November 2015 - firedrake - imperial.ac.uk

Re: [firedrake] Preconditioner for Least-squares FEM
by Colin Cotter 08 Dec '15

08 Dec '15

This is a tricky system to precondition in general. Approaches that are known to work well are: 1) Use a block H(div) preconditioner, i.e. precondition by the operator ( 1 - grad(div .) 0 ) ( 0 1 ) This converges fast with direct solvers to obtain the inverse of this operator, but classical iterative methods do not converge well for the top block, and so you need to use Schwartz preconditioners to get a fast parallel implementation. We don't currently have Firedrake support for that. 2) Hybridise the equation. The resulting reduced system behaves well with classical iterative methods or AMG. We have been able to construct the hybridised system using petsc4py. all the best --cjc On 15 October 2015 at 14:18, Andrew McRae <A.T.T.McRae(a)bath.ac.uk> wrote: > As usual, the Firedrake person in the best position to answer your > question is Lawrence, but he's currently on holiday. > > I won't try to answer your question, but in an old email > <http://mailman.ic.ac.uk/pipermail/firedrake/2015-July/000953.html>, > Lawrence suggested you try using multigrid on the blocks; e.g. > > 'fieldsplit_0_pc_type': 'gamg', > 'fieldsplit_1_pc_type': 'gamg', > > My linear solver knowledge is pretty dire, so I'll stop talking here :) > > Andrew > > > On 15 October 2015 at 13:58, Justin Chang <jychang48(a)gmail.com> wrote: > >> I did ask them, but it went unanswered for the last 1.5 months... >> >> Let me clarify my question a bit though, I tried using -ksp_type cg >> -pc_type hypre -pc_hypre_type boomeramg, but it is not supported for >> matnest. I added nest=False but it seems to break my solver somehow. >> >> The only thing that works for me is -ksp_type cg -pc_type bjacobi plus >> MixedVectorSpaceBasis(W, ...) but the number of solver iterations increases >> with problem size... >> >> Thanks, >> Justin >> >> On Thu, Oct 15, 2015 at 6:50 AM, Andrew McRae <A.T.T.McRae(a)bath.ac.uk> >> wrote: >> >>> I think the PETSc mailing list is a much better place to ask questions >>> like this? >>> >>> On 15 October 2015 at 13:47, Justin Chang <jychang48(a)gmail.com> wrote: >>> >>>> Hi all, >>>> >>>> I am attempting to solve Darcy's equation: >>>> >>>> u + grad[p] = 0 >>>> div[u] = f >>>> >>>> The weak form under the least-squares finite element method (LSFEM) >>>> looks like this: >>>> >>>> (u + grad[p]; v + grad[q]) + div[u]*div[v] = (f; div[v]) >>>> >>>> For H(div) elements like RT0, these options worked nicely: >>>> >>>> -ksp_type gmres >>>> -pc_type fieldsplit >>>> -pc_fieldsplit_type schur >>>> -pc_fieldsplit_schur_precondition selfp >>>> -fieldsplit_0_ksp_type preonly >>>> -fieldsplit_0_pc_type bjacobi >>>> -fieldsplit_0_sub_pc_type ilu >>>> -fieldsplir_1_ksp_type preonly >>>> -fieldsplit_1_pc_type hypre >>>> >>>> but for the above LSFEM, it does not work (get a DIVERGED_MAX_IT >>>> error). I heard multigrid methods are good for these types of problems, so >>>> how do I tweak the above parameters? >>>> >>>> Thanks, >>>> Justin >>>> >>> >>> >>> _______________________________________________ >>> firedrake mailing list >>> firedrake(a)imperial.ac.uk >>> https://mailman.ic.ac.uk/mailman/listinfo/firedrake >>> >>> >> > -- http://www.imperial.ac.uk/people/colin.cotter www.cambridge.org/9781107663916

5 11

Re: [firedrake] FD on mac
by David Ham 05 Dec '15

05 Dec '15

On Mon, 30 Nov 2015 at 18:26 Mitchell, Lawrence < lawrence.mitchell(a)imperial.ac.uk> wrote: > > > On 30 Nov 2015, at 18:01, David Ham <David.Ham(a)imperial.ac.uk> wrote: > > > > This is quite possible. Unfortunately we have no access to Mavericks > machines which makes it difficult to work fix this problem. > > > > The script does currently work for me on El Capitan! > > The proximate cause of Onno's problem appears to be the lack of fortran > compiler. In the past, this came via the brew package "gfortran" is this > still the case? Those of us for whom everything works have had it all in > place for a while, no? > gfortran is currently provided by the plain gcc package. ls -l $(which gfortran) lrwxr-xr-x 1 dham admin 32 23 Jul 12:44 /usr/local/bin/gfortran -> ../Cellar/gcc/5.2.0/bin/gfortran > > Lawrence > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake >

5 23

Re: [firedrake] FD on mac
by David Ham 30 Nov '15

30 Nov '15

On Mon, 30 Nov 2015 at 17:34 Anna Kalogirou <a.kalogirou(a)leeds.ac.uk> wrote: > I don't think Onno is using the installation script but the old > step-by-step instructions. I am right, Onno? > > The installation script does not work on my Mac either; I reported it > some time ago but did not get any update from you. It seems that it > might not work on Mavericks OS (?) > This is quite possible. Unfortunately we have no access to Mavericks machines which makes it difficult to work fix this problem. The script does currently work for me on El Capitan! Regards, David > > Best, Anna. > > > On 30/11/15 17:27, Onno Bokhove wrote: > > with me but florin has it working now she forgot to remove the prompt I > think > > I have the same os as Floriane, installed anaconda, reinstalled > > home-brew but that did not work. > > > > ________________________________________ > > From: firedrake-bounces(a)imperial.ac.uk <firedrake-bounces(a)imperial.ac.uk> > on behalf of Lawrence Mitchell <lawrence.mitchell(a)imperial.ac.uk> > > Sent: Monday, November 30, 2015 5:25 PM > > To: firedrake(a)imperial.ac.uk > > Subject: Re: [firedrake] FD on mac > > > >> On 30 Nov 2015, at 17:04, Onno Bokhove <O.Bokhove(a)leeds.ac.uk> wrote: > >> > >> ------------------------------------- > >> No underlying compiler was specified in the wrapper compiler data file > >> (e.g., mpicc-wrapper-data > > Ok, so there is no fortran compiler available. I thought our install > script installed one, but plausibly that failed to work. > > > > Lawrence > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > -- > > Dr Anna Kalogirou > Research Fellow > School of Mathematics > University of Leeds > > http://www1.maths.leeds.ac.uk/~matak/ > > > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake >

3 2

Re: [firedrake] FD on mac
by David Ham 30 Nov '15

30 Nov '15

Well El Capitan at least has the advantage that you're on the same platform as us, which (a) increases the chance of things working, and (b) increases the chance us being able to reproduce any issues you encounter. On Mon, 30 Nov 2015 at 18:10 Onno Bokhove <O.Bokhove(a)leeds.ac.uk> wrote: > Worthwhile switching all 3 to el-capitan then? > > > ------------------------------ > *From:* firedrake-bounces(a)imperial.ac.uk <firedrake-bounces(a)imperial.ac.uk> > on behalf of David Ham <David.Ham(a)imperial.ac.uk> > *Sent:* Monday, November 30, 2015 6:01 PM > *To:* firedrake > > *Subject:* Re: [firedrake] FD on mac > On Mon, 30 Nov 2015 at 17:34 Anna Kalogirou <a.kalogirou(a)leeds.ac.uk> > wrote: > >> I don't think Onno is using the installation script but the old >> step-by-step instructions. I am right, Onno? >> >> The installation script does not work on my Mac either; I reported it >> some time ago but did not get any update from you. It seems that it >> might not work on Mavericks OS (?) >> > > This is quite possible. Unfortunately we have no access to Mavericks > machines which makes it difficult to work fix this problem. > > The script does currently work for me on El Capitan! > > Regards, > > David > > > >> >> Best, Anna. >> >> >> On 30/11/15 17:27, Onno Bokhove wrote: >> > with me but florin has it working now she forgot to remove the prompt I >> think >> > I have the same os as Floriane, installed anaconda, reinstalled >> > home-brew but that did not work. >> > >> > ________________________________________ >> > From: firedrake-bounces(a)imperial.ac.uk < >> firedrake-bounces(a)imperial.ac.uk> on behalf of Lawrence Mitchell < >> lawrence.mitchell(a)imperial.ac.uk> >> > Sent: Monday, November 30, 2015 5:25 PM >> > To: firedrake(a)imperial.ac.uk >> > Subject: Re: [firedrake] FD on mac >> > >> >> On 30 Nov 2015, at 17:04, Onno Bokhove <O.Bokhove(a)leeds.ac.uk> wrote: >> >> >> >> ------------------------------------- >> >> No underlying compiler was specified in the wrapper compiler data file >> >> (e.g., mpicc-wrapper-data >> > Ok, so there is no fortran compiler available. I thought our install >> script installed one, but plausibly that failed to work. >> > >> > Lawrence >> > _______________________________________________ >> > firedrake mailing list >> > firedrake(a)imperial.ac.uk >> > https://mailman.ic.ac.uk/mailman/listinfo/firedrake >> > _______________________________________________ >> > firedrake mailing list >> > firedrake(a)imperial.ac.uk >> > https://mailman.ic.ac.uk/mailman/listinfo/firedrake >> >> -- >> >> Dr Anna Kalogirou >> Research Fellow >> School of Mathematics >> University of Leeds >> >> http://www1.maths.leeds.ac.uk/~matak/ >> >> >> _______________________________________________ >> firedrake mailing list >> firedrake(a)imperial.ac.uk >> https://mailman.ic.ac.uk/mailman/listinfo/firedrake >> >

1 0

Re: [firedrake] FD on mac
by Colin Cotter 30 Nov '15

30 Nov '15

Hi Onno, What goes wrong? all the best --cjc On 30 November 2015 at 17:57, Onno Bokhove <O.Bokhove(a)leeds.ac.uk> wrote: > Yes, the script does not work either. > > ________________________________________ > From: firedrake-bounces(a)imperial.ac.uk <firedrake-bounces(a)imperial.ac.uk> > on behalf of Anna Kalogirou <a.kalogirou(a)leeds.ac.uk> > Sent: Monday, November 30, 2015 5:34 PM > To: firedrake(a)imperial.ac.uk > Subject: Re: [firedrake] FD on mac > > I don't think Onno is using the installation script but the old > step-by-step instructions. I am right, Onno? > > The installation script does not work on my Mac either; I reported it > some time ago but did not get any update from you. It seems that it > might not work on Mavericks OS (?) > > Best, Anna. > > > On 30/11/15 17:27, Onno Bokhove wrote: > > with me but florin has it working now she forgot to remove the prompt I > think > > I have the same os as Floriane, installed anaconda, reinstalled > > home-brew but that did not work. > > > > ________________________________________ > > From: firedrake-bounces(a)imperial.ac.uk <firedrake-bounces(a)imperial.ac.uk> > on behalf of Lawrence Mitchell <lawrence.mitchell(a)imperial.ac.uk> > > Sent: Monday, November 30, 2015 5:25 PM > > To: firedrake(a)imperial.ac.uk > > Subject: Re: [firedrake] FD on mac > > > >> On 30 Nov 2015, at 17:04, Onno Bokhove <O.Bokhove(a)leeds.ac.uk> wrote: > >> > >> ------------------------------------- > >> No underlying compiler was specified in the wrapper compiler data file > >> (e.g., mpicc-wrapper-data > > Ok, so there is no fortran compiler available. I thought our install > script installed one, but plausibly that failed to work. > > > > Lawrence > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > -- > > Dr Anna Kalogirou > Research Fellow > School of Mathematics > University of Leeds > > http://www1.maths.leeds.ac.uk/~matak/ > > > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > -- http://www.imperial.ac.uk/people/colin.cotter www.cambridge.org/9781107663916

1 0

Re: [firedrake] FD on mac
by Colin Cotter 30 Nov '15

30 Nov '15

Can you try Lawrence's suggestion of trying to compile a simple Fortran code using mpif90: $ cat > hello.f90 << EOF program main use mph implicit none write(*, *)'Hello world' end program EOF $ mpif90 hello.f90 -o hello $ ./hello (The $s indicate a shell prompt). all the best --cjc On 30 November 2015 at 16:05, Onno Bokhove <O.Bokhove(a)leeds.ac.uk> wrote: > Tried: > > > 1) Updating .profile as suggested by Justin > > 2) Reinstalled home-brew after uninstalling it. > > 3) But the hello.world fortran code still crashes. > > 4) Hence, same error message holds. > > > ******************************************************************************* > > UNABLE to CONFIGURE with GIVEN OPTIONS (see configure.log > for details): > > > ------------------------------------------------------------------------------- > > Fortran compiler you provided with --with-fc=/usr/local/bin/mpif90 > does not work. > > Cannot compile FC with /usr/local/bin/mpif90. > > > ******************************************************************************* > > > > ... > > > > ---------------------------------------- > > Command "//anaconda/bin/python -c "import setuptools, > tokenize;__file__='/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-TL5bgv-build/setup.py';exec(compile(getattr(tokenize, > 'open', open)(__file__).read().replace('\r\n', '\n'), __file__, 'exec'))" > install --record > /var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-YcsuFy-record/install-record.txt > --single-version-externally-managed --compile" failed with error code 1 in > /var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-TL5bgv-build > > Any suggestions? > > ------------------------------ > *From:* firedrake-bounces(a)imperial.ac.uk <firedrake-bounces(a)imperial.ac.uk> > on behalf of Justin Chang <jychang48(a)gmail.com> > *Sent:* Monday, November 30, 2015 3:53 PM > *To:* Firedrake Project > > *Subject:* Re: [firedrake] FD on mac > > Miklos, > > Yes my error has been resolved by recompiling spatialindex with the same > homebrew compilers as was used for firedrake > > Thanks > Justin > > On Mon, Nov 30, 2015 at 3:29 AM, Miklós Homolya < > m.homolya14(a)imperial.ac.uk> wrote: > >> Please let me know if that fixes your problem. >> >> >> On 28/11/15 20:08, Justin Chang wrote: >> >> Ah right, didn't think about that. Got it, thanks! >> >> On Sat, Nov 28, 2015 at 3:35 AM, Miklós Homolya < >> m.homolya14(a)imperial.ac.uk> wrote: >> >>> With respect to the libspatialindex problem: >>> >>> Yes, see https://github.com/firedrakeproject/firedrake/issues/640 >>> >>> I suggest you to try rebuilding libspatialindex from source (you can do >>> that with homebrew) using the same compiler as you used to build firedrake. >>> >>> >>> On 28/11/15 07:11, Justin Chang wrote: >>> >>> Hi all, >>> >>> I also had the mpif90 error on my El Capitan Macbook, but worked around >>> it by rebuilding the openmpi with the latest brew gcc compilers (GCC >>> 5.2.0). It seems the default configuration provided by Homebrew is built >>> with Apple's compilers (and thus has no wrapper for the mpif90). I put >>> these in my ~/.profile: >>> >>> export HOMEBREW_CC=gcc-5 >>> export HOMEBREW_CXX=g++-5 >>> export FC=gfortran-5 >>> >>> and rebuilt openmpi via: >>> >>> >>> brew reinstall openmpi --build-from-source >>> >>> >>> Firedrake installs successfully, but now I am running into a >>> spatialindex related problem whenever I try to import the firedrake module: >>> >>> >>> python -c 'from firedrake import *' >>> >>> Traceback (most recent call last): >>> >>> File "<string>", line 1, in <module> >>> >>> File >>> "/Users/justin/Software/firedrake/src/firedrake/firedrake/__init__.py", >>> line 31, in <module> >>> >>> from firedrake.assemble import * >>> >>> File >>> "/Users/justin/Software/firedrake/src/firedrake/firedrake/assemble.py", >>> line 10, in <module> >>> >>> from firedrake import assemble_expressions >>> >>> File >>> "/Users/justin/Software/firedrake/src/firedrake/firedrake/assemble_expressions.py", >>> line 17, in <module> >>> >>> from firedrake import function >>> >>> File >>> "/Users/justin/Software/firedrake/src/firedrake/firedrake/function.py", >>> line 14, in <module> >>> >>> from firedrake import functionspace >>> >>> File >>> "/Users/justin/Software/firedrake/src/firedrake/firedrake/functionspace.py", >>> line 17, in <module> >>> >>> import firedrake.mesh as mesh_t >>> >>> File >>> "/Users/justin/Software/firedrake/src/firedrake/firedrake/mesh.py", line >>> 16, in <module> >>> >>> import firedrake.spatialindex as spatialindex >>> >>> ImportError: >>> dlopen(/Users/justin/Software/firedrake/src/firedrake/firedrake/spatialindex.so, >>> 2): Symbol not found: >>> __ZN5Tools11PropertySet11setPropertyENSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKNS_7VariantE >>> >>> Referenced from: >>> /Users/justin/Software/firedrake/src/firedrake/firedrake/spatialindex.so >>> >>> Expected in: flat namespace >>> >>> in >>> /Users/justin/Software/firedrake/src/firedrake/firedrake/spatialindex.so >>> >>> Looks like something went awry with spatialindex. Anyone else having >>> this problem? >>> >>> Thanks, >>> Justin >>> >>> On Fri, Nov 27, 2015 at 11:26 AM, Lawrence Mitchell < >>> <lawrence.mitchell(a)imperial.ac.uk>lawrence.mitchell(a)imperial.ac.uk> >>> wrote: >>> >>>> >>>> > On 27 Nov 2015, at 15:44, Onno Bokhove <O.Bokhove(a)leeds.ac.uk> wrote: >>>> > >>>> > That is what I got yesterday, but Floriane Gidel went past it on her >>>> mac laptop. >>>> > Computer IT staff and I could not figure out what it was. >>>> > We tried anaconda (worked with Floriane), but not with me. >>>> > He thought my mpif90 was probably slightly older: so reloading Xcode >>>> an option? >>>> > Updating the OS? >>>> >>>> Hopefully not! >>>> >>>> > From: firedrake-bounces(a)imperial.ac.uk < >>>> firedrake-bounces(a)imperial.ac.uk> on behalf of Wingate, Beth < >>>> <B.Wingate(a)exeter.ac.uk>B.Wingate(a)exeter.ac.uk> >>>> > Sent: Friday, November 27, 2015 3:41 PM >>>> > To: firedrake(a)imperial.ac.uk >>>> > Subject: Re: [firedrake] FD on mac >>>> > >>>> > I’m getting the same thing — firedrake now doesn’t install due to >>>> “mpif90” not working: >>>> > >>>> > TESTING: checkFortranCompiler from >>>> config.setCompilers(/private/var/folders/zn/46_9pn0x2k32_z59z6fw5j2w0000gn/T/pip-AO07XK-build/config/BuildSystem/config/setCompilers.py:919) >>>> > >>>> ******************************************************************************* >>>> > UNABLE to CONFIGURE with GIVEN OPTIONS (see >>>> configure.log for details): >>>> > >>>> ------------------------------------------------------------------------------- >>>> > Fortran compiler you provided with >>>> --with-fc=/usr/local/bin/mpif90 does not work. >>>> > Cannot compile FC with /usr/local/bin/mpif90. >>>> > >>>> ******************************************************************************* >>>> >>>> >>>> If you try to compile the following simple Fortran hello world code >>>> with mpif90, what happens? >>>> >>>> $ cat > hello.f90 << EOF >>>> program main >>>> use mph >>>> implicit none >>>> write(*, *)'Hello world' >>>> end program >>>> EOF >>>> $ mpif90 hello.f90 -o hello >>>> $ ./hello >>>> >>>> Cheers, >>>> >>>> Lawrence >>>> >>>> >>>> _______________________________________________ >>>> firedrake mailing list >>>> firedrake(a)imperial.ac.uk >>>> https://mailman.ic.ac.uk/mailman/listinfo/firedrake >>>> >>>> >>> >>> >>> _______________________________________________ >>> firedrake mailing listfiredrake@imperial.ac.ukhttps://mailman.ic.ac.uk/mailman/listinfo/firedrake >>> >>> >>> >>> _______________________________________________ >>> firedrake mailing list >>> firedrake(a)imperial.ac.uk >>> https://mailman.ic.ac.uk/mailman/listinfo/firedrake >>> >>> >> >> >> _______________________________________________ >> firedrake mailing listfiredrake@imperial.ac.ukhttps://mailman.ic.ac.uk/mailman/listinfo/firedrake >> >> >> >> _______________________________________________ >> firedrake mailing list >> firedrake(a)imperial.ac.uk >> https://mailman.ic.ac.uk/mailman/listinfo/firedrake >> >> > -- http://www.imperial.ac.uk/people/colin.cotter www.cambridge.org/9781107663916

4 7

Re: [firedrake] FD on mac
by Colin Cotter 30 Nov '15

30 Nov '15

The dollar is just a symbol for your bash prompt. On 30 November 2015 at 17:04, Onno Bokhove <O.Bokhove(a)leeds.ac.uk> wrote: > > On Onno's macbook: > cat > hello.f90 << EOF > > program main > > use mpi > > implicit none > > write(*, *)'Hello world' > > end program > > EOF > math-mc1097:werk amtob$ mpif90 hello.f90 -o hello > -------------------------------------------------------------------------- > No underlying compiler was specified in the wrapper compiler data file > (e.g., mpicc-wrapper-data.txt) > -------------------------------------------------------------------------- > math-mc1097:werk amtob$ > > On Floriane's: > With mpi I don't have an error at 'EOF', but the next command ('$ mpif90 > hello.f90 -o hello') does not work. Error: " -bash: $: command not found" > > ________________________________________ > From: firedrake-bounces(a)imperial.ac.uk <firedrake-bounces(a)imperial.ac.uk> > on behalf of Lawrence Mitchell <lawrence.mitchell(a)imperial.ac.uk> > Sent: Monday, November 30, 2015 4:11 PM > To: firedrake(a)imperial.ac.uk > Subject: Re: [firedrake] FD on mac > > On 30/11/15 16:09, Colin Cotter wrote: > > use mph > > use mpi > > > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > -- http://www.imperial.ac.uk/people/colin.cotter www.cambridge.org/9781107663916

1 0

Re: [firedrake] FD on mac
by David Ham 30 Nov '15

30 Nov '15

OK. I'm doing a scratch rebuild to see if I can reproduce this. David On Tue, 24 Nov 2015 at 21:33 Justin Chang <jychang48(a)gmail.com> wrote: > Hi all, > > Just wanted to say that I am also now having these exact same issues as > others after updating Firedrake on my mac. > > Thanks, > Justin > > On Tue, Nov 24, 2015 at 6:11 AM, Lawrence Mitchell < > lawrence.mitchell(a)imperial.ac.uk> wrote: > >> >> On 23/11/15 16:52, Wingate, Beth wrote: >> > module = imp.load_module(fullname, fo, fn, stuff) >> > ImportError: >> > >> dlopen(/Users/wingate/firedrake/lib/python2.7/site-packages/petsc4py/lib/PETSc.so, >> > 2): Library not loaded: >> > >> /private/var/folders/rw/1wgmdvd50sv7bb1dksy4y_l00000gq/T/pip-DOH6bI-build/arch-python-macosx-10.11-x86_64/lib/libpetsc.3.06.dylib >> > Referenced from: >> > >> /Users/wingate/firedrake/lib/python2.7/site-packages/petsc4py/lib/PETSc.so >> > Reason: image not found >> >> >> Hmph, petsc4py has linked against the temporary location of libpetsc >> (which appeared via pip presumably), rather than its final location. >> >> It is not immediately obvious to me why this would be the case. I do >> not think it is likely anything wrong with your build setup though. >> >> Lawrence >> >> >> _______________________________________________ >> firedrake mailing list >> firedrake(a)imperial.ac.uk >> https://mailman.ic.ac.uk/mailman/listinfo/firedrake >> >> >

7 15

Re: [firedrake] FD on mac
by Colin Cotter 30 Nov '15

30 Nov '15

Hi Onno, I copied in the mailing list. There was a big merge last week which Lawrence says is mostly fixed now. You might have tried to download at an unfortunate moment. Please can you try again and report errors into this mailing list? all the best --cjc On 23 November 2015 at 11:52, Onno Bokhove <O.Bokhove(a)leeds.ac.uk> wrote: > Hi David and Colin, > > > Sorry to use the email, but what is the status of firedrake downloads for > the mac, please? > > All of us mac users are crashing while downloading the new version of > firedrake. > > > We are meeting Thu in Leeds from 12-1 to tackle this issue with the > > laptops around one table in our meeting room. Is there any chance someone > of the firedrake team > > is available on Skype or team view, or email, to help us along? > > > Or any other tips regarding the mac firedrake update? > > > Best wishes, > > > Onno > > > > -- http://www.imperial.ac.uk/people/colin.cotter www.cambridge.org/9781107663916

6 7

Re: [firedrake] FD on mac
by David Ham 26 Nov '15

26 Nov '15

On Thu, 26 Nov 2015 at 13:12 Floriane Gidel [RPG] <mmfg(a)leeds.ac.uk> wrote: > > Dear all, > > > Onno and I still have problem when trying to update/install Firedrake on > our mac, at different stages. Here are the errors we get: > > > I first tried to update FD, but the command to update petsc4py (pip > install -U --no-deps git+ > https://bitbucket.org/mapdes/petsc4py.git@firedrake#egg=petsc4py) was > taking all the application memory of the computer, and the computer just > crashed in the end. > > > So I tried to re-install Firedrake from scratch, but I still have an issue > when trying to install petsc4py: > > > - when installing Petsc, it installs version 3.7.0; I get the following > message : > > > Installing collected packages: petsc > > Found existing installation: petsc 3.6.2 > > Uninstalling petsc-3.6.2: > > Successfully uninstalled petsc-3.6.2 > > Running setup.py install for petsc > > Successfully installed petsc-3.7.0.dev0 > > - when I then try to install petsc4py, I've got an error saying that the > version of petsc needed must be lower than 3.7.0 : > > Collecting petsc<3.7,>=3.6 (from petsc4py) > > Could not find a version that satisfies the requirement petsc<3.7,>=3.6 > (from petsc4py) (from versions: ) > > Some externally hosted files were ignored as access to them may be > unreliable (use --allow-external petsc to allow). > > No matching distribution found for petsc<3.7,>=3.6 (from petsc4py) > > This error is usually a symptom of failing to pass --no-deps to the petsc4py build (it's a bug in upstream petsc4py really but --no-deps is usually an adequate work around). Regards, David > I tried to allow external petsc by adding the option '--allow-external > petsc', but I then have a fatal error : > > fatal error: too many errors emitted, stopping now [-ferror-limit=] > > 1 warning and 20 errors generated. > > error: command '/usr/local/bin/mpicc' failed with exit status 1 > > and this comment: > > Command "/Users/mmfg/anaconda/bin/python -c "import setuptools, > tokenize;__file__='/private/tmp/pip-build-8fgPi3/petsc4py/setup.py';exec(compile(getattr(tokenize, > 'open', open)(__file__).read().replace('\r\n', '\n'), __file__, 'exec'))" > install --record /tmp/pip-u2WItt-record/install-record.txt > --single-version-externally-managed --compile" failed with error code 1 in > /private/tmp/pip-build-8fgPi3/petsc4py > > I have also tried to install FD using your script, but I get error as well > : > > Creating firedrake virtualenv. > > Traceback (most recent call last): > > File "firedrake-install", line 527, in <module> > > virtualenv.create_environment("firedrake", site_packages=True) > > File "/Users/mmfg/.local/lib/python2.7/site-packages/virtualenv.py", > line 1004, in create_environment > > install_wheel(to_install, py_executable, search_dirs) > > File "/Users/mmfg/.local/lib/python2.7/site-packages/virtualenv.py", > line 969, in install_wheel > > 'PIP_NO_INDEX': '1' > > File "/Users/mmfg/.local/lib/python2.7/site-packages/virtualenv.py", > line 910, in call_subprocess > > % (cmd_desc, proc.returncode)) > > OSError: Command /Users/mmfg/anaconda/bin/firedrake/bin/python -c "import > sys, pip; sys...d\"] + sys.argv[1:]))" setuptools pip wheel failed with > error code 1 > > Moreover, Onno gets stalled as well at his macbook pro at: > > > campus-10-41-94-68:werk amtob$ PETSC_CONFIGURE_OPTIONS="--download-ctetgen > --download-triangle --download-chaco --download-hdf5" \ > > > pip install https://bitbucket.org/mapdes/petsc/get/firedrake.tar.bz2 > > Collecting https://bitbucket.org/mapdes/petsc/get/firedrake.tar.bz2 > > Downloading https://bitbucket.org/mapdes/petsc/get/firedrake.tar.bz2 > (12.1MB) > > 100% |████████████████████████████████| 12.1MB 40kB/s > > Installing collected packages: petsc > > Running setup.py install for petsc > > Complete output from command /usr/local/opt/python/bin/python2.7 -c > "import setuptools, > tokenize;__file__='/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/setup.py';exec(compile(getattr(tokenize, > 'open', open)(__file__).read().replace('\r\n', '\n'), __file__, 'exec'))" > install --record > /var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-I6eRXQ-record/install-record.txt > --single-version-externally-managed --compile: > > running install > > PETSc: configure > > configure options: > > --prefix=/usr/local/lib/python2.7/site-packages/petsc > > PETSC_ARCH=arch-python-macosx-10.9-x86_64 > > --with-shared-libraries=1 > > --with-debugging=0 > > --with-c2html=0 > > --with-cc=/usr/local/bin/mpicc > > --with-cxx=/usr/local/bin/mpicxx > > --with-fc=/usr/local/bin/mpif90 > > --download-ctetgen > > --download-triangle > > --download-chaco > > --download-hdf5 > > > =============================================================================== > > Configuring PETSc to compile on your system > > > =============================================================================== > > TESTING: configureExternalPackagesDir from > config.framework(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/framework.py:833) > > TESTING: configureDebuggers from > config.utilities.debuggers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/utilities/debuggers.py:22) > > TESTING: configureGit from > config.sourceControl(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/sourceControl.py:24) > > TESTING: configureCLanguage from > PETSc.options.languages(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/languages.py:27) > > TESTING: configureDirectories from > PETSc.options.petscdir(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/petscdir.py:23) > > TESTING: configureInstallationMethod from > PETSc.options.petscclone(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/petscclone.py:20) > > TESTING: getDatafilespath from > PETSc.options.dataFilesPath(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/dataFilesPath.py:29) > > TESTING: configureArchitecture from > PETSc.options.arch(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/arch.py:25) > > TESTING: setInstallDir from > PETSc.options.installDir(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/installDir.py:35) > > TESTING: saveReconfigure from > PETSc.options.installDir(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/installDir.py:74) > > TESTING: cleanInstallDir from > PETSc.options.installDir(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/installDir.py:67) > > TESTING: configureInstallDir from > PETSc.options.installDir(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/installDir.py:51) > > TESTING: restoreReconfigure from > PETSc.options.installDir(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/installDir.py:87) > > TESTING: setExternalPackagesDir from > PETSc.options.externalpackagesdir(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/externalpackagesdir.py:15) > > TESTING: cleanExternalpackagesDir from > PETSc.options.externalpackagesdir(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/PETSc/options/externalpackagesdir.py:22) > > TESTING: printEnvVariables from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:1572) > > TESTING: resetEnvCompilers from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:1579) > > TESTING: checkEnvCompilers from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:1609) > > TESTING: checkMPICompilerOverride from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:1544) > > TESTING: requireMpiLdPath from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:1563) > > TESTING: checkVendor from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:402) > > TESTING: checkInitialFlags from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:412) > > TESTING: checkCCompiler from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:538) > > TESTING: checkCPreprocessor from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:571) > > TESTING: checkCUDACompiler from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:612) > > TESTING: checkCUDAPreprocessor from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:652) > > TESTING: checkCxxCompiler from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:764) > > TESTING: checkCxxPreprocessor from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:802) > > TESTING: checkFortranCompiler from > config.setCompilers(/private/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/config/BuildSystem/config/setCompilers.py:919) > > > ******************************************************************************* > > UNABLE to EXECUTE BINARIES for ./configure > > > ------------------------------------------------------------------------------- > > Cannot run executables created with FC. If this machine uses a batch > system > > to submit jobs you will need to configure using ./configure with the > additional option --with-batch. > > Otherwise there is problem with the compilers. Can you compile and > run code with your compiler '/usr/local/bin/mpif90'? > > > ******************************************************************************* > > > > Traceback (most recent call last): > > File "<string>", line 1, in <module> > > File > "/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/setup.py", > line 302, in <module> > > **metadata) > > File > "/usr/local/Cellar/python/2.7.10_2/Frameworks/Python.framework/Versions/2.7/lib/python2.7/distutils/core.py", > line 151, in setup > > dist.run_commands() > > File > "/usr/local/Cellar/python/2.7.10_2/Frameworks/Python.framework/Versions/2.7/lib/python2.7/distutils/dist.py", > line 953, in run_commands > > self.run_command(cmd) > > File > "/usr/local/Cellar/python/2.7.10_2/Frameworks/Python.framework/Versions/2.7/lib/python2.7/distutils/dist.py", > line 972, in run_command > > cmd_obj.run() > > File > "/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/setup.py", > line 218, in run > > config(prefix, self.dry_run) > > File > "/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/setup.py", > line 148, in config > > if status != 0: raise RuntimeError(status) > > RuntimeError: 256 > > > > ---------------------------------------- > > Command "/usr/local/opt/python/bin/python2.7 -c "import setuptools, > tokenize;__file__='/var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build/setup.py';exec(compile(getattr(tokenize, > 'open', open)(__file__).read().replace('\r\n', '\n'), __file__, 'exec'))" > install --record > /var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-I6eRXQ-record/install-record.txt > --single-version-externally-managed --compile" failed with error code 1 in > /var/folders/c0/rzc18gxj0619yv1l5zqlbrssh33xh5/T/pip-5zNG1S-build > > > Best regards, > Floriane > > ------------------------------ > *De :* firedrake-bounces(a)imperial.ac.uk <firedrake-bounces(a)imperial.ac.uk> > de la part de David Ham <David.Ham(a)imperial.ac.uk> > *Envoyé :* mercredi 25 novembre 2015 12:03 > *À :* firedrake > *Objet :* Re: [firedrake] FD on mac > > Hi all, > > It turns out that this is a problem in PETSc caused by a "fix" to a > different OSX problem. We have reverted the offending commit on the > firedrake/petsc fork so hopefully updating now should fix the issue. Please > scream immediately if it does not. > > Regards, > > David > > On Tue, 24 Nov 2015 at 22:23 David Ham <David.Ham(a)imperial.ac.uk> wrote: > >> Reproduced. Very odd, it seems petsc4py is currently borked. I don't >> understand where this library location is coming from. It certainly isn't >> where the petsc python module reports the python being installed. >> >> On Tue, 24 Nov 2015 at 22:06 David Ham <David.Ham(a)imperial.ac.uk> wrote: >> >>> OK. I'm doing a scratch rebuild to see if I can reproduce this. >>> >>> David >>> >>> On Tue, 24 Nov 2015 at 21:33 Justin Chang <jychang48(a)gmail.com> wrote: >>> >>>> Hi all, >>>> >>>> Just wanted to say that I am also now having these exact same issues as >>>> others after updating Firedrake on my mac. >>>> >>>> Thanks, >>>> Justin >>>> >>>> On Tue, Nov 24, 2015 at 6:11 AM, Lawrence Mitchell < >>>> lawrence.mitchell(a)imperial.ac.uk> wrote: >>>> >>>>> >>>>> On 23/11/15 16:52, Wingate, Beth wrote: >>>>> > module = imp.load_module(fullname, fo, fn, stuff) >>>>> > ImportError: >>>>> > >>>>> dlopen(/Users/wingate/firedrake/lib/python2.7/site-packages/petsc4py/lib/PETSc.so, >>>>> > 2): Library not loaded: >>>>> > >>>>> /private/var/folders/rw/1wgmdvd50sv7bb1dksy4y_l00000gq/T/pip-DOH6bI-build/arch-python-macosx-10.11-x86_64/lib/libpetsc.3.06.dylib >>>>> > Referenced from: >>>>> > >>>>> /Users/wingate/firedrake/lib/python2.7/site-packages/petsc4py/lib/PETSc.so >>>>> > Reason: image not found >>>>> >>>>> >>>>> Hmph, petsc4py has linked against the temporary location of libpetsc >>>>> (which appeared via pip presumably), rather than its final location. >>>>> >>>>> It is not immediately obvious to me why this would be the case. I do >>>>> not think it is likely anything wrong with your build setup though. >>>>> >>>>> Lawrence >>>>> >>>>> >>>>> _______________________________________________ >>>>> firedrake mailing list >>>>> firedrake(a)imperial.ac.uk >>>>> https://mailman.ic.ac.uk/mailman/listinfo/firedrake >>>>> >>>>> >>>>

2 1