[petsc-users] Product of matrix row times a vector

[Matthew Knepley]

On Fri, Jan 31, 2020 at 4:19 AM Jeremy Theler <jeremy at seamplex.com> wrote:

>
> On Thu, 2020-01-30 at 18:05 -0500, Matthew Knepley wrote:
> > On Thu, Jan 30, 2020 at 6:04 PM Jeremy Theler <jeremy at seamplex.com>
> > wrote:
> > > On Thu, 2020-01-30 at 21:10 +0000, Smith, Barry F. wrote:
> > >
> > > >   MatGetSubMatrix() and then do the product on the sub matrix
> > > then
> > > > VecSum
> > > >
> > >
> > > Ok, I have it working in a single-processor and throws the expected
> > > value. Yet I have a segfault in parallel when I ask for the IS
> > > corresponding to the rows. When I call this instruction in parallel
> > > I
> > > get a segfault (I can post the full debug output if needed):
> > >
> > > ISCreateStride(PETSC_COMM_WORLD, size_local, first_row, 1,
> > > &set_cols);
> > >
> > > I tried also
> > >
> > > ISCreateStride(PETSC_COMM_WORLD, size_global, 0, 1, &set_cols));
> > >
> > > but it also fails with the same segfault.
> > >
> > > What am I getting wrong?
> >
> > Show the entire error, including stack.
> >
>
> I run in two processes with start_on_debugger. Main terminal says
>
> malloc(): corrupted top size
>

The arguments look fine. I would run in valgrind, since it seems like you
have memory corruption somewhere else in the code.

  Thanks,

     Matt


> and the gdb output with the stack trace is
>
>
> Attaching to program:
> /home/gtheler/codigos/wasora-suite/fino/examples/fino, process 31192
> [New LWP 31196]
> [New LWP 31198]
> [Thread debugging using libthread_db enabled]
> Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
> 0x00007fd236019720 in __GI___nanosleep (
>     requested_time=requested_time at entry=0x7ffdf9838190,
>     remaining=remaining at entry=0x7ffdf9838190)
>     at ../sysdeps/unix/sysv/linux/nanosleep.c:28
> 28      ../sysdeps/unix/sysv/linux/nanosleep.c: No such file or directory.
> (gdb) c
> Continuing.
> [New Thread 0x7fd233034700 (LWP 31212)]
> Thread 1 "fino" received signal SIGABRT, Aborted.
> __GI_raise (sig=sig at entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
> 50      ../sysdeps/unix/sysv/linux/raise.c: No such file or directory.
> (gdb) where
> #0  __GI_raise (sig=sig at entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
> #1  0x00007fd235f75535 in __GI_abort () at abort.c:79
> #2  0x00007fd235fcc508 in __libc_message (action=action at entry=do_abort,
>     fmt=fmt at entry=0x7fd2360d728d "%s\n") at
> ../sysdeps/posix/libc_fatal.c:181
> #3  0x00007fd235fd2c1a in malloc_printerr (
>     str=str at entry=0x7fd2360d5518 "malloc(): corrupted top size")
>     at malloc.c:5341
> #4  0x00007fd235fd620d in _int_malloc (
>     av=av at entry=0x7fd23610ec40 <main_arena>, bytes=bytes at entry=2372)
>     at malloc.c:4099
> #5  0x00007fd235fd756a in __GI___libc_malloc (bytes=2372) at malloc.c:3057
> #6  0x00007fd2377dd42d in PetscMallocAlign (mem=2372, clear=PETSC_TRUE,
>     line=37, func=0x7fd239450408 <__func__.14861> "ISCreate",
>     file=0x7fd239450298
> "/home/gtheler/libs/petsc-3.12.3/src/vec/is/is/interface/isreg.c",
> result=0x7ffdf983c618)
>     at /home/gtheler/libs/petsc-3.12.3/src/sys/memory/mal.c:49
> #7  0x00007fd2377e0875 in PetscTrMallocDefault (a=768, clear=PETSC_TRUE,
>     lineno=37, function=0x7fd239450408 <__func__.14861> "ISCreate",
>     filename=0x7fd239450298
> "/home/gtheler/libs/petsc-3.12.3/src/vec/is/is/interface/isreg.c",
> result=0x7ffdf983c8b0)
>     at /home/gtheler/libs/petsc-3.12.3/src/sys/memory/mtr.c:164
> #8  0x00007fd2377dedf0 in PetscMallocA (n=1, clear=PETSC_TRUE, lineno=37,
>     function=0x7fd239450408 <__func__.14861> "ISCreate",
> --Type <RET> for more, q to quit, c to continue without paging--
>     filename=0x7fd239450298
> "/home/gtheler/libs/petsc-3.12.3/src/vec/is/is/interface/isreg.c",
> bytes0=768, ptr0=0x7ffdf983c8b0)
>     at /home/gtheler/libs/petsc-3.12.3/src/sys/memory/mal.c:422
> #9  0x00007fd237c67760 in ISCreate (comm=0x556dbe8c9ee0
> <ompi_mpi_comm_world>,
>     is=0x7ffdf983c8b0)
>     at /home/gtheler/libs/petsc-3.12.3/src/vec/is/is/interface/isreg.c:37
> #10 0x00007fd237c4589a in ISCreateStride (
>     comm=0x556dbe8c9ee0 <ompi_mpi_comm_world>, n=70188, first=70191,
> step=1,
>     is=0x7ffdf983c8b0)
> #11 0x0000556dbe82aab4 in fino_instruction_reaction (arg=0x556dbf8b4460)
>     at ./reactions.c:72
> #12 0x0000556dbe8a8aa2 in wasora_step (whence=0) at
> ../wasora/src/wasora.c:406
> #13 0x0000556dbe8a81b3 in wasora_standard_run () at
> ../wasora/src/wasora.c:206
> #14 0x0000556dbe8a80a5 in main (argc=3, argv=0x7ffdf983cbd8)
>     at ../wasora/src/wasora.c:166
> (gdb)
>
>

-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener

https://www.cse.buffalo.edu/~knepley/ <http://www.cse.buffalo.edu/~knepley/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20200131/ae90daf3/attachment-0001.html>