I encountered this problem on my linux machine. Everything works fine in
debug mode; however, as soon as I switch to release mode using
-DCMAKE_BUILD_TYPE=Release, the code still compiles, but bumps into a
segmentation violation when it runs.
The following code reproduces this problem for me
(The CMakeLists.txt simply calls the DEAL_II_INVOKE_AUTOPILOT() macro):
#include <iostream>
#include <deal.II/base/mpi.h>
int main (int argc, char *argv[]) {
dealii::Utilities::MPI::MPI_InitFinalize mpi_init (argc, argv);
std::cout << "Program runs normally." << std::endl;
return 0;
}
Initially I suspected that the mpi_init object might be optimized away by
the compiler. But adding volatile to its
declaration does not fix this. Now I still do not know what is causing the
problem. Any help or suggestion is appreciated.
P.S.: I also tried running the binary in valgrind, which reports an invalid
read. The full report is as follows:
==14930== Memcheck, a memory error detector
==14930== Copyright (C) 2002-2015, and GNU GPL'd, by Julian Seward et al.
==14930== Using Valgrind-3.12.0 and LibVEX; rerun with -h for copyright info
==14930== Command: ./test.o
==14930==
==14930== Invalid read of size 8
==14930== at 0xADD6692:
tbb::internal::generic_scheduler::allocate_task(unsigned long, tbb::task*,
tbb::task_group_context*) (scheduler.cpp:315)
==14930== by 0xADD675A:
tbb::internal::generic_scheduler::generic_scheduler(tbb::internal::market&)
(scheduler.cpp:100)
==14930== by 0xADD8207: custom_scheduler (custom_scheduler.h:59)
==14930== by 0xADD8207:
tbb::internal::custom_scheduler<tbb::internal::IntelSchedulerTraits>::allocate_scheduler(tbb::internal::market&)
(custom_scheduler.h:115)
==14930== by 0xADD7051: allocate_scheduler (scheduler.cpp:42)
==14930== by 0xADD7051:
tbb::internal::generic_scheduler::create_master(tbb::internal::arena*)
(scheduler.cpp:1032)
==14930== by 0xADD1B53: tbb::internal::governor::init_scheduler(int,
unsigned long, bool) (governor.cpp:206)
==14930== by 0xADD1C0B: tbb::task_scheduler_init::initialize(int, unsigned
long) (governor.cpp:349)
==14930== by 0x97A96C2: dealii::MultithreadInfo::set_thread_limit(unsigned
int) (in
/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/dealii-develop-e5t4qkpvvpzbbomgfymanf2z5ja2yxfr/lib/libdeal_II.so.8.5.0-pre)
==14930== by 0x97A3D0B:
dealii::Utilities::MPI::MPI_InitFinalize::MPI_InitFinalize(int&, char**&,
unsigned int) (in
/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/dealii-develop-e5t4qkpvvpzbbomgfymanf2z5ja2yxfr/lib/libdeal_II.so.8.5.0-pre)
==14930== by 0x409C77: main (in /home/xywei/release_debug/build/test.o)
==14930== Address 0xfffffffffffffff7 is not stack'd, malloc'd or (recently)
free'd
==14930==
[0]PETSC ERROR:
------------------------------------------------------------------------
[0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably
memory access out of range
[0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger
[0]PETSC ERROR: or see
http://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind
[0]PETSC ERROR: or try http://valgrind.org on GNU/linux and Apple Mac OS X to
find memory corruption errors
[0]PETSC ERROR: configure using --with-debugging=yes, recompile, link, and run
[0]PETSC ERROR: to get more information on the crash.
[0]PETSC ERROR: --------------------- Error Message
--------------------------------------------------------------
[0]PETSC ERROR: Signal received
[0]PETSC ERROR: See http://www.mcs.anl.gov/petsc/documentation/faq.html for
trouble shooting.
[0]PETSC ERROR: Petsc Release Version 3.7.4, Oct, 02, 2016
[0]PETSC ERROR: ./test.o on a arch-linux2-c-opt named WXYZG_under_Arch by xywei
Fri Dec 2 19:01:12 2016
[0]PETSC ERROR: Configure options
--prefix=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/petsc-3.7.4-gdocdh6szvdge7ybgtujtibhu3xdvhc6
--with-ssl=0 --with-mpi=1
--with-mpi-dir=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/openmpi-2.0.1-tqpsaspkym4fdbxzzr2g7qfbaf52qz2u
--with-cpp=cpp --with-cxxcpp=cpp --with-precision=double
--with-scalar-type=real --with-shared-libraries=1 --with-debugging=0
--with-blas-lapack-lib=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/openblas-0.2.19-443ize6rvixorrj5ftlz4brlmju2lrfq/lib/libopenblas.so
--with-metis=1
--with-metis-dir=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/metis-5.1.0-p255v6f6j47usne5perkzwkrcqjzyllg
--with-boost=1
--with-boost-dir=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/boost-1.62.0-fmjpoz5juo2quof5vmwkxq7l64jlvam5
--with-hdf5=1
--with-hdf5-dir=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/hdf5-1.10.0-patch1-viwtpeovrofsuhsll2wccgqulpnx4zhx
--with-hypre=1
--with-hypre-dir=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/hypre-2.10.1-eyk2nctsm3lf5kzx67isilgboqtqkxgo
--with-parmetis=1
--with-parmetis-dir=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/parmetis-4.0.3-dwetxnnqiilkpubh3rwzia3224ww4cqc
--with-mumps=1
--with-mumps-dir=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/mumps-5.0.1-6mivsraqmu72gd6zadnx63526hqiaxqk
--with-scalapack=1
--with-scalapack-dir=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/netlib-scalapack-2.0.2-tieocctr7jazlz47blq5ff4gkyluer73
--with-superlu_dist-include=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/superlu-dist-5.0.0-sxkek4cbgxo7ea6imi4bfbygqidsfhen/include
--with-superlu_dist-lib=/home/xywei/spack/opt/spack/linux-arch-x86_64/gcc-6.2.1/superlu-dist-5.0.0-sxkek4cbgxo7ea6imi4bfbygqidsfhen/lib/libsuperlu_dist.a
--with-superlu_dist=1
[0]PETSC ERROR: #1 User provided function() line 0 in unknown file
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
with errorcode 59.
NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
==14930== Warning: invalid file descriptor -1 in syscall close()
==14930==
==14930== HEAP SUMMARY:
==14930== in use at exit: 4,955,503 bytes in 18,505 blocks
==14930== total heap usage: 28,634 allocs, 10,129 frees, 6,265,419 bytes
allocated
==14930==
==14930== LEAK SUMMARY:
==14930== definitely lost: 844,584 bytes in 65 blocks
==14930== indirectly lost: 0 bytes in 0 blocks
==14930== possibly lost: 704 bytes in 2 blocks
==14930== still reachable: 4,110,215 bytes in 18,438 blocks
==14930== of which reachable via heuristic:
==14930== newarray : 1,560 bytes in 3 blocks
==14930== suppressed: 0 bytes in 0 blocks
==14930== Rerun with --leak-check=full to see details of leaked memory
==14930==
==14930== For counts of detected and suppressed errors, rerun with: -v
==14930== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 0 from 0)
--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see
https://groups.google.com/d/forum/dealii?hl=en
---
You received this message because you are subscribed to the Google Groups
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
