[CIG-SHORT] PyLith Installer PETSc Issue
Brad Aagaard
baagaard at usgs.gov
Mon Sep 24 12:57:36 PDT 2018
Robert,
Matt merged the PETSc master branch into the PETSc knepley/pylith branch
this morning, so it should have all of the updates and resolve your
issue building PETSc.
Note: I do not think the PyLith poroelasticity branch is up-to-date with
the PyLith multiphysics branch (knepley/feature-petsc-fe), so it
probably won't build with the latest PETSc. Josimar or someone else with
write access to the poroelasticity branch will need to merge
knepley/feature-petsc-fe into the poroelasticity branch --or-- create a
pull request on that branch with this change.
Brad
On 09/18/2018 10:49 PM, Robert Walker wrote:
> Trying again, and hard setting PETSC_DIR, no that is not the issue.
>
> Robert L. Walker
> PhD Candidate Petroleum Engineering
> Mork Family Department of Chemicals and Materials Sciences
> University of Southern California
> ----------------------------------------------
> Mobile US: +1 (213) - 290 -7101
> Mobile EU: +34 62 274 66 40
> rlwalker at usc.edu <mailto:rlwalker at usc.edu>
>
> On Tue, Sep 18, 2018 at 10:43 PM, Robert Walker <rlwalker at usc.edu
> <mailto:rlwalker at usc.edu>> wrote:
>
> I tried on a local machine, and on a virgin cluster install, the
> same....here is the result from the local machine.
>
> ${PETSC_DIR} is not set, period. Might this be the issue?
>
> Robert
>
> Robert L. Walker
> PhD Candidate Petroleum Engineering
> Mork Family Department of Chemicals and Materials Sciences
> University of Southern California
> ----------------------------------------------
> Mobile US: +1 (213) - 290 -7101
> Mobile EU: +34 62 274 66 40
> rlwalker at usc.edu <mailto:rlwalker at usc.edu>
>
> On Tue, Sep 18, 2018 at 4:07 PM, Brad Aagaard <baagaard at usgs.gov
> <mailto:baagaard at usgs.gov>> wrote:
>
> Robert,
>
> Do you have another PETSc installed on your machine or PETSC_DIR
> set in your environment to another location?
>
> Can you verify that the PETSc source has this hash
> c3d7e8a3961e2564b0ee47a7f7034285c2c0f620 with the most recent
> commit on Aug 29?
>
> Brad
>
>
> On 09/18/2018 03:53 PM, Robert Walker wrote:
>
> Brad,
>
> I tried again, the installer indicated that it pulled the
> knepley/pylith branch, still I get the exact same error.
>
> Robert
>
> Robert L. Walker
> PhD Candidate Petroleum Engineering
> Mork Family Department of Chemicals and Materials Sciences
> University of Southern California
> ----------------------------------------------
> Mobile US: +1 (213) - 290 -7101
> Mobile EU: +34 62 274 66 40
> rlwalker at usc.edu <mailto:rlwalker at usc.edu>
> <mailto:rlwalker at usc.edu <mailto:rlwalker at usc.edu>>
>
>
> On Tue, Sep 18, 2018 at 9:40 AM, Brad Aagaard
> <baagaard at usgs.gov <mailto:baagaard at usgs.gov>
> <mailto:baagaard at usgs.gov <mailto:baagaard at usgs.gov>>> wrote:
>
> My original reply didn't go to the cig-short list, so I
> am sending
> it again.
>
> It sounds like your build issue is the branch of PETSc
> being used.
> You need the knepley/pylith branch of PETSc, not the
> master branch.
> Matt sometimes adds new features in the knepley/pylith
> branch before
> they get merged into master. Likewise, he controls when
> master gets
> merged into the knepley/pylith branch so that we know
> when to expect
> PETSc updates to break things in PyLith.
>
> Matt also mentioned that there was a short time period
> last week
> when he broke the knepley/pylith PETSc branch. This is
> fixed now.
>
> Brad
>
>
> On 9/14/18 3:23 AM, Robert Walker wrote:
>
> Hello,
>
> Not sure if this is the right place to ask this, or
> it would be
> better to kick it over to petsc-users.
> I am trying to use the installer to build PyLith
> from source,
> and I am encountering the following error when
> trying to make
> pylith-petsc, using the configure settings that are
> stock for
> the installer script, but listed below anyway:
>
> ./configure --prefix=/home/rcf-40/rlwalker/pylith \
> --with-c2html=0 --with-x=0 \
> --with-clanguage=C \
> --with-mpicompilers=1 \
> --with-shared-libraries=1 --with-64-bit-points=1
> --with-large-file-io=1 \
> --download-chaco=1 --download-ml=1
> --download-f2cblaslapack=1
> --with-hdf5=1
> --with-hdf5-dir=/home/rcf-40/rlwalker/pylith
> --with-zlib=1 --LIBS=-lz --with-debugging=0
> --with-fc=0
> CPPFLAGS="-I/home/rcf-40/rlwalker/pylith/include
> -I/home/rcf-40/rlwalker/pylith/include "
> LDFLAGS="-L/home/rcf-40/rlwalker/pylith/lib
> -L/home/rcf-40/rlwalker/pylith/lib64
> -L/home/rcf-40/rlwalker/pylith/lib
> -L/home/rcf-40/rlwalker/pylith/lib64 " CFLAGS="-fPIC"
> CXXFLAGS="-fPIC -DMPICH_IGNORE_CXX_SEEK" FCFLAGS="" \
>
> PETSC_DIR=/home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith
> PETSC_ARCH=arch-pylith && \
>
>
> /*
>
> ==================================================================
> */
> /* ( End PETSc Config....Next is the error) */
> /*
>
> ==================================================================
> */
>
>
>
> In file included from
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmpleximpl.h:8:0,
> from
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/petscfeimpl.h:7,
> from
>
> /auto/rcf-proj/rlw1/build/pylith/petsc-pylith/src/dm/field/impls/ds/dmfieldds.c:2:
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:
> In function ‘DMGetLocalOffset_Private’:
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:335:15:
> error: dereferencing pointer to incomplete type
> *start = s->atlasOff[point - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:335:35:
> error: dereferencing pointer to incomplete type
> *start = s->atlasOff[point - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:336:24:
> error: dereferencing pointer to incomplete type
> *end = *start + s->atlasDof[point -
> s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:336:44:
> error: dereferencing pointer to incomplete type
> *end = *start + s->atlasDof[point -
> s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:
> In function ‘DMGetLocalFieldOffset_Private’:
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:357:46:
> error: dereferencing pointer to incomplete type
> const PetscSection s =
> dm->defaultSection->field[field];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:358:15:
> error: dereferencing pointer to incomplete type
> *start = s->atlasOff[point - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:358:35:
> error: dereferencing pointer to incomplete type
> *start = s->atlasOff[point - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:359:24:
> error: dereferencing pointer to incomplete type
> *end = *start + s->atlasDof[point -
> s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:359:44:
> error: dereferencing pointer to incomplete type
> *end = *start + s->atlasDof[point -
> s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:
> In function ‘DMGetGlobalOffset_Private’:
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:383:32:
> error: dereferencing pointer to incomplete type
> const PetscInt dof = s->atlasDof[point
> - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:383:52:
> error: dereferencing pointer to incomplete type
> const PetscInt dof = s->atlasDof[point
> - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:384:32:
> error: dereferencing pointer to incomplete type
> const PetscInt cdof = s->bc ?
> s->bc->atlasDof[point -
> s->bc->pStart] : 0;
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:384:40:
> error: dereferencing pointer to incomplete type
> const PetscInt cdof = s->bc ?
> s->bc->atlasDof[point -
> s->bc->pStart] : 0;
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:384:64:
> error: dereferencing pointer to incomplete type
> const PetscInt cdof = s->bc ?
> s->bc->atlasDof[point -
> s->bc->pStart] : 0;
>
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:385:15:
> error: dereferencing pointer to incomplete type
> *start = s->atlasOff[point - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:385:35:
> error: dereferencing pointer to incomplete type
> *start = s->atlasOff[point - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:
> In function ‘DMGetGlobalFieldOffset_Private’:
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:417:50:
> error: dereferencing pointer to incomplete type
> const PetscSection fs =
> dm->defaultSection->field[field];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:419:33:
> error: dereferencing pointer to incomplete type
> const PetscInt loff = s->atlasOff[point
> - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:419:53:
> error: dereferencing pointer to incomplete type
> const PetscInt loff = s->atlasOff[point
> - s->pStart];
>
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:420:34:
> error: dereferencing pointer to incomplete type
> const PetscInt goff =
> gs->atlasOff[point - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:420:54:
> error: dereferencing pointer to incomplete type
> const PetscInt goff =
> gs->atlasOff[point - s->pStart];
>
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:421:34:
> error: dereferencing pointer to incomplete type
> const PetscInt lfoff =
> fs->atlasOff[point - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:421:54:
> error: dereferencing pointer to incomplete type
> const PetscInt lfoff =
> fs->atlasOff[point - s->pStart];
>
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:422:34:
> error: dereferencing pointer to incomplete type
> const PetscInt fdof =
> fs->atlasDof[point - s->pStart];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:422:54:
> error: dereferencing pointer to incomplete type
> const PetscInt fdof =
> fs->atlasDof[point - s->pStart];
>
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:423:34:
> error: dereferencing pointer to incomplete type
> const PetscInt fcdof = fs->bc ?
> fs->bc->atlasDof[point - fs->bc->pStart] : 0;
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:423:43:
> error: dereferencing pointer to incomplete type
> const PetscInt fcdof = fs->bc ?
> fs->bc->atlasDof[point - fs->bc->pStart] : 0;
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:423:68:
> error: dereferencing pointer to incomplete type
> const PetscInt fcdof = fs->bc ?
> fs->bc->atlasDof[point - fs->bc->pStart] : 0;
>
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:427:50:
> error: dereferencing pointer to incomplete type
> const PetscSection ffs =
> dm->defaultSection->field[f];
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:428:20:
> error: dereferencing pointer to incomplete type
> ffcdof += ffs->bc ? ffs->bc->atlasDof[point -
> ffs->bc->pStart] : 0;
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:428:30:
> error: dereferencing pointer to incomplete type
> ffcdof += ffs->bc ? ffs->bc->atlasDof[point -
> ffs->bc->pStart] : 0;
> ^
>
> /home/rcf-40/rlwalker/NTS/build/pylith/petsc-pylith/include/petsc/private/dmimpl.h:428:56:
> error: dereferencing pointer to incomplete type
> ffcdof += ffs->bc ? ffs->bc->atlasDof[point -
> ffs->bc->pStart] : 0;
>
> ^
> make[3]: ***
> [arch-pylith/obj/dm/field/impls/ds/dmfieldds.o] Error 1
> make[3]: Leaving directory
> `/auto/rcf-proj/rlw1/build/pylith/petsc-pylith'
> make[2]: *** [petsc] Error 2
> make[2]: Leaving directory
> `/auto/rcf-proj/rlw1/build/pylith'
> make[1]: *** [installed_petsc] Error 2
> make[1]: Leaving directory
> `/auto/rcf-proj/rlw1/build/pylith'
> make: *** [installed_pylith] Error 2
>
> /*
> ========================================================== */
> /* End Error */
> /*
> ========================================================== */
>
> Looking at ìt my thought is that there is a goof in the
> interaction between the PetscSection type and the
> functions
> outlined in dmimpl.h....it's a private function so
> maybe a PETSc
> bug, but again I don't remember what version PETSc
> that the
> pylith installer script pulls.
>
> Thanks,
>
> Robert
>
>
>
>
> Robert L. Walker
> MS Petroleum Engineering
> Mork Family Department of Chemicals and Materials
> Sciences
> University of Southern California
> ----------------------------------------------
> Mobile US: +1 (213) - 290 -7101
> Mobile EU: +34 62 274 66 40
> rlwalker at usc.edu <mailto:rlwalker at usc.edu>
> <mailto:rlwalker at usc.edu <mailto:rlwalker at usc.edu>>
> <mailto:rlwalker at usc.edu <mailto:rlwalker at usc.edu>
> <mailto:rlwalker at usc.edu <mailto:rlwalker at usc.edu>>>
>
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org <mailto:CIG-SHORT at geodynamics.org>
> <mailto:CIG-SHORT at geodynamics.org
> <mailto:CIG-SHORT at geodynamics.org>>
> https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.geodynamics.org_cgi-2Dbin_mailman_listinfo_cig-2Dshort&d=DwIGaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=BWOpUsAp5dvI5cUiP4uZxQ&m=saijv-PR0-VMJ14p4-dtFFzUC8H3QkdxQwMsw3L0aO8&s=ZUv2APUcfRgnrbEdOXuuviw3GwlnghID20tIjE3Hbjo&e=
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.geodynamics.org_cgi-2Dbin_mailman_listinfo_cig-2Dshort&d=DwIGaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=BWOpUsAp5dvI5cUiP4uZxQ&m=saijv-PR0-VMJ14p4-dtFFzUC8H3QkdxQwMsw3L0aO8&s=ZUv2APUcfRgnrbEdOXuuviw3GwlnghID20tIjE3Hbjo&e=>
>
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.geodynamics.org_cgi-2Dbin_mailman_listinfo_cig-2Dshort&d=DwIGaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=BWOpUsAp5dvI5cUiP4uZxQ&m=saijv-PR0-VMJ14p4-dtFFzUC8H3QkdxQwMsw3L0aO8&s=ZUv2APUcfRgnrbEdOXuuviw3GwlnghID20tIjE3Hbjo&e=
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.geodynamics.org_cgi-2Dbin_mailman_listinfo_cig-2Dshort&d=DwIGaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=BWOpUsAp5dvI5cUiP4uZxQ&m=saijv-PR0-VMJ14p4-dtFFzUC8H3QkdxQwMsw3L0aO8&s=ZUv2APUcfRgnrbEdOXuuviw3GwlnghID20tIjE3Hbjo&e=>>
>
>
>
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org <mailto:CIG-SHORT at geodynamics.org>
> https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.geodynamics.org_cgi-2Dbin_mailman_listinfo_cig-2Dshort&d=DwIGaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=BWOpUsAp5dvI5cUiP4uZxQ&m=SspbmP9pzoEaLMfTiH6nGz0ocPP6YG5U3ecbg2kUee0&s=8nP2dBAnovcNhik-CsYcRI4Wc0nEdAY6vD2zFRbvP-I&e=
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.geodynamics.org_cgi-2Dbin_mailman_listinfo_cig-2Dshort&d=DwIGaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=BWOpUsAp5dvI5cUiP4uZxQ&m=SspbmP9pzoEaLMfTiH6nGz0ocPP6YG5U3ecbg2kUee0&s=8nP2dBAnovcNhik-CsYcRI4Wc0nEdAY6vD2zFRbvP-I&e=>
>
>
>
More information about the CIG-SHORT
mailing list