April 2013 - petsc-dev - mailman3.cels.anl.gov

Re: [petsc-dev] build problem with suitesparse 4.2
by Satish Balay 24 Apr '13

24 Apr '13

On Wed, 24 Apr 2013, Chaffra Affouda wrote: > Hello, > > I am getting the attached error when building with the latest suitesparse on > ubuntu (suitesparse 4.2). I think umfpack examples must be linked with > libcholmod.so but I might be mistaking. I am building on ubuntu precise --with-umfpack=1 --with-umfpack-include=/usr/include/suitesparse --with-umfpack-lib="[/usr/lib/libumfpack.so,/usr/lib/libamd.so]" --with-cholmod=1 --with-cholmod-include=/usr/include/suitesparse --with-cholmod-lib=/usr/lib/libcholmod.so >> --with-umfpack-lib=['/usr/lib/libumfpack.so', '/usr/lib/libamd.so'] and --with-umfpack-include=['/usr/include/suitesparse'] did not work <<< Possible ERROR while running linker: /usr/lib/libumfpack.so: undefined reference to `cholmod_l_print_common' /usr/lib/libumfpack.so: undefined reference to `cholmod_l_analyze' /usr/lib/libumfpack.so: undefined reference to `cholmod_l_finish' /usr/lib/libumfpack.so: undefined reference to `cholmod_l_free_factor' How about just using: --with-umfpack=1 --with-cholmod=1 [and not specifying the includes/libs - and let configure figure this out from its default configs] Satish

1 0

build problem with suitesparse 4.2
by Chaffra Affouda 24 Apr '13

24 Apr '13

Hello, I am getting the attached error when building with the latest suitesparse on ubuntu (suitesparse 4.2). I think umfpack examples must be linked with libcholmod.so but I might be mistaking. I am building on ubuntu precise Thanks, Chaffra -- Dr. Chaffra Affouda U.S. Naval Research Laboratory Solid State Devices Branch code 6818 4555 Overlook Ave., SW Washington, DC 20375 Ph: (202)-404-2904

1 0

single/double precision return types in blas interface
by Chetan Jhurani 23 Apr '13

23 Apr '13

Hi, There was a discussion few months ago where the problem of single precision f2cblas routines returning double precision values was fixed by creating a new f2cblaslapack package. However, the maint branch still uses f2cblaslapack-3.1.1.q.tar.gz where the signature is wrong: doublereal sdot_ and all that. The fix is in master, which uses f2cblaslapack-3.4.1.q.tar.gz. Perhaps maint should be updated? maint - https://bitbucket.org/petsc/petsc/src/fa9e63e24d90a9e99336ca74ae004b9b5f1da… t=maint master - https://bitbucket.org/petsc/petsc/src/cee652cbe889c53a5c397ca74d3ef6fdcdf8b… t=master Chetan

1 0

[PATCH] config: test functions and headers only once
by Jed Brown 19 Apr '13

19 Apr '13

Many functions and headers are mentioned in both BuildSystem and PETSc, thus being tested more than once. Removing this redundancy speeds up my configure from 125 seconds to 110 seconds. --- The full list of redundant configure tests is shown below. HAVE_LIBLAPACK is anomalous because it is actually testing for different functions. $ grep 'Defined .HAVE_' mpich-basic/conf/configure.log | sed 's,^ \+,,' | sort | uniq -cd 2 Defined "HAVE_ACCESS" to "1" 2 Defined "HAVE_BZERO" to "1" 2 Defined "HAVE_CLOCK" to "1" 2 Defined "HAVE_DLCLOSE" to "1" 2 Defined "HAVE_DLERROR" to "1" 2 Defined "HAVE_DLOPEN" to "1" 2 Defined "HAVE_DLSYM" to "1" 2 Defined "HAVE_DRAND48" to "1" 2 Defined "HAVE_ENDIAN_H" to "1" 2 Defined "HAVE_FCNTL_H" to "1" 2 Defined "HAVE_FENV_H" to "1" 4 Defined "HAVE_FLOAT_H" to "1" 2 Defined "HAVE_FORK" to "1" 2 Defined "HAVE_GETCWD" to "1" 2 Defined "HAVE_GETDOMAINNAME" to "1" 2 Defined "HAVE_GETHOSTBYNAME" to "1" 2 Defined "HAVE_GETHOSTNAME" to "1" 2 Defined "HAVE_GET_NPROCS" to "1" 4 Defined "HAVE_GETPAGESIZE" to "1" 2 Defined "HAVE_GETPWUID" to "1" 2 Defined "HAVE_GETRUSAGE" to "1" 2 Defined "HAVE_GETTIMEOFDAY" to "1" 2 Defined "HAVE_GETWD" to "1" 2 Defined "HAVE_LIBBLAS" to "1" 3 Defined "HAVE_LIBDL" to "1" 27 Defined "HAVE_LIBLAPACK" to "1" 5 Defined "HAVE_LIBM" to "1" 2 Defined "HAVE_LIBZ" to "1" 2 Defined "HAVE_LIMITS_H" to "1" 2 Defined "HAVE_LSEEK" to "1" 2 Defined "HAVE_MALLOC_H" to "1" 2 Defined "HAVE_MEMALIGN" to "1" 2 Defined "HAVE_MEMMOVE" to "1" 2 Defined "HAVE_MEMORY_H" to "1" 2 Defined "HAVE_MKSTEMP" to "1" 2 Defined "HAVE_MPI_FINALIZED" to "1" 2 Defined "HAVE_NETDB_H" to "1" 2 Defined "HAVE_NETINET_IN_H" to "1" 2 Defined "HAVE_POPEN" to "1" 2 Defined "HAVE_PTHREAD_H" to "1" 2 Defined "HAVE_PWD_H" to "1" 2 Defined "HAVE_RAND" to "1" 2 Defined "HAVE_READLINK" to "1" 2 Defined "HAVE_REALPATH" to "1" 2 Defined "HAVE_SCHED_H" to "1" 2 Defined "HAVE_SEARCH_H" to "1" 2 Defined "HAVE_SETJMP_H" to "1" 2 Defined "HAVE_SIGACTION" to "1" 2 Defined "HAVE_SIGNAL" to "1" 2 Defined "HAVE_SIGSET" to "1" 2 Defined "HAVE_SLEEP" to "1" 2 Defined "HAVE_SNPRINTF" to "1" 2 Defined "HAVE_SOCKET" to "1" 2 Defined "HAVE_STDINT_H" to "1" 2 Defined "HAVE_STDLIB_H" to "1" 2 Defined "HAVE_STRCASECMP" to "1" 2 Defined "HAVE_STRING_H" to "1" 2 Defined "HAVE_STRINGS_H" to "1" 2 Defined "HAVE_SYS_PARAM_H" to "1" 2 Defined "HAVE_SYS_PROCFS_H" to "1" 2 Defined "HAVE_SYS_RESOURCE_H" to "1" 2 Defined "HAVE_SYS_SOCKET_H" to "1" 3 Defined "HAVE_SYS_SYSINFO_H" to "1" 2 Defined "HAVE_SYS_TIMES_H" to "1" 2 Defined "HAVE_SYS_TYPES_H" to "1" 2 Defined "HAVE_SYS_UTSNAME_H" to "1" 3 Defined "HAVE_SYS_WAIT_H" to "1" 3 Defined "HAVE_TIME_H" to "1" 2 Defined "HAVE_TIMES" to "1" 2 Defined "HAVE_TIME" to "1" 2 Defined "HAVE_UNAME" to "1" 2 Defined "HAVE_UNISTD_H" to "1" 2 Defined "HAVE_USLEEP" to "1" After this patch, we have a few "redundant" tests left. 2 Defined "HAVE_LIBBLAS" to "1" 3 Defined "HAVE_LIBDL" to "1" 27 Defined "HAVE_LIBLAPACK" to "1" 5 Defined "HAVE_LIBM" to "1" 2 Defined "HAVE_LIBZ" to "1" 2 Defined "HAVE_MPI_FINALIZED" to "1" 2 Defined "HAVE_SYS_SYSINFO_H" to "1" 2 Defined "HAVE_SYS_WAIT_H" to "1" 2 Defined "HAVE_TIME_H" to "1" I have not de-duplicated libraries.py yet because it passes in more structured arguments. It only runs a few tests multiple times: [(['socket', 'nsl'], 'socket'), (['fpe'], 'handle_sigfpes'), (['socket', 'nsl'], 'socket'), (['fpe'], 'handle_sigfpes')] These are actually all defined in one place, but it is visited multiple times: /home/jed/petsc/mpich-basic/conf/reconfigure-mpich-basic.py(12)<module>() -> configure.petsc_configure(configure_options) /home/jed/petsc/config/configure.py(290)petsc_configure() -> framework = config.framework.Framework(['--configModules=PETSc.Configure','--optionsModule=PETSc.compilerOptions']+sys.argv[1:], loadArgDB = 0) /home/jed/petsc/config/BuildSystem/config/framework.py(110)__init__() -> self.createChildren() /home/jed/petsc/config/BuildSystem/config/framework.py(319)createChildren() -> self.getChild(moduleName) /home/jed/petsc/config/BuildSystem/config/framework.py(304)getChild() -> config.setupDependencies(self) > /home/jed/petsc/config/PETSc/Configure.py(105)setupDependencies() -> self.libraries.libraries.extend(libraries1) /home/jed/petsc/mpich-basic/conf/reconfigure-mpich-basic.py(12)<module>() -> configure.petsc_configure(configure_options) /home/jed/petsc/config/configure.py(293)petsc_configure() -> framework.configure(out = sys.stdout) /home/jed/petsc/config/BuildSystem/config/framework.py(929)configure() -> self.updateDependencies() /home/jed/petsc/config/BuildSystem/config/framework.py(385)updateDependencies() -> child.setupDependencies(self) > /home/jed/petsc/config/PETSc/Configure.py(105)setupDependencies() -> self.libraries.libraries.extend(libraries1) config/BuildSystem/config/functions.py | 4 ++++ config/BuildSystem/config/headers.py | 4 ++++ 2 files changed, 8 insertions(+) diff --git a/config/BuildSystem/config/functions.py b/config/BuildSystem/config/functions.py index 30da45e..7b700cb 100644 --- a/config/BuildSystem/config/functions.py +++ b/config/BuildSystem/config/functions.py @@ -197,5 +197,9 @@ choke me self.executeTest(self.checkSignalHandlerType) self.executeTest(self.checkFreeReturnType) self.executeTest(self.checkVariableArgumentLists) + try: + self.functions = list(set(self.functions)) + except: + pass map(lambda function: self.executeTest(self.check, function), self.functions) return diff --git a/config/BuildSystem/config/headers.py b/config/BuildSystem/config/headers.py index e180eaa..f3da45c 100644 --- a/config/BuildSystem/config/headers.py +++ b/config/BuildSystem/config/headers.py @@ -230,6 +230,10 @@ class Configure(config.base.Configure): self.executeTest(self.checkSysWait) self.executeTest(self.checkTime) self.executeTest(self.checkMath) + try: + self.headers = list(set(self.headers)) + except: + pass map(lambda header: self.executeTest(self.check, header), self.headers) self.executeTest(self.checkRecursiveMacros) return -- 1.8.2.1

1 0

Re: [petsc-dev] Should MatRestoreRow zeroes cols and vals?
by Jed Brown 19 Apr '13

19 Apr '13

Patrick Lacasse <patrick.m.lacasse(a)gmail.com> writes: > Hi, > > For some types of matrices, > MatRestoreRow zeroes pointer cols and vals (MatRestoreRow_SeqBAIJ), > and for other not (MatRestoreRow_SeqAIJ). > Should it always put them to NULL ? > It would help us prevent misuses. Yes, thanks for pointing this out. We had a few questionable constructs ourselves. These are in 'next' now. commit 4b4408bbb6af7ebe4b0bfb87714e390b6edcba7c Author: Jed Brown <jed(a)59A2.org> Date: Fri Apr 19 10:52:20 2013 -0500 MatRestoreRow: zero arguments after restore to catch accidental misuse Explicitly pass NULL when we are intentionally using the number of columns after restoring the row. src/dm/impls/composite/packm.c | 2 +- src/ksp/pc/impls/asa/asa.c | 2 +- src/mat/interface/matrix.c | 3 +++ 3 files changed, 5 insertions(+), 2 deletions(-) commit 9383aa0518031aa36d3d785a7263101f32ea5006 Author: Jed Brown <jed(a)59A2.org> Date: Fri Apr 19 10:55:56 2013 -0500 MatRestoreRowIJ/ColumnIJ: zero pointers to catch accidental misuse Only zero in case where the operation is "done" so that user's uninitialized arguments are not overwritten. I don't know if it's best, but this is consistent with done=FALSE meaning that the arguments are not changed. Explicitly pass NULL when intentionally using the number of rows after restoring. src/dm/impls/da/hypre/mhyp.c | 2 +- src/mat/color/color.c | 12 ++++++------ src/mat/examples/tests/ex22.c | 2 +- src/mat/interface/matrix.c | 6 ++++++ src/mat/order/amd/amd.c | 2 +- src/mat/order/sorder.c | 4 ++-- src/mat/order/sp1wd.c | 2 +- src/mat/order/spnd.c | 2 +- src/mat/order/spqmd.c | 2 +- src/mat/order/sprcm.c | 2 +- 10 files changed, 21 insertions(+), 15 deletions(-)

1 0

Should MatRestoreRow zeroes cols and vals?
by Patrick Lacasse 19 Apr '13

19 Apr '13

Hi, For some types of matrices, MatRestoreRow zeroes pointer cols and vals (MatRestoreRow_SeqBAIJ), and for other not (MatRestoreRow_SeqAIJ). Should it always put them to NULL ? It would help us prevent misuses. Patrick Lacasse

1 0

Re: [petsc-dev] [petsc-users] PCView only for one block ?
by Barry Smith 18 Apr '13

18 Apr '13

On Apr 16, 2013, at 9:54 PM, Jed Brown <jedbrown(a)mcs.anl.gov> wrote: > Barry Smith <bsmith(a)mcs.anl.gov> writes: > >>> FWIW, src/snes/examples/tutorials/ex7.c also produces verbose output. I >>> understand the desire to show details when someone configures each >>> subdomain differently, but I'd still rather have scalable output by >>> default. >> >> > Typo. The example I actually ran was src/ksp/ksp/examples/tutorials/ex7.c. What a nightmare, the output between processes is all scrambled. Looks very amateur hour. I'll look at the code, Barry

1 0

Re: [petsc-dev] [petsc-users] Any changes in ML usage between 3.1-p8 -> 3.3-p6?
by Mark F. Adams 18 Apr '13

18 Apr '13

> > On the related issues with the ICC preconditioner, I do suspect that ML has a ( or another) bug in it that is being triggered by the recent ICC changes, but we don't have time to chase it down right now. Note that we have spotted a number of bugs/memory issues in ML over the past couple years, so this wouldn't be surprising. > This does not have anything to do with ML. Its not obvious but PETSc just uses ML to get the AMG prolongation operators and then uses this in its MG framework. So smoothers are all in PETSc and do not interact with ML. hypre, on the other hand, is more monolithic and you get all hypre stuff as a PC: smoothers, etc. The outer Krylov solver is in PETSc. hypre would be worth trying. > Related: I know Bob Taylor fairly well and have followed your work for some time. I was one of the people behind the push to try and use your preconditioner at SIMULA for Abaqus/Standard and possibly Abaqus/CFD. > > Do you think that your preconditioner would do a good job on this seemingly benign, but typically very stiff system? We haven't taken the time to try it with our PETSc interface. > My solver (gamg) uses the same algorithm as ML. You should be able to get the same results, more or less. One important parameter in ML and gamg is the threshold dropping tolerance. This can be very useful for anisotropic problems (grids in your case) because it drops weak connections and gives you semi-coarsening (very useful for anisotropic problems). ML defaults to zero and gamg has a small default value that could make a big difference. You might try playing with "-pc_ml_Threshold <0>". I would try numbers like: 0.0 (the default, just to check) .0001 .001 .01 .05 .1 I would expect that you would see, perhaps large, drops in iteration counts with this but the cost of each iteration does go up. Mark > Thanks again. > > - Mark > > -- > Mark A. Christon > Computational Physics Group (CCS-2) > Computer, Computational and Statistical Sciences Division > Los Alamos National Laboratory > MS D413, P.O. Box 1663 > Los Alamos, NM 87545 > > E-mail: christon(a)lanl.gov > Phone: (505) 663-5124 > Mobile: (505) 695-5649 (voice mail) > > International Journal for Numerical Methods in Fluids > > From: "Mark F. Adams" <mark.adams(a)columbia.edu> > Date: Thu, 18 Apr 2013 10:12:51 -0400 > To: "Mark A. Christon" <christon(a)lanl.gov> > Cc: PETSc users list <petsc-users(a)mcs.anl.gov>, "Mark F. Adams" <mark.adams(a)columbia.edu> > Subject: Re: [petsc-users] Any changes in ML usage between 3.1-p8 -> 3.3-p6? > >> Note, you need to damp jacobi for a smoother with something like this: >> >> -mg_levels_ksp_type chebyshev >> -mg_levels_ksp_chebyshev_estimate_eigenvalues 0,0.1,0,1.05 >> >> >> On Apr 18, 2013, at 9:49 AM, "Christon, Mark A" <christon(a)lanl.gov> wrote: >> >>> HI Mark, >>> >>> Thanks for the information. We thought something had changed and could see it the effect, but couldn't quite pin it down. >>> >>> To be clear our pressure-Poisson equation is a warm and fluffy Laplacian, but typically quite stiff from a spectral point of view when dealing with boundary-layer meshes and complex geometry — our norm. This is the first-order computational cost in our flow solver, so hit's like the recent change are very problematic, and particularly so when they break a number of regression tests that run nightly across multiple platflorms. >>> >>> So, unfortunately, while I'd like to use something as Jacobi, it's completely ineffective in for the operator (and RHS) in question. >>> >>> Thanks. >>> >>> - Mark >>> >>> -- >>> Mark A. Christon >>> Computational Physics Group (CCS-2) >>> Computer, Computational and Statistical Sciences Division >>> Los Alamos National Laboratory >>> MS D413, P.O. Box 1663 >>> Los Alamos, NM 87545 >>> >>> E-mail: christon(a)lanl.gov >>> Phone: (505) 663-5124 >>> Mobile: (505) 695-5649 (voice mail) >>> >>> International Journal for Numerical Methods in Fluids >>> >>> From: "Mark F. Adams" <mark.adams(a)columbia.edu> >>> Reply-To: PETSc users list <petsc-users(a)mcs.anl.gov> >>> Date: Wed, 17 Apr 2013 19:42:47 -0400 >>> To: PETSc users list <petsc-users(a)mcs.anl.gov> >>> Subject: Re: [petsc-users] Fwd: Any changes in ML usage between 3.1-p8 -> 3.3-p6? >>> >>>> In looking at the logs for icc it looks like Hong has done a little messing around with the shifting tolerance: >>>> >>>> - ((PC_Factor*)icc)->info.shiftamount = 1.e-12; >>>> - ((PC_Factor*)icc)->info.zeropivot = 1.e-12; >>>> + ((PC_Factor*)icc)->info.shiftamount = 100.0*PETSC_MACHINE_EPSILON; >>>> + ((PC_Factor*)icc)->info.zeropivot = 100.0*PETSC_MACHINE_EPSILON; >>>> >>>> >>>> This looks like it would lower the shifting and drop tolerance. You might set these back to 1e-12. >>>> >>>> http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/PC/PCFactorSetZ… >>>> >>>> http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/PC/PCFactorSetS… >>>> >>>> BTW, using an indefinite preconditioner, that has to be fixed with is-this-a-small-number kind of code, on a warm and fluffy Laplacian is not recommended. As I said before I would just use jacobi -- god gave you an easy problem. Exploit it. >>>> >>>> On Apr 17, 2013, at 7:22 PM, "Mark F. Adams" <mark.adams(a)columbia.edu> wrote: >>>> >>>>> >>>>> >>>>> Begin forwarded message: >>>>> >>>>>> From: "Christon, Mark A" <christon(a)lanl.gov> >>>>>> Subject: Re: [petsc-users] Any changes in ML usage between 3.1-p8 -> 3.3-p6? >>>>>> Date: April 17, 2013 7:06:11 PM EDT >>>>>> To: "Mark F. Adams" <mark.adams(a)columbia.edu>, "Bakosi, Jozsef" <jbakosi(a)lanl.gov> >>>>>> >>>>>> Hi Mark, >>>>>> >>>>>> Yes, looks like the new version does a little better after 2 iterations, but at the 8th iteration, the residuals increase:( >>>>>> >>>>>> I suspect this is why PETSc is whining about an indefinite preconditioner. >>>>>> >>>>>> Something definitely changes as we've had about 6-8 regression tests start failing that have been running flawlessly with ML + PETSc 3.1-p8 for almost two years. >>>>>> >>>>>> If we can understand what changed, we probably have a fighting chance of correcting it — assuming it's some solver setting for PETSc that we're not currently using. >>>>>> >>>>>> - Mark >>>>>> >>>>>> -- >>>>>> Mark A. Christon >>>>>> Computational Physics Group (CCS-2) >>>>>> Computer, Computational and Statistical Sciences Division >>>>>> Los Alamos National Laboratory >>>>>> MS D413, P.O. Box 1663 >>>>>> Los Alamos, NM 87545 >>>>>> >>>>>> E-mail: christon(a)lanl.gov >>>>>> Phone: (505) 663-5124 >>>>>> Mobile: (505) 695-5649 (voice mail) >>>>>> >>>>>> International Journal for Numerical Methods in Fluids >>>>>> >>>>>> From: "Mark F. Adams" <mark.adams(a)columbia.edu> >>>>>> Date: Wed, 17 Apr 2013 18:51:02 -0400 >>>>>> To: PETSc users list <petsc-users(a)mcs.anl.gov> >>>>>> Cc: "Mark A. Christon" <christon(a)lanl.gov> >>>>>> Subject: Re: [petsc-users] Any changes in ML usage between 3.1-p8 -> 3.3-p6? >>>>>> >>>>>>> I see you are using icc. Perhaps our icc changed a bit between versions. These results look like both solves are working and the old does a little better (after two iterations). >>>>>>> >>>>>>> Try using jacobi instead of icc. >>>>>>> >>>>>>> >>>>>>> On Apr 17, 2013, at 6:21 PM, Jozsef Bakosi <jbakosi(a)lanl.gov> wrote: >>>>>>> >>>>>>>>> On 04.17.2013 15:38, Matthew Knepley wrote: >>>>>>>>>> On 04.17.2013 14:26, Jozsef Bakosi wrote: >>>>>>>>>>> Mark F. Adams mark.adams at columbia.edu >>>>>>>>>>> Wed Apr 17 14:25:04 CDT 2013 >>>>>>>>>>> 2) If you get "Indefinite PC" (I am guessing from using CG) it is because the >>>>>>>>>>> preconditioner >>>>>>>>>>> really is indefinite (or possible non-symmetric). We improved the checking >>>>>>>>>>> for this in one >>>>>>>>>>> of those releases. >>>>>>>>>>> AMG does not guarantee an SPD preconditioner so why persist in trying to use >>>>>>>>>>> CG? >>>>>>>>>>> AMG is positive if everything is working correctly. >>>>>>>>>>> Are these problems only semidefinite? Singular systems can give erratic >>>>>>>>>>> behavior. >>>>>>>>>> It is a Laplace operator from Galerkin finite elements. And the PC is fine on >>>>>>>>>> ranks 1, 2, 3, and 5 -- indefinite only on 4. I think we can safely say that the >>>>>>>>>> same PC should be positive on 4 as well. >>>>>>>>> Why is it safe? Because it sounds plausible? Mathematics is replete with things >>>>>>>>> that sound plausible and are false. Are there proofs that suggest this? Is there >>>>>>>>> computational evidence? Why would I believe you? >>>>>>>> Okay, so here is some additional information: >>>>>>>> I tried both old and new PETSc versions again, but now only taking 2 iterations >>>>>>>> (both with 4 CPUs) and checked the residuals. I get the same exact PC from ML in >>>>>>>> both cases, however, the residuals are different after both iterations: >>>>>>>> Please do a diff on the attached files and you can verify that the ML >>>>>>>> diagnostics are exactly the same: same max eigenvalues, nodes aggregated, etc, >>>>>>>> while the norm coming out of the solver at the end at both iterations are >>>>>>>> different. >>>>>>>> We reproduced the same exact behavior on two different linux platforms. >>>>>>>> Once again: same application source code, same ML source code, different PETSc: >>>>>>>> 3.1-p8 vs. 3.3-p6. >>>>>>>> <old.out><new.out> >>>>>>> >>>>>>> >>>>> >>>> >>

1 0

Re: [petsc-dev] Fwd: [petsc-users] PCView only for one block ?
by Jed Brown 16 Apr '13

16 Apr '13

Barry Smith <bsmith(a)mcs.anl.gov> writes: >> FWIW, src/snes/examples/tutorials/ex7.c also produces verbose output. I >> understand the desire to show details when someone configures each >> subdomain differently, but I'd still rather have scalable output by >> default. > > It doesn't for me. In fact in next it doesn't even run in parallel, is this expected behavior? Typo. The example I actually ran was src/ksp/ksp/examples/tutorials/ex7.c. > ~/Src/petsc/src/snes/examples/tutorials next $ petscmpiexec -n 2 ./ex7 -snes_view You're right, this example isn't set up for parallel. I'll add a guard. > [1]PETSC ERROR: ------------------------------------------------------------------------ > [1]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably memory access out of range > [1]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger > [1]PETSC ERROR: or see http://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind[1]PETSC ERROR: or try http://valgrind.org on GNU/linux and Apple Mac OS X to find memory corruption errors > [1]PETSC ERROR: likely location of problem given in stack below > [1]PETSC ERROR: --------------------- Stack Frames ------------------------------------ > [1]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, > [1]PETSC ERROR: INSTEAD the line number of the start of the function > [1]PETSC ERROR: is given. > [1]PETSC ERROR: [1] FormFunctionLocal line 377 src/snes/examples/tutorials/ex7.c > [1]PETSC ERROR: [1] SNESComputeFunction_DMDA line 79 /Users/barrysmith/Src/PETSc/src/snes/utils/dmdasnes.c > [1]PETSC ERROR: [1] SNES user function line 0 unknownunknown > [1]PETSC ERROR: [1] SNESComputeFunction line 2016 /Users/barrysmith/Src/PETSc/src/snes/interface/snes.c > [1]PETSC ERROR: [1] SNESSolve_NEWTONLS line 145 /Users/barrysmith/Src/PETSc/src/snes/impls/ls/ls.c > [1]PETSC ERROR: [1] SNESSolve line 3708 /Users/barrysmith/Src/PETSc/src/snes/interface/snes.c > [1]PETSC ERROR: --------------------- Error Message ------------------------------------ > [1]PETSC ERROR: Signal received! > [1]PETSC ERROR: ------------------------------------------------------------------------ > [1]PETSC ERROR: Petsc Development GIT revision: acc80a61d235fd808c6194a016f8a1e2727e30e4 GIT Date: 2013-04-07 17:28:52 -0500 > [1]PETSC ERROR: See docs/changes/index.html for recent updates. > [1]PETSC ERROR: See docs/faq.html for hints about trouble shooting. > [1]PETSC ERROR: See docs/index.html for manual pages. > [1]PETSC ERROR: ------------------------------------------------------------------------ > [1]PETSC ERROR: ./ex7 on a arch-g++ named Barrys-MacBook-Pro.local by barrysmith Tue Apr 16 21:05:19 2013 > [1]PETSC ERROR: Libraries linked from /Users/barrysmith/Src/PETSc/arch-g++/lib > [1]PETSC ERROR: Configure run at Sun Apr 7 17:35:21 2013 > [1]PETSC ERROR: Configure options --download-mpich --with-clanguage=c++ --with-dynamic-loading --with-shared-libraries --with-single-library=0 PETSC_ARCH=arch-g++ --with-fc=0 > [1]PETSC ERROR: ------------------------------------------------------------------------ > [1]PETSC ERROR: User provided function() line 0 in unknown directory unknown file > application called MPI_Abort(MPI_COMM_WORLD, 59) - process 1 > [cli_1]: aborting job: > application called MPI_Abort(MPI_COMM_WORLD, 59) - process 1 > > =================================================================================== > = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES > = EXIT CODE: 59 > = CLEANING UP REMAINING PROCESSES > = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES > ===================================================================================

1 0

Re: [petsc-dev] Fwd: [petsc-users] PCView only for one block ?
by Barry Smith 16 Apr '13

16 Apr '13

On Apr 16, 2013, at 8:51 PM, Jed Brown <jedbrown(a)mcs.anl.gov> wrote: > Barry Smith <bsmith(a)mcs.anl.gov> writes: > >> Someone has been mucking with PCView_BJacobi? Under normal >> > > Mikolaj, can you show the code that you use for setting up this bjacobi? > >> > > FWIW, src/snes/examples/tutorials/ex7.c also produces verbose output. I > understand the desire to show details when someone configures each > subdomain differently, but I'd still rather have scalable output by > default. It doesn't for me. In fact in next it doesn't even run in parallel, is this expected behavior? ~/Src/petsc/src/snes/examples/tutorials next $ petscmpiexec -n 2 ./ex7 -snes_view [1]PETSC ERROR: ------------------------------------------------------------------------ [1]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably memory access out of range [1]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger [1]PETSC ERROR: or see http://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind[1]PETSC ERROR: or try http://valgrind.org on GNU/linux and Apple Mac OS X to find memory corruption errors [1]PETSC ERROR: likely location of problem given in stack below [1]PETSC ERROR: --------------------- Stack Frames ------------------------------------ [1]PETSC ERROR: Note: The EXACT line numbers in the stack are not available, [1]PETSC ERROR: INSTEAD the line number of the start of the function [1]PETSC ERROR: is given. [1]PETSC ERROR: [1] FormFunctionLocal line 377 src/snes/examples/tutorials/ex7.c [1]PETSC ERROR: [1] SNESComputeFunction_DMDA line 79 /Users/barrysmith/Src/PETSc/src/snes/utils/dmdasnes.c [1]PETSC ERROR: [1] SNES user function line 0 unknownunknown [1]PETSC ERROR: [1] SNESComputeFunction line 2016 /Users/barrysmith/Src/PETSc/src/snes/interface/snes.c [1]PETSC ERROR: [1] SNESSolve_NEWTONLS line 145 /Users/barrysmith/Src/PETSc/src/snes/impls/ls/ls.c [1]PETSC ERROR: [1] SNESSolve line 3708 /Users/barrysmith/Src/PETSc/src/snes/interface/snes.c [1]PETSC ERROR: --------------------- Error Message ------------------------------------ [1]PETSC ERROR: Signal received! [1]PETSC ERROR: ------------------------------------------------------------------------ [1]PETSC ERROR: Petsc Development GIT revision: acc80a61d235fd808c6194a016f8a1e2727e30e4 GIT Date: 2013-04-07 17:28:52 -0500 [1]PETSC ERROR: See docs/changes/index.html for recent updates. [1]PETSC ERROR: See docs/faq.html for hints about trouble shooting. [1]PETSC ERROR: See docs/index.html for manual pages. [1]PETSC ERROR: ------------------------------------------------------------------------ [1]PETSC ERROR: ./ex7 on a arch-g++ named Barrys-MacBook-Pro.local by barrysmith Tue Apr 16 21:05:19 2013 [1]PETSC ERROR: Libraries linked from /Users/barrysmith/Src/PETSc/arch-g++/lib [1]PETSC ERROR: Configure run at Sun Apr 7 17:35:21 2013 [1]PETSC ERROR: Configure options --download-mpich --with-clanguage=c++ --with-dynamic-loading --with-shared-libraries --with-single-library=0 PETSC_ARCH=arch-g++ --with-fc=0 [1]PETSC ERROR: ------------------------------------------------------------------------ [1]PETSC ERROR: User provided function() line 0 in unknown directory unknown file application called MPI_Abort(MPI_COMM_WORLD, 59) - process 1 [cli_1]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 59) - process 1 =================================================================================== = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES = EXIT CODE: 59 = CLEANING UP REMAINING PROCESSES = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES ===================================================================================

1 0