March 2018 - petsc-dev - mailman3.cels.anl.gov

Re: [petsc-dev] [PETSc-3.8.4] src/mat/impls/hypre/mhypre.c compilation failed
by Antonio 31 Mar '18

31 Mar '18

On 30/03/2018 17:07, Satish Balay wrote: > current petsc code is tested with hypre-v2.12.0. > > So this must be due to changes in hypre-v2.14.0. > > Ok looks like we need the following change - but this update is likely > to go into 3.9 [and not 3.8 - which we hope to release next week] > > Satish Your patch fixes the error effectively. Thanks. > > ----------- > commit e6de09342ce9c4562cc062ff2c1bac4bd956bda0 (HEAD -> balay/update-hypre-v2.14.0) > Author: Satish Balay <balay(a)mcs.anl.gov> > Date: Fri Mar 30 09:53:12 2018 -0500 > > hypre: update to version 2.14.0 > > Reported-by: Antonio Trande <anto.trande(a)gmail.com> > > diff --git a/config/BuildSystem/config/packages/hypre.py b/config/BuildSystem/config/packages/hypre.py > index 1249b39374..620e51602a 100644 > --- a/config/BuildSystem/config/packages/hypre.py > +++ b/config/BuildSystem/config/packages/hypre.py > @@ -4,7 +4,7 @@ import os > class Configure(config.package.GNUPackage): > def __init__(self, framework): > config.package.GNUPackage.__init__(self, framework) > - self.gitcommit = 'v2.12.0' > + self.gitcommit = 'v2.14.0' > self.download = ['git://https://github.com/LLNL/hypre','https://github.com/LLNL/hypre/archive/'+self.gitcommit+'.tar.gz'] > self.functions = ['HYPRE_IJMatrixCreate'] > self.includes = ['HYPRE.h'] > diff --git a/src/mat/impls/hypre/mhypre.c b/src/mat/impls/hypre/mhypre.c > index 8a51581a09..65595f2af4 100644 > --- a/src/mat/impls/hypre/mhypre.c > +++ b/src/mat/impls/hypre/mhypre.c > @@ -1450,6 +1450,6 @@ PETSC_EXTERN PetscErrorCode MatCreate_HYPRE(Mat B) > static PetscErrorCode hypre_array_destroy(void *ptr) > { > PetscFunctionBegin; > - hypre_TFree(ptr); > + hypre_TFree(ptr,HYPRE_MEMORY_HOST); > PetscFunctionReturn(0); > } > > > On Fri, 30 Mar 2018, Antonio Trande wrote: > >> Hello everyone. >> >> 'src/mat/impls/hypre/mhypre.c' file compilation is failed (full log [1]) >> with this error: >> >> mpicc -c -O2 -g -pipe -Wall -Werror=format-security >> -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions >> -fstack-protector-strong -grecord-gcc-switches >> -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic >> -fasynchronous-unwind-tables -fstack-clash-protection -mcet >> -fcf-protection -O3 -Wl,-z,now -fPIC >> -I/builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/include >> -I/builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/x86_64/include >> -I/usr/include/openmpi-x86_64/hypre -I/usr/include/openmpi-x86_64 >> -MMD -MP >> /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c >> -o x86_64/obj/src/mat/impls/hypre/mhypre.o >> /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c: >> In function 'hypre_array_destroy': >> /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c:1453:19: >> error: macro "hypre_TFree" requires 2 arguments, but only 1 given >> hypre_TFree(ptr); >> ^ >> /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c:1453:4: >> error: 'hypre_TFree' undeclared (first use in this function); did you >> mean 'hypre_Free'? >> hypre_TFree(ptr); >> ^~~~~~~~~~~ >> hypre_Free >> /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c:1453:4: >> note: each undeclared identifier is reported only once for each function >> it appears in >> gmake[2]: *** [gmakefile:227: x86_64/obj/src/mat/impls/hypre/mhypre.o] >> Error 1 >> >> Please, take a look. >> >> [1] https://kojipkgs.fedoraproject.org//work/tasks/6330/26056330/build.log >> >> System: Fedora 29 (devel branch) >> GCC 8.0.1 >> OpenMPI 2.1.1 >> Hypre 2.14.0 >> >> > -- --- Antonio Trande Fedora Project mailto 'sagitter at fedoraproject dot org' GPG key: 0x5E212EE1D35568BE GPG key server: https://keys.fedoraproject.org/

1 0

Re: [petsc-dev] [PETSc-3.8.4] src/mat/impls/hypre/mhypre.c compilation failed
by Satish Balay 30 Mar '18

30 Mar '18

current petsc code is tested with hypre-v2.12.0. So this must be due to changes in hypre-v2.14.0. Ok looks like we need the following change - but this update is likely to go into 3.9 [and not 3.8 - which we hope to release next week] Satish ----------- commit e6de09342ce9c4562cc062ff2c1bac4bd956bda0 (HEAD -> balay/update-hypre-v2.14.0) Author: Satish Balay <balay(a)mcs.anl.gov> Date: Fri Mar 30 09:53:12 2018 -0500 hypre: update to version 2.14.0 Reported-by: Antonio Trande <anto.trande(a)gmail.com> diff --git a/config/BuildSystem/config/packages/hypre.py b/config/BuildSystem/config/packages/hypre.py index 1249b39374..620e51602a 100644 --- a/config/BuildSystem/config/packages/hypre.py +++ b/config/BuildSystem/config/packages/hypre.py @@ -4,7 +4,7 @@ import os class Configure(config.package.GNUPackage): def __init__(self, framework): config.package.GNUPackage.__init__(self, framework) - self.gitcommit = 'v2.12.0' + self.gitcommit = 'v2.14.0' self.download = ['git://https://github.com/LLNL/hypre','https://github.com/LLNL/hypre/archive/'+self.gitcommit+'.tar.gz'] self.functions = ['HYPRE_IJMatrixCreate'] self.includes = ['HYPRE.h'] diff --git a/src/mat/impls/hypre/mhypre.c b/src/mat/impls/hypre/mhypre.c index 8a51581a09..65595f2af4 100644 --- a/src/mat/impls/hypre/mhypre.c +++ b/src/mat/impls/hypre/mhypre.c @@ -1450,6 +1450,6 @@ PETSC_EXTERN PetscErrorCode MatCreate_HYPRE(Mat B) static PetscErrorCode hypre_array_destroy(void *ptr) { PetscFunctionBegin; - hypre_TFree(ptr); + hypre_TFree(ptr,HYPRE_MEMORY_HOST); PetscFunctionReturn(0); } On Fri, 30 Mar 2018, Antonio Trande wrote: > Hello everyone. > > 'src/mat/impls/hypre/mhypre.c' file compilation is failed (full log [1]) > with this error: > > mpicc -c -O2 -g -pipe -Wall -Werror=format-security > -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions > -fstack-protector-strong -grecord-gcc-switches > -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic > -fasynchronous-unwind-tables -fstack-clash-protection -mcet > -fcf-protection -O3 -Wl,-z,now -fPIC > -I/builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/include > -I/builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/x86_64/include > -I/usr/include/openmpi-x86_64/hypre -I/usr/include/openmpi-x86_64 > -MMD -MP > /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c > -o x86_64/obj/src/mat/impls/hypre/mhypre.o > /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c: > In function 'hypre_array_destroy': > /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c:1453:19: > error: macro "hypre_TFree" requires 2 arguments, but only 1 given > hypre_TFree(ptr); > ^ > /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c:1453:4: > error: 'hypre_TFree' undeclared (first use in this function); did you > mean 'hypre_Free'? > hypre_TFree(ptr); > ^~~~~~~~~~~ > hypre_Free > /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c:1453:4: > note: each undeclared identifier is reported only once for each function > it appears in > gmake[2]: *** [gmakefile:227: x86_64/obj/src/mat/impls/hypre/mhypre.o] > Error 1 > > Please, take a look. > > [1] https://kojipkgs.fedoraproject.org//work/tasks/6330/26056330/build.log > > System: Fedora 29 (devel branch) > GCC 8.0.1 > OpenMPI 2.1.1 > Hypre 2.14.0 > >

1 0

[PETSc-3.8.4] src/mat/impls/hypre/mhypre.c compilation failed
by Antonio Trande 30 Mar '18

30 Mar '18

Hello everyone. 'src/mat/impls/hypre/mhypre.c' file compilation is failed (full log [1]) with this error: mpicc -c -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -mcet -fcf-protection -O3 -Wl,-z,now -fPIC -I/builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/include -I/builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/x86_64/include -I/usr/include/openmpi-x86_64/hypre -I/usr/include/openmpi-x86_64 -MMD -MP /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c -o x86_64/obj/src/mat/impls/hypre/mhypre.o /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c: In function 'hypre_array_destroy': /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c:1453:19: error: macro "hypre_TFree" requires 2 arguments, but only 1 given hypre_TFree(ptr); ^ /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c:1453:4: error: 'hypre_TFree' undeclared (first use in this function); did you mean 'hypre_Free'? hypre_TFree(ptr); ^~~~~~~~~~~ hypre_Free /builddir/build/BUILD/petsc-3.8.4/buildopenmpi_dir/src/mat/impls/hypre/mhypre.c:1453:4: note: each undeclared identifier is reported only once for each function it appears in gmake[2]: *** [gmakefile:227: x86_64/obj/src/mat/impls/hypre/mhypre.o] Error 1 Please, take a look. [1] https://kojipkgs.fedoraproject.org//work/tasks/6330/26056330/build.log System: Fedora 29 (devel branch) GCC 8.0.1 OpenMPI 2.1.1 Hypre 2.14.0 -- --- Antonio Trande Fedora Project mailto 'sagitter at fedoraproject dot org' GPG key: 0x5E212EE1D35568BE GPG key server: https://keys.fedoraproject.org/

1 0

Re: [petsc-dev] Target next PETSc/TAO release for March 30
by Richard Tran Mills 30 Mar '18

30 Mar '18

All: The recent downtime for the Argonne FTP server threw a wrench into our release plans by causing most nightly builds to fail. We will do the release next week. Please finish your merges as soon as possible, anyway, and then we will (fortuitously) have the few days of feature freeze that Jed asked for. =) --Richard On Fri, Mar 23, 2018 at 10:21 AM, Jed Brown <jed(a)jedbrown.org> wrote: > I think it's feasible, but are we ever going to have more than a week's > notice so that we can have a known merge window followed by a few days > of feature freeze? > > Richard Tran Mills <rtmills(a)anl.gov> writes: > > > All, > > > > To meet deliverables for one the DOE projects providing major funding for > > PETSc development, Barry, Todd, and I agree that we should target Friday, > > March 30 for the next PETSc/TAO release. > > > > Does anyone see any major obstacles to this? Does anyone have anything > that > > will require some extra help/coordination? > > > > Thanks, > > Richard >

1 0

Re: [petsc-dev] TS: -info_exclude and -log_exclude
by Smith, Barry F. 29 Mar '18

29 Mar '18

I think implementing both would be good. > On Mar 29, 2018, at 8:05 AM, Lisandro Dalcin <dalcinl(a)gmail.com> wrote: > > Question for Barry: > > Right now, `-info_exclude ts and `-log_exclude ts` deactivate TS, but > not DMTS, TSAdapt, and TSTrajectory, and there is no way to deactivate > them from cmd line. I want to somehow fix this. > > Option 1: Passing `-xxx_exclude ts` deactivates everything TS-related. > > Option 2: We add `-xxx_exclude <dmts,tsadapt,tstrajectory>` for > fine-grained deactivation. > > Option 3: We implement both option 1 and option 2. > > So, Barry, what's your preference? The quickest one is just (1) but > the others are also rather trivial to implement. > > > > > -- > Lisandro Dalcin > ============ > Research Scientist > Computer, Electrical and Mathematical Sciences & Engineering (CEMSE) > Extreme Computing Research Center (ECRC) > King Abdullah University of Science and Technology (KAUST) > http://ecrc.kaust.edu.sa/ > > 4700 King Abdullah University of Science and Technology > al-Khawarizmi Bldg (Bldg 1), Office # 0109 > Thuwal 23955-6900, Kingdom of Saudi Arabia > http://www.kaust.edu.sa > > Office Phone: +966 12 808-0459

1 0

Re: [petsc-dev] ftp.mcs.anl.gov is down - breaking petsc builds with --download-package
by Satish Balay 29 Mar '18

29 Mar '18

ftp.mcs.anl.gov is now back online [serving ftp:// - http:// is not yet enabled] petsc configure builds with --download-package option should now work. For direct tarball download - switching the URL from http://ftp.mcs.anl.gov to ftp://ftp.mcs.anl.gov notation is the temporary workaround. For eg: change: http://ftp.mcs.anl.gov/pub/petsc/release-snapshots/petsc-3.8.4.tar.gz to: ftp://ftp.mcs.anl.gov/pub/petsc/release-snapshots/petsc-3.8.4.tar.gz Satish On Thu, 29 Mar 2018, Satish Balay wrote: > > We've been informed by the site admin that ftp.mcs.anl.gov is taken > down due to 'cyber incident' - there is no ETA on when it will be > back. > > We've been using it the primary host for distributing petsc tarballs > and hosting some externalpackage tarballs - so some of the petsc buils > with '--download-package' option are failing. > > Obtaining petsc [and some most externalpackages] via git should work. > > I will update when ftp.mcs.anl.gov is brought back up. > > Satish > >

1 0

ftp.mcs.anl.gov is down - breaking petsc builds with --download-package
by Satish Balay 29 Mar '18

29 Mar '18

We've been informed by the site admin that ftp.mcs.anl.gov is taken down due to 'cyber incident' - there is no ETA on when it will be back. We've been using it the primary host for distributing petsc tarballs and hosting some externalpackage tarballs - so some of the petsc buils with '--download-package' option are failing. Obtaining petsc [and some most externalpackages] via git should work. I will update when ftp.mcs.anl.gov is brought back up. Satish

1 0

PETSc 2018, extended abstract deadline: 11th April
by Lawrence Mitchell 29 Mar '18

29 Mar '18

Dear all, a reminder that the PETSc users meeting is in London this summer. The early registration deadline of 11th April is fast approaching, so if you have not yet registered and wish to attend, now is a good time. Also note that the abstract submission deadline has been extended until 11th April as well. All details at www.mcs.anl.gov/petsc/meetings/2018 We're delighted to confirm a number of exciting invited speakers: - Amanda Bienz (UIUC) - Vaclav Hapla (ETH Zurich) - Laetitia Le Pourhiet (UPMC) - Nicole Spillane (Ecole Polytechnique) Looking forward to seeing you in London. Lawrence (for the organising committee)

1 0

TS: -info_exclude and -log_exclude
by Lisandro Dalcin 29 Mar '18

29 Mar '18

Question for Barry: Right now, `-info_exclude ts and `-log_exclude ts` deactivate TS, but not DMTS, TSAdapt, and TSTrajectory, and there is no way to deactivate them from cmd line. I want to somehow fix this. Option 1: Passing `-xxx_exclude ts` deactivates everything TS-related. Option 2: We add `-xxx_exclude <dmts,tsadapt,tstrajectory>` for fine-grained deactivation. Option 3: We implement both option 1 and option 2. So, Barry, what's your preference? The quickest one is just (1) but the others are also rather trivial to implement. -- Lisandro Dalcin ============ Research Scientist Computer, Electrical and Mathematical Sciences & Engineering (CEMSE) Extreme Computing Research Center (ECRC) King Abdullah University of Science and Technology (KAUST) http://ecrc.kaust.edu.sa/ 4700 King Abdullah University of Science and Technology al-Khawarizmi Bldg (Bldg 1), Office # 0109 Thuwal 23955-6900, Kingdom of Saudi Arabia http://www.kaust.edu.sa Office Phone: +966 12 808-0459

1 0

Re: [petsc-dev] how many users subscribe to [email protected]?
by Satish Balay 28 Mar '18

28 Mar '18

There are 734 subscribers - with about 107 disabling e-mail delivery. Satish On Wed, 28 Mar 2018, Fande Kong wrote: > Hi Developers, > > Just be curious about how many users subscribe currently in > petsc-users(a)mcs.anl.gov? In other words, how many users will receive > messages from petsc-users(a)mcs.anl.gov? > > > Thanks, > > Fande >

1 0