Joost, i created #4581 and attached a patch (for the trunk) in order to solve this leak (and two similar ones)
Cheers, Gilles On 2014/04/29 5:18, VandeVondele Joost wrote: > Hi, > > I applied the patch from ticket #4569 (to 1.8.1), and things improved (in > particular the reported issue is gone). The next big leaks seems to relate to > Cartesian communicators. > > Direct leak of 9600 byte(s) in 300 object(s) allocated from: > #0 0x7f7cd2c8e793 in __interceptor_calloc > ../../../../gcc/libsanitizer/lsan/lsan_interceptors.cc:89 > #1 0x7f7cd3a92552 in mca_topo_base_cart_create > ../../../../ompi/mca/topo/base/topo_base_cart_create.c:89 > #2 0x7f7cd3a52bfd in PMPI_Cart_create > /data/vjoost/openmpi-1.8.1/build/ompi/mpi/c/profile/pcart_create.c:103 > #3 0x7f7cd3d0f4bf in ompi_cart_create_f > /data/vjoost/openmpi-1.8.1/build/ompi/mpi/fortran/mpif-h/profile/pcart_create_f.c:82 > #4 0x1bfdf6a in __message_passing_MOD_mp_cart_create > /data/vjoost/clean/cp2k/cp2k/src/common/message_passing.F:984 > > Direct leak of 21600 byte(s) in 300 object(s) allocated from: > #0 0x7f7cd2c8e3a8 in __interceptor_malloc > ../../../../gcc/libsanitizer/lsan/lsan_interceptors.cc:66 > #1 0x7f7cd3a3501f in opal_obj_new ../../opal/class/opal_object.h:467 > #2 0x7f7cd3a3501f in ompi_group_allocate ../../ompi/group/group_init.c:63 > #3 0x7f7cd3a2b192 in ompi_comm_fill_rest > ../../ompi/communicator/comm.c:1827 > #4 0x7f7cd3a2b192 in ompi_comm_enable ../../ompi/communicator/comm.c:1782 > #5 0x7f7cd3a924d2 in mca_topo_base_cart_create > ../../../../ompi/mca/topo/base/topo_base_cart_create.c:164 > #6 0x7f7cd3a52bfd in PMPI_Cart_create > /data/vjoost/openmpi-1.8.1/build/ompi/mpi/c/profile/pcart_create.c:103 > #7 0x7f7cd3d0f4bf in ompi_cart_create_f > /data/vjoost/openmpi-1.8.1/build/ompi/mpi/fortran/mpif-h/profile/pcart_create_f.c:82 > #8 0x1bfdf6a in __message_passing_MOD_mp_cart_create > /data/vjoost/clean/cp2k/cp2k/src/common/message_passing.F:984 > > Joost >
