Re: [OMPI users] MPI Error

2008-12-21 Thread Asad Ali
Hi Jeff,

I have changed the position of malloc.h in header files  list. I moved it up
above mpi.h. Now I am getting a different error message see following,
[asad@stat74 T]$ mpirun  --np 4 nice -10 ./lisa09EMRIT-P
 +---[ lisa14.c ]---
 | This is process #0 of 4
 | initialising  :  21.12.2008  21:59:28
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
-+--> INJECTING SIGNAL...
Allocating 134217728 bytes for read buffer...
-+--> INJECTING SIGNAL...
-+--> INJECTING SIGNAL...
-+--> INJECTING SIGNAL...
524765
524765
524765
524765
 | initialising `TDI framework' (might take several minutes)...
Signal:6 info.si_errno:0(Success) si_code:-6()
[0] func:/usr/lib/openmpi/libopal.so.0 [0x46efd6]
[1] func:[0x110440]
[2] func:[0x110402]
[3] func:/lib/libc.so.6(gsignal+0x50) [0x2c1fa0]
[4] func:/lib/libc.so.6(abort+0x101) [0x2c38b1]
[5] func:/usr/lib/openmpi/libopal.so.0 [0x473d34]
[6] func:/usr/lib/openmpi/libopal.so.0(free+0x36) [0x473786]
[7] func:./lisa09EMRIT-P(main+0x454) [0x80566f4]
[8] func:/lib/libc.so.6(__libc_start_main+0xe0) [0x2aef70]
[9] func:./lisa09EMRIT-P [0x804af21]
*** End of error message ***

Its good that now no user defined functions in code are involved in error
(it was giving error messages for some of my defined functions too).
malloc.h has also been disappeared in this error message as was in previous
message.

I have used valgrind to see where the error is but I am unable to interpret
the following valgrind output error message.


--16515--object doesn't have a symbol table
--16515-- Reading syms from /usr/lib/openmpi/openmpi/mca_osc_pt2pt.so
(0x4114000)
--16515--object doesn't have a symbol table
==16513== Warning: set address range perms: large range 536870912 (defined)
==16512== Warning: set address range perms: large range 536870912 (defined)
==16514== Warning: set address range perms: large range 536870912 (defined)
==16515== Warning: set address range perms: large range 536870912 (defined)
 +---[ lisa14.c ]---
 | This is process #0 of 4
 | initialising  :  21.12.2008  21:15:37
Allocating 134217728 bytes for read buffer...
==16515== Warning: set address range perms: large range 134221824 (defined)
Allocating 134217728 bytes for read buffer...
==16514== Warning: set address range perms: large range 134221824 (defined)
Allocating 134217728 bytes for read buffer...
==16513== Warning: set address range perms: large range 134221824 (defined)
Allocating 134217728 bytes for read buffer...
==16512== Warning: set address range perms: large range 134221824 (defined)
==16515== Warning: set address range perms: large range 134217728 (defined)
==16514== Warning: set address range perms: large range 134217728 (defined)
==16513== Warning: set address range perms: large range 134217728 (defined)
==16512== Warning: set address range perms: large range 134217728 (defined)
==16515== Warning: set address range perms: large range 134221824 (noaccess)
==16514== Warning: set address range perms: large range 134221824 (noaccess)
==16513== Warning: set address range perms: large range 134221824 (noaccess)
==16512== Warning: set address range perms: large range 134221824 (noaccess)
--16515-- memcheck GC: 1024 nodes, 1024 survivors (100.0%)
--16515-- memcheck GC: increase table size to 2048
--16513-- memcheck GC: 1024 nodes, 1024 survivors (100.0%)
--16513-- memcheck GC: increase table size to 2048
--16514-- memcheck GC: 1024 nodes, 1024 survivors (100.0%)
--16514-- memcheck GC: increase table size to 2048
--16512-- memcheck GC: 1024 nodes, 1024 survivors (100.0%)
--16512-- memcheck GC: increase table size to 2048
--16513-- memcheck GC: 2048 nodes, 2048 survivors (100.0%)
--16513-- memcheck GC: increase table size to 4096
--16514-- memcheck GC: 2048 nodes, 2048 survivors (100.0%)
--16514-- memcheck GC: increase table size to 4096
--16515-- memcheck GC: 2048 nodes, 2048 survivors (100.0%)
--16515-- memcheck GC: increase table size to 4096
--16512-- memcheck GC: 2048 nodes, 2048 survivors (100.0%)
--16512-- memcheck GC: increase table size to 4096
Allocating 134217728 bytes for read buffer...
==16515== Warning: set address range perms: large range 134221824 (defined)
Allocating 134217728 bytes for read buffer...
==16513== Warning: set address range perms: large range 134221824 (defined)
Allocating 134217728 bytes for read buffer...
==16514== Warning: set address range perms: large range 134221824 (defined)
==16515== Warning: set address range perms: large range 134217728 (defined)
==16513== Warning: set address range perms: large range 134217728 (defined)
==16514== Warning: set address range perms: large range 134217728 (defined)
==16515== Warning:

Re: [OMPI users] MPI Error

2008-12-21 Thread Jeff Squyres
It's not the valgrind that are telling here -- it's the fact that  
something is saying "INJECTING SIGNAL..."


It looks like something in your program is calling abort() (signal 6  
is the abort signal).



On Dec 21, 2008, at 4:08 AM, Asad Ali wrote:


--===1267833741==
Content-Type: multipart/alternative;
boundary="=_Part_48219_1955003.1229850505418"

--=_Part_48219_1955003.1229850505418
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

Hi Jeff,

I have changed the position of malloc.h in header files  list. I  
moved it up

above mpi.h. Now I am getting a different error message see following,
[asad@stat74 T]$ mpirun  --np 4 nice -10 ./lisa09EMRIT-P
+---[ lisa14.c ]---
| This is process #0 of 4
| initialising  :  21.12.2008  21:59:28
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
Allocating 134217728 bytes for read buffer...
-+--> INJECTING SIGNAL...
Allocating 134217728 bytes for read buffer...
-+--> INJECTING SIGNAL...
-+--> INJECTING SIGNAL...
-+--> INJECTING SIGNAL...
524765
524765
524765
524765
| initialising `TDI framework' (might take several minutes)...
Signal:6 info.si_errno:0(Success) si_code:-6()
[0] func:/usr/lib/openmpi/libopal.so.0 [0x46efd6]
[1] func:[0x110440]
[2] func:[0x110402]
[3] func:/lib/libc.so.6(gsignal+0x50) [0x2c1fa0]
[4] func:/lib/libc.so.6(abort+0x101) [0x2c38b1]
[5] func:/usr/lib/openmpi/libopal.so.0 [0x473d34]
[6] func:/usr/lib/openmpi/libopal.so.0(free+0x36) [0x473786]
[7] func:./lisa09EMRIT-P(main+0x454) [0x80566f4]
[8] func:/lib/libc.so.6(__libc_start_main+0xe0) [0x2aef70]
[9] func:./lisa09EMRIT-P [0x804af21]
*** End of error message ***

Its good that now no user defined functions in code are involved in  
error

(it was giving error messages for some of my defined functions too).
malloc.h has also been disappeared in this error message as was in  
previous

message.

I have used valgrind to see where the error is but I am unable to  
interpret

the following valgrind output error message.


--16515--object doesn't have a symbol table
--16515-- Reading syms from /usr/lib/openmpi/openmpi/mca_osc_pt2pt.so
(0x4114000)
--16515--object doesn't have a symbol table
==16513== Warning: set address range perms: large range 536870912  
(defined)
==16512== Warning: set address range perms: large range 536870912  
(defined)
==16514== Warning: set address range perms: large range 536870912  
(defined)
==16515== Warning: set address range perms: large range 536870912  
(defined)

+---[ lisa14.c ]---
| This is process #0 of 4
| initialising  :  21.12.2008  21:15:37
Allocating 134217728 bytes for read buffer...
==16515== Warning: set address range perms: large range 134221824  
(defined)

Allocating 134217728 bytes for read buffer...
==16514== Warning: set address range perms: large range 134221824  
(defined)

Allocating 134217728 bytes for read buffer...
==16513== Warning: set address range perms: large range 134221824  
(defined)

Allocating 134217728 bytes for read buffer...
==16512== Warning: set address range perms: large range 134221824  
(defined)
==16515== Warning: set address range perms: large range 134217728  
(defined)
==16514== Warning: set address range perms: large range 134217728  
(defined)
==16513== Warning: set address range perms: large range 134217728  
(defined)
==16512== Warning: set address range perms: large range 134217728  
(defined)
==16515== Warning: set address range perms: large range 134221824  
(noaccess)
==16514== Warning: set address range perms: large range 134221824  
(noaccess)
==16513== Warning: set address range perms: large range 134221824  
(noaccess)
==16512== Warning: set address range perms: large range 134221824  
(noaccess)

--16515-- memcheck GC: 1024 nodes, 1024 survivors (100.0%)
--16515-- memcheck GC: increase table size to 2048
--16513-- memcheck GC: 1024 nodes, 1024 survivors (100.0%)
--16513-- memcheck GC: increase table size to 2048
--16514-- memcheck GC: 1024 nodes, 1024 survivors (100.0%)
--16514-- memcheck GC: increase table size to 2048
--16512-- memcheck GC: 1024 nodes, 1024 survivors (100.0%)
--16512-- memcheck GC: increase table size to 2048
--16513-- memcheck GC: 2048 nodes, 2048 survivors (100.0%)
--16513-- memcheck GC: increase table size to 4096
--16514-- memcheck GC: 2048 nodes, 2048 survivors (100.0%)
--16514-- memcheck GC: increase table size to 4096
--16515-- memcheck GC: 2048 nodes, 2048 survivors (100.0%)
--16515-- memcheck GC: increase table size to 4096
--16512-- memcheck GC: 2048 nodes, 2048 survivors (100.0%)
--16512-- memcheck GC: increase table size to 4096
Allocating 134217728 bytes for read buffer

Re: [OMPI users] Bug in 1.3 nightly

2008-12-21 Thread Jeff Squyres
Can you provide any further details?  See http://www.open-mpi.org/community/help/ 
. 


On Dec 16, 2008, at 3:27 AM, Gabriele Fatigati wrote:


Dear OpenMPI developers,
trying to compile 1.3 nightly version , i get the follow error:

../../../orte/.libs/libopen-rte.so: undefined reference to  
`ORTE_NAME_PRINT'
../../../orte/.libs/libopen-rte.so: undefined reference to  
`ORTE_JOBID_PRINT'



The version affected are:

openmpi-1.3rc3r20130
openmpi-1.3rc3r20107
openmpi-1.3rc3r20092
openmpi-1.3rc2r20084

Thanks in advance.


--
Ing. Gabriele Fatigati

Parallel programmer

CINECA Systems & Tecnologies Department

Supercomputing Group

Via Magnanelli 6/3, Casalecchio di Reno (BO) Italy

www.cineca.itTel:   +39 051 6171722

g.fatig...@cineca.it
___
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users



--
Jeff Squyres
Cisco Systems



[OMPI users] Mi Yan is out of the office.

2008-12-21 Thread Mi Yan

I will be out of the office starting  12/21/2008 and will not return until
01/02/2009.