-
Notifications
You must be signed in to change notification settings - Fork 930
Closed
Labels
Description
This problem is in master only and I tracked it down to this commit using git bisect.
[rvandevaart@drossetti-ivy5 ompi-master-rolfv]$ git bisect good
Bisecting: 0 revisions left to test after this (roughly 1 step)
[720fa860eee7df4f06eb5604058da57453e88b46] changing the destruct function of list release API to release list items caused a bug in oshmem application. Fixing the bug with this patch
[rvandevaart@drossetti-ivy5 ompi-master-rolfv]$
Although I cannot quite figure out what that commit is when I look at the log. In any event, the errors can be seen on many different buffered send tests.
[rvandevaart@drossetti-ivy4 src]$ mpirun -np 2 MPI_Bsend_ator_c
--------------------------------------------------------------------------
MPITEST info (0): Starting MPI_Bsend_ator: All Bsend TO Root test
MPITEST info (0): Node spec MPITEST_comm_sizes[6]=2 too large, using 1
MPITEST info (0): Node spec MPITEST_comm_sizes[22]=2 too large, using 1
MPITEST info (0): Node spec MPITEST_comm_sizes[32]=2 too large, using 1
--------------------------------------------------------------------------
mpirun noticed that process rank 1 with PID 0 on node drossetti-ivy4 exited on signal 11 (Segmentation fault).
--------------------------------------------------------------------------
(gdb) where
#0 0x00007f16a17ef1d0 in ?? ()
#1 <signal handler called>
#2 0x00007f16a81e3abb in opal_memory_ptmalloc2_int_free (av=0x7f16a843f3a0, mem=0xb8e500)
at ../../../../../opal/mca/memory/linux/malloc.c:4377
#3 0x00007f16a81e2317 in opal_memory_ptmalloc2_free (mem=0xb8e500) at ../../../../../opal/mca/memory/linux/malloc.c:3513
#4 0x00007f16a81e1051 in opal_memory_linux_free_hook (__ptr=0xb8e500, caller=0x7f16a8132687)
at ../../../../../opal/mca/memory/linux/hooks.c:708
#5 0x00007f16a8132687 in opal_list_destruct (list=0xb8d250) at ../../opal/class/opal_list.c:108
#6 0x00007f16a4e6eeeb in opal_obj_run_destructors (object=0xb8d250) at ../../../../../opal/class/opal_object.h:460
#7 0x00007f16a4e6ff44 in mca_allocator_basic_finalize (base=0xb8d210)
at ../../../../../opal/mca/allocator/basic/allocator_basic.c:362
#8 0x00007f16a8dd9fc4 in mca_pml_base_bsend_detach (addr=0x7fffa44e12c8, size=0x7fffa44e12c4)
at ../../../../ompi/mca/pml/base/pml_base_bsend.c:190
#9 0x00007f16a8d52001 in PMPI_Buffer_detach (buffer=0x7fffa44e12c8, size=0x7fffa44e12c4) at pbuffer_detach.c:52
#10 0x0000000000402b76 in main (argc=1, argv=0x7fffa44e1668) at MPI_Bsend_ator_c.c:625
(gdb)