[mpich-discuss] Errors: In direct memory block for handle type REQUEST, 2 handles are still allocated

Borrelli, Steven steven.borrelli at citi.com
Wed Oct 26 16:12:36 CDT 2011


I'm seeing the following issues when I attempt to run a program that used to run on 1.2.1p1/mpd.  What should be my next step in debugging this issue? 


Results for launching mpiexec with -n

N=1, run completes without any issues
N=2,3, run completes, but has the "In direct memory block for handle type REQUEST, 2 handles are still allocated" warning
N=4+, run fails with a signal 11. 


Platform: Linux x86_64,g++44 (RedHat 5.6, gcc44-c++-4.4.4-13.el5) 

MPICH built with the following options: ./configure --prefix=/app/mpich2-1.4.1p1-64-DEBUG/ --enable-debuginfo --enable-g=all --with-device=ch3:nemesis

--------------------------------------------------------------------------------
Run results for n = 2, 3 and 4:

# /app/mpich2-1.4.1p1-64-DEBUG/bin/mpiexec  -n 2 ./myapp 

In direct memory block for handle type REQUEST, 2 handles are still allocated
[1] 48 at [0x00002aaac0813408], ch3u_handle_recv_pkt.c[248]
[1] 8 at [0x00002aaac0005a98], ch3u_eager.c[439]



# /app/mpich2-1.4.1p1-64-DEBUG/bin/mpiexec  -n 3 ./myapp

In direct memory block for handle type REQUEST, 2 handles are still allocated
[2] 48 at [0x00002aaab4d62688], ch3u_handle_recv_pkt.c[248]
[2] 8 at [0x00002aaab40018c8], ch3u_eager.c[439]
In direct memory block for handle type REQUEST, 2 handles are still allocated
[1] 48 at [0x00002aaab1082218], ch3u_handle_recv_pkt.c[248]
[1] 8 at [0x00002aaab1083c28], ch3u_eager.c[439]



# /app/mpich2-1.4.1p1-64-DEBUG/bin/mpiexec  -n 4 ./myapp

=====================================================================================
=   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
=   EXIT CODE: 11
=   CLEANING UP REMAINING PROCESSES
=   YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
=====================================================================================
APPLICATION TERMINATED WITH THE EXIT STRING: Segmentation fault (signal 11)

--------------------------------------------------------------------------------


tail of debug.log 

0       0       2b6fc2769760[24377]     32      1.941493        bcast.c 1190    Leaving MPID_STATE_MPIR_BCAST
0       0       2b6fc2769760[24377]     32      1.941496        bcast.c 1469    Leaving MPID_STATE_MPI_BCAST
0       0       2b6fc2769760[24377]     16      1.941513        recv.c  76      Entering MPID_STATE_MPI_RECV
0       0       2b6fc2769760[24377]     16      1.941519        mpid_recv.c     29      Entering MPID_STATE_MPID_RECV
0       0       2b6fc2769760[24377]     16      1.941525        ch3u_recvq.c    273     Entering MPID_STATE_MPIDI_CH3U_RECVQ_FDU_OR_AEP
0       0       2b6fc2769760[24377]     16      1.941530        ch3_progress.c  1148    Entering MPID_STATE_MPIDI_CH3I_POSTED_RECV_ENQUEUED
0       0       2b6fc2769760[24377]     32      1.941534        ch3_progress.c  1202    Leaving MPID_STATE_MPIDI_CH3I_POSTED_RECV_ENQUEUED
0       0       2b6fc2769760[24377]     32      1.941538        ch3u_recvq.c    407     Leaving MPID_STATE_MPIDI_CH3U_RECVQ_FDU_OR_AEP
0       0       2b6fc2769760[24377]     32      1.941542        mpid_recv.c     190     Leaving MPID_STATE_MPID_RECV
0       0       2b6fc2769760[24377]     16      1.941545        ch3_progress.c  296     Entering MPID_STATE_MPIDI_CH3I_PROGRESS







More information about the mpich-discuss mailing list