

On Sun, Apr 24, 2011 at 03:46, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > Looks like we need a patch to deal with the "illegal" placement of > XOPEN_SOURCE and BSD_SOURCE in the source code so this issue doesn't keep > coming up. What exactly is the error? Note that we already changed _XOPEN_SOURCE to _POSIX_C_SOURCE (in dev) because Open Solaris did not like the former. One possible fix could be the ugly #ifndef _BSD_SOURCE # define _BSD_SOURCE #endif I really do not want to put feature test macros in public headers (our usual model with finding which header declares certain functions) because they change which symbols are available in headers and introduce order dependence. That is, a user writing #include <stdlib.h> #include <petsc.h> would not bring in atoll(), but including petsc.h first would have it (and lots of other symbols). I don't think a library should ever change the language "dialect" that a user writes their code in. I don't really understand why Solaris is having a problem with our defining feature test macros. As far as I know, we are using them as recommended. Searching for solaris feature test macros brings up results like the following man page that explains how to set them. http://www.cims.nyu.edu/cgi-systems/man.cgi?section=5&topic=standards

1 0

Re: [petsc-dev] [petsc-maint #71297] Problem installing PETSc
by Barry Smith 23 Apr '11

23 Apr '11

Robert, Unfortunately we don't have much access to Solaris for testing. You will need to do two things to get it compiled 1) edit $PETSC_ARCH/include/petscconf.h and remove all the lines that mention ISINF 2) edit src/sys/objects/options.c and remove the two lines #define _XOPEN_SOURCE 600 #define _BSD_SOURCE then rerun make; do not rerun ./configure. If you want to make another PETSC_ARCH you will need to remove the ISNF after running the second configure. Satish and Jed, Looks like we need a patch to deal with the "illegal" placement of XOPEN_SOURCE and BSD_SOURCE in the source code so this issue doesn't keep coming up. Satish, Also backport the patch for isinf test that Matt and I recently added to petsc-dev Thanks Barry On Apr 23, 2011, at 8:02 PM, Robert Gill wrote: > I'm trying to install PETSc for a computer science class, but I was unable to get it to install. Here is the list of commands I performed: > > gunzip -c petsc-3.1-p8.tar.gz | tar -xof - > cd petsc-3.1-p8 > setenv PETSC_DIR $PWD > setenv PETSC_ARCH solaris2.10-c-debug > ./config/configure.py --with-mpi=0 > make all > make test > > Also, please find attached the make.log and test.log files. > > I tried going to the troubleshooting URL listed in the test.log file, but the page was not found. In addition, the Troubleshooting link (which seems to only appear on the Installation page) is broken. > > Please help! Thanks! > > Robert Gill > > I'm trying to install PETSc for a computer science class, but I was unable to get it to install. Here is the list of commands I performed: > > gunzip -c petsc-3.1-p8.tar.gz | tar -xof - > cd petsc-3.1-p8 > setenv PETSC_DIR $PWD > setenv PETSC_ARCH solaris2.10-c-debug > ./config/configure.py --with-mpi=0 > make all > make test > > Also, please find attached the make.log and test.log files. > > I tried going to the troubleshooting URL listed in the test.log file, but the page was not found. In addition, the Troubleshooting link (which seems to only appear on the Installation page) is broken. > > Please help! Thanks! > > Robert Gill > > <make.log><test.log>

1 0

What should MatGetRowIJ(..., symmetric=TRUE, ...) do with empty diagonal?
by Jed Brown 23 Apr '11

23 Apr '11

MatGetOrdering is failing with the following matrix. Should it fail (i.e. diagonal is required) or not? It seems to work for many larger sizes even without setting a diagonal. Matrix Object: 1 MPI processes type: seqaij row 0: (2, 1) row 1: (2, 1) row 2: (0, 1) (1, 1) #0 0x00007ffff5a70ffc in MatGetRowIJ_SeqAIJ_Inode_Symmetric (A=0x1ee9a00, iia=0x7fffffffbe68, jja=0x7fffffffbe60, ishift=0, oshift=1) at /home/jed/petsc/src/mat/impls/aij/seq/inode.c:108 #1 0x00007ffff5a72808 in MatGetRowIJ_SeqAIJ_Inode (A=0x1ee9a00, oshift=1, symmetric=PETSC_TRUE, blockcompressed=PETSC_TRUE, n=0x7fffffffbe74, ia=0x7fffffffbe68, ja=0x7fffffffbe60, done=0x7fffffffbe54) at /home/jed/petsc/src/mat/impls/aij/seq/inode.c:244 #2 0x00007ffff5705629 in MatGetRowIJ (mat=0x1ee9a00, shift=1, symmetric=PETSC_TRUE, inodecompressed=PETSC_TRUE, n=0x7fffffffbe74, ia=0x7fffffffbe68, ja=0x7fffffffbe60, done=0x7fffffffbe54) at /home/jed/petsc/src/mat/interface/matrix.c:6650 #3 0x00007ffff5b2318d in MatGetOrdering_RCM (mat=0x1ee9a00, type=0xa875b0 "rcm", row=0x7fffffffc1d0, col=0x7fffffffc1c8) at /home/jed/petsc/src/mat/order/sprcm.c:18 #4 0x00007ffff5b260fb in MatGetOrdering (mat=0x1ee9a00, type=0xa875b0 "rcm", rperm=0x7fffffffc1d0, cperm=0x7fffffffc1c8) at /home/jed/petsc/src/mat/order/sorder.c:229 #5 0x00007ffff7b8312d in dMeshPopulateOrderedSet_Private (mesh=0xa01920, orderedSet=0xb000000000000018, explicitSet=0xb000000000000019, dirichletSet=0xb00000000000001a, ghostSet=0xb00000000000001b, orderingtype=0xa875b0 "rcm") at /home/jed/dohp/src/fs/impls/cont/cont.c:159 #6 0x00007ffff7b84f60 in dFSBuildSpace_Cont (fs=0xa864e0) at /home/jed/dohp/src/fs/impls/cont/cont.c:288 #7 0x00007ffff7b533f9 in dFSBuildSpace (fs=0xa864e0) at /home/jed/dohp/src/fs/interface/fs.c:351 #8 0x00007ffff7b61d00 in dFSSetFromOptions (fs=0xa864e0) at /home/jed/dohp/src/fs/interface/fsreg.c:127 #9 0x0000000000409572 in EllipSetFromOptions (elp=0x9fdc80) at /home/jed/dohp/src/fs/tests/ellip.c:256 #10 0x000000000040e95f in main (argc=15, argv=0x7fffffffd658) at /home/jed/dohp/src/fs/tests/ellip.c:644

1 0

Re: [petsc-dev] detecting inappropriate mpi.h?
by Jed Brown 23 Apr '11

23 Apr '11

On Sat, Apr 23, 2011 at 18:45, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > Is there any way ./configure can detect an inappropriate mpi.h (like > /usr/include/mpi.h) in the search path and warning the user? Even something > as crude as ls in those locations? I want to use that mpi.h when I configure --with-mpi-dir=/usr. It's tricky to reproduce the compiler's "system paths" so it's not clear how to portably determine which mpi.h the compiler is going to pull in.

1 0

detecting inappropriate mpi.h?
by Barry Smith 23 Apr '11

23 Apr '11

Is there any way ./configure can detect an inappropriate mpi.h (like /usr/include/mpi.h) in the search path and warning the user? Even something as crude as ls in those locations? Thanks Barry

1 0