[Myrinet] Cannot use totatalview debuger with mpich-gm.

??????? ?????? ??????????? <[email protected]>
Newsgroups gmane.network.myrinet.general
Organization Kvant
Message-ID <[email protected]>
Hi!

My program is aborted then I try to run it under totalview. Here goes the log
of debug session:

[maximh@M257E:~/testinteract]$ /common/mpich/bin/mpirun.ch_gm --gm-np 8 --gm-f gmpilist -tv
/home/maximh/testinteract/intertest
/usr/bin/totalviewcli
TVROOT = /common/toolworks/totalview.5.0.0-1
exectarget = /common/toolworks/totalview.5.0.0-1/linux-alpha/bin/tv5climain
Linux Alpha TotalView 5.0.0-1

Copyright 1999-2001 by Etnus, LLC. ALL RIGHTS RESERVED.
Copyright 1999 by Etnus, Inc.
Copyright 1996-1998 by Dolphin Interconnect Solutions, Inc.
Copyright 1989-1996 by BBN Inc.

Mapping 16196 bytes of ELF string data from '/home/maximh/testinteract/intertest'...done
Digesting 1021 ELF symbols from '/home/maximh/testinteract/intertest'...done
Skimming 9852 bytes of DWARF '.debug_info' symbols from '/home/maximh/testinteract/intertest'...done
Indexing 600 bytes of DWARF '.debug_frame' symbols from '/home/maximh/testinteract/intertest'...done
Digesting 4177 ECOFF external symbols...done
Digesting 307 ECOFF local symbols...done
Loading dynamic symbols for /home/maximh/testinteract/intertest
Mapping 53450 bytes of ELF string data from '/lib/libc.so.6.1'...done
Digesting 9661 ELF symbols from '/lib/libc.so.6.1'...done
Digesting 2790 ECOFF external symbols...done
Digesting 269212 ECOFF local symbols...done
Library /lib/libc.so.6.1 is loaded at 0x2000011e000
Mapping 7380 bytes of ELF string data from '/usr/lib/libots.so'...done
Digesting 1056 ELF symbols from '/usr/lib/libots.so'...done
Digesting 98 ECOFF external symbols...done
Digesting 1030 ECOFF local symbols...done
Library /usr/lib/libots.so is loaded at 0x20000354000
Mapping 5768 bytes of ELF string data from '/lib/ld-linux.so.2'...done
Digesting 1234 ELF symbols from '/lib/ld-linux.so.2'...done
Digesting 269 ECOFF external symbols...done
Digesting 18624 ECOFF local symbols...done
Library /lib/ld-linux.so.2 is loaded at 0x20000000000
Thread 1.1 has appeared
Process 1 has exited

{ Here I made quite a long pause (about 30 sec). I didn't entered any TotalView
  command (like dstep, dcont ..) and after that period of time i've got the following
  messages. When they appears, debugging is not possible anymore.
  NOTE: If I execute dcont command right after the debugging session start
  everything is Ok and debugging works fine }

d1.<> WARNING: libgm/gm_send.c:176:???():userland:
send failed: target port was closed
_gm_sent: gmID=336 'M258E' msg_cnt=1 send failed to complete (see kernel log for details)WARNING:
libgm/gm_send.c:176:???():userland:
send failed: target port was closed
_gm_sent: gmID=336 'M258E' msg_cnt=1 send failed to complete (see kernel log for details)WARNING:
libgm/gm_send.c:176:???():userland:
WARNING: libgm/gm_send.c:176:???():userland:
send failed: target port was closed
send failed: target port was closed
_gm_sent: gmID=337 'M259E' msg_cnt=1 send failed to complete (see kernel log for details)_gm_sent: gmID=337 'M259E'
msg_cnt=1 send failed to complete (see kernel log for details)WARNING: libgm/gm_send.c:176:???():userland:
send failed: target port was closed
WARNING: libgm/gm_send.c:176:???():userland:
send failed: target port was closed
_gm_sent: gmID=338 'M260E' msg_cnt=1 send failed to complete (see kernel log for details)_gm_sent: gmID=338 'M260E'
msg_cnt=1 send failed to complete (see kernel log for details)
d1.<> 
d1.<> interacttest:Node noini[0/0] Receive SIGINT signal 2


{ Here I pressed Ctrl-C. }
 
Anyone has some ideas ?

 Kirill
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.