On Mon, Oct 8, 2012 at 3:31 AM, Aron Ahmadia <span dir="ltr"><<a href="mailto:aron@ahmadia.net" target="_blank">aron@ahmadia.net</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Dang, that's almost a perfect match to my gcc:<br>
<br>
i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)<br>
<br>
We can definitely reproduce this on two 10.6 systems out here at KAUST.<br>
Ketch and I are both using "brew" on these boxen, but this is fairly<br>
non-invasive with regards to this stack.<br>
<br>
Matt, Satish, can you guys get your config and make logs to me somehow?<br>
When I get some time I will try to bisect back to the commit in -dev that<br>
is causing this.<br></blockquote><div><br></div><div>I will send it when not in the airport. However, the fastest way to debug this is</div><div>to run with the debugger and see what the SEGV is actually coming from.</div>
<div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
A<br>
<br>
On Sun, Oct 7, 2012 at 10:16 PM, Matthew Knepley <<a href="mailto:petsc-maint@mcs.anl.gov">petsc-maint@mcs.anl.gov</a>>wrote:<br>
<br>
> On Sun, Oct 7, 2012 at 2:36 PM, Aron Ahmadia <<a href="mailto:aron.ahmadia@gmail.com">aron.ahmadia@gmail.com</a>>wrote:<br>
><br>
>> Please report your gcc compiler versions.<br>
>><br>
><br>
> knepley:/PETSc3/petsc/petsc-dev$ mpicc --show<br>
> mpicc --show<br>
> gcc -D_REENTRANT -Wl,-u,_munmap -Wl,-multiply_defined,suppress -lmpi<br>
> -lopen-rte -lopen-pal<br>
> knepley:/PETSc3/petsc/petsc-dev$ gcc --version<br>
> gcc --version<br>
> i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5664)<br>
> Copyright (C) 2007 Free Software Foundation, Inc.<br>
> This is free software; see the source for copying conditions. There is NO<br>
> warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.<br>
><br>
> Matt<br>
><br>
><br>
>> A<br>
>><br>
>> Sent from my iPhone<br>
>><br>
>> On Oct 7, 2012, at 7:49 PM, Matthew Knepley <<a href="mailto:petsc-maint@mcs.anl.gov">petsc-maint@mcs.anl.gov</a>><br>
>> wrote:<br>
>><br>
>> > On Sun, Oct 7, 2012 at 12:48 PM, Satish Balay <<a href="mailto:petsc-maint@mcs.anl.gov">petsc-maint@mcs.anl.gov</a>><br>
>> wrote:<br>
>> > My test ran fine again.<br>
>> ><br>
>> > Perhaps you can run the example manually in a debugger to get a better<br>
>> > trace.<br>
>> ><br>
>> > Yes, I can't reproduce this either.<br>
>> ><br>
>> > Matt<br>
>> ><br>
>> ><br>
>> > Satish<br>
>> ><br>
>> > On Sun, 7 Oct 2012, Satish Balay wrote:<br>
>> ><br>
>> > > ok - then the issue is unrelated to fortran. I'll redo the 10.6 build<br>
>> > > and see if its reproduceable.<br>
>> > ><br>
>> > > and secondly there is some wierd bug where fortran interface is not<br>
>> > > getting built [you could try 'make all-legacy' and see if it gets<br>
>> > > built correctly]<br>
>> > ><br>
>> > > Also gfortran-4.2* is buggy [wrt some of the f90 interface<br>
>> > > functionality]. gfortran >= 4.3 is prefered for that. [but if not<br>
>> > > using f90 interfaces - it doesn't matter]<br>
>> > ><br>
>> > > Satish<br>
>> > ><br>
>> > > On Sun, 7 Oct 2012, Aron Ahmadia wrote:<br>
>> > ><br>
>> > > > reproduced with fortran enabled, additionally:<br>
>> > > ><br>
>> > > > /Users/aron/sandbox/petsc/arch-darwin-c-debug/bin/mpif90 -c -Wall<br>
>> > > > -Wno-unused-variable -g -I/Users/aron/sandbox/petsc/include<br>
>> > > > -I/Users/aron/sandbox/petsc/arch-darwin-c-debug/include -o<br>
>> ex5f.o ex5f.F<br>
>> > > > Warning: Nonconforming tab character in column 1 of line 2757<br>
>> > > > Warning: Nonconforming tab character in column 1 of line 4937<br>
>> > > > Warning: Nonconforming tab character in column 1 of line 6914<br>
>> > > > Warning: Nonconforming tab character in column 1 of line 8887<br>
>> > > > Warning: Nonconforming tab character in column 1 of line 10901<br>
>> > > > /Users/aron/sandbox/petsc/arch-darwin-c-debug/bin/mpif90<br>
>> > > > -Wl,-multiply_defined,suppress -Wl,-multiply_defined -Wl,suppress<br>
>> > > > -Wl,-commons,use_dylibs -Wl,-search_paths_first<br>
>> > > > -Wl,-multiply_defined,suppress -Wl,-multiply_defined -Wl,suppress<br>
>> > > > -Wl,-commons,use_dylibs -Wl,-search_paths_first -Wall<br>
>> > > > -Wno-unused-variable -g -o ex5f ex5f.o<br>
>> > > > -L/Users/aron/sandbox/petsc/arch-darwin-c-debug/lib -lpetsc<br>
>> > > > -L/usr/X11R6/lib -lX11 -lpthread -llapack -lblas<br>
>> > > > -L/usr/lib/gcc/i686-apple-darwin10/4.2.1/x86_64<br>
>> > > > -L/usr/lib/i686-apple-darwin10/4.2.1<br>
>> > > > -L/usr/lib/gcc/i686-apple-darwin10/4.2.1 -lmpichf90 -lgfortran<br>
>> > > ><br>
>> -L/usr/local/Cellar/gfortran/4.2.4-5664/lib/gcc/i686-apple-darwin10/4.2.1/x86_64<br>
>> > > ><br>
>> -L/usr/local/Cellar/gfortran/4.2.4-5664/lib/gcc/i686-apple-darwin10/4.2.1<br>
>> > > > -L/usr/local/Cellar/gfortran/4.2.4-5664/lib/gcc<br>
>> > > > -L/usr/local/Cellar/gfortran/4.2.4-5664/lib -ldl -lpmpich -lmpich<br>
>> -lopa<br>
>> > > > -lmpl -lpthread -lSystem -ldl<br>
>> > > > Undefined symbols:<br>
>> > > > "_dmlocaltoglobalbegin_", referenced from:<br>
>> > > > _forminitialguess_ in ex5f.o<br>
>> > > > "_dmgetlocalvector_", referenced from:<br>
>> > > > _forminitialguess_ in ex5f.o<br>
>> > > > "_matassemblybegin_", referenced from:<br>
>> > > > _formjacobianlocal_ in ex5f.o<br>
>> > > > "_dmdestroy_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_snessolve_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_vecrestorearray_", referenced from:<br>
>> > > > _forminitialguess_ in ex5f.o<br>
>> > > > "_vecdestroy_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_matsetvalueslocal_", referenced from:<br>
>> > > > _formjacobianlocal_ in ex5f.o<br>
>> > > > _formjacobianlocal_ in ex5f.o<br>
>> > > > "_matassemblyend_", referenced from:<br>
>> > > > _formjacobianlocal_ in ex5f.o<br>
>> > > > "_dmdagetcorners_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_dmdasetlocaljacobian_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_petscinitialize_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_dmcreateglobalvector_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_petscfinalize_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_snessetdm_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_snessetfromoptions_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_snescreate_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_snesdestroy_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_dmdacreate2d_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_vecduplicate_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_petsclogflops_", referenced from:<br>
>> > > > _formfunctionlocal_ in ex5f.o<br>
>> > > > "_dmdagetghostcorners_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_dmrestorelocalvector_", referenced from:<br>
>> > > > _forminitialguess_ in ex5f.o<br>
>> > > > "_snesgetiterationnumber_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_dmlocaltoglobalend_", referenced from:<br>
>> > > > _forminitialguess_ in ex5f.o<br>
>> > > > "_dmdasetlocalfunction_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_dmdagetinfo_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_petscoptionsgetreal_", referenced from:<br>
>> > > > _MAIN__ in ex5f.o<br>
>> > > > "_vecgetarray_", referenced from:<br>
>> > > > _forminitialguess_ in ex5f.o<br>
>> > > > ld: symbol(s) not found<br>
>> > > > collect2: ld returned 1 exit status<br>
>> > > > make[3]: [ex5f] Error 1 (ignored)<br>
>> > > ><br>
>> > > > On Sun, Oct 7, 2012 at 7:29 PM, Satish Balay <<br>
>> <a href="mailto:petsc-maint@mcs.anl.gov">petsc-maint@mcs.anl.gov</a>>wrote:<br>
>> > > ><br>
>> > > > > Hm - nightlybuilds use --download-mpich=1 - but I don't see this<br>
>> error<br>
>> > > > > on OSX build.<br>
>> > > > ><br>
>> > > > > wrt binsection - the following worked - the last time I tried.<br>
>> > > > ><br>
>> > > > > <a href="http://www.mcs.anl.gov/petsc/documentation/faq.html#bisect" target="_blank">http://www.mcs.anl.gov/petsc/documentation/faq.html#bisect</a><br>
>> > > > ><br>
>> > > > > Satish<br>
>> > > > ><br>
>> > > > > On Sun, 7 Oct 2012, Aron Ahmadia wrote:<br>
>> > > > ><br>
>> > > > > > reproduced with --download-mpich=1<br>
>> > > > > ><br>
>> > > > > > how the heck do you guys bisect with two independent repos?<br>
>> > > > > ><br>
>> > > > > > A<br>
>> > > > > ><br>
>> > > > > > On Sun, Oct 7, 2012 at 6:28 PM, Aron Ahmadia <<a href="mailto:aron@ahmadia.net">aron@ahmadia.net</a>><br>
>> wrote:<br>
>> > > > > ><br>
>> > > > > > > I'm able to reproduce with shared libraries enabled and<br>
>> disabled. Your<br>
>> > > > > > > nightly logs aren't indicating failures as far as I can tell.<br>
>> > > > > > ><br>
>> > > > > > > changeset: 24541:729963bc657c<br>
>> > > > > > > tag: tip<br>
>> > > > > > > user: Stefano Zampini <<a href="mailto:stefano.zampini@gmail.com">stefano.zampini@gmail.com</a>><br>
>> > > > > > > date: Sat Oct 06 13:39:31 2012 +0200<br>
>> > > > > > > summary: Rearranged some deallocations when building<br>
>> FETIDP mat.<br>
>> > > > > > > (same failure with 2 processes), I'm looking into whether any<br>
>> of the<br>
>> > > > > > > specific configure options (python/shared libraries) triggers<br>
>> the<br>
>> > > > > failure.<br>
>> > > > > > ><br>
>> > > > > > > ~/s/petsc ❯❯❯ make PETSC_DIR=/Users/aron/sandbox/petsc<br>
>> > > > > > > PETSC_ARCH=arch-darwin-c-debug test<br>
>> > > > > > > Running test examples to verify correct installation<br>
>> > > > > > > Using PETSC_DIR=/Users/aron/sandbox/petsc and<br>
>> > > > > > > PETSC_ARCH=arch-darwin-c-debug<br>
>> > > > > > > Possible error running C/C++ src/snes/examples/tutorials/ex19<br>
>> with 1<br>
>> > > > > MPI<br>
>> > > > > > > process<br>
>> > > > > > > See <a href="http://www.mcs.anl.gov/petsc/documentation/faq.html" target="_blank">http://www.mcs.anl.gov/petsc/documentation/faq.html</a><br>
>> > > > > > > [peregrine:31992] *** Process received signal ***<br>
>> > > > > > > [peregrine:31992] Signal: Segmentation fault (11)<br>
>> > > > > > > [peregrine:31992] Signal code: Address not mapped (1)<br>
>> > > > > > > [peregrine:31992] Failing at address: 0x440000b0<br>
>> > > > > > > [peregrine:31992] [ 0] 2 libSystem.B.dylib<br>
>> > > > > > > 0x00007fff846391ba _sigtramp + 26<br>
>> > > > > > > [peregrine:31992] [ 1] 3 ???<br>
>> > > > > > > 0x0000000101434fd0 0x0 + 4316155856<br>
>> > > > > > > [peregrine:31992] [ 2] 4 libpetsc.dylib<br>
>> > > > > > > 0x0000000100118456 PetscInitialize + 1424<br>
>> > > > > > > [peregrine:31992] [ 3] 5 ex19<br>
>> > > > > > > 0x000000010000110f main + 43<br>
>> > > > > > > [peregrine:31992] [ 4] 6 ex19<br>
>> > > > > > > 0x00000001000010dc start + 52<br>
>> > > > > > > [peregrine:31992] [ 5] 7 ???<br>
>> > > > > > > 0x0000000000000007 0x0 + 7<br>
>> > > > > > > [peregrine:31992] *** End of error message ***<br>
>> > > > > > > mpiexec noticed that job rank 0 with PID 31992 on node<br>
>> peregrine.local<br>
>> > > > > > > exited on signal 11 (Segmentation fault).<br>
>> > > > > > ><br>
>> > > > > ><br>
>> > > > > ><br>
>> > > > ><br>
>> > > ><br>
>> > > ><br>
>> > ><br>
>> > ><br>
>> ><br>
>> ><br>
>> ><br>
>> ><br>
>> > --<br>
>> > What most experimenters take for granted before they begin their<br>
>> experiments is infinitely more interesting than any results to which their<br>
>> experiments lead.<br>
>> > -- Norbert Wiener<br>
>><br>
>><br>
><br>
<span class="HOEnZb"><font color="#888888">><br>
> --<br>
> What most experimenters take for granted before they begin their<br>
> experiments is infinitely more interesting than any results to which their<br>
> experiments lead.<br>
> -- Norbert Wiener<br>
><br>
<br>
</font></span></blockquote></div><br><br clear="all"><div><br></div>-- <br>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>
-- Norbert Wiener<br>