[mpich2-dev] about Ticket #337 increment MAX_FINALIZE_FUNC

Lisandro Dalcin dalcinl at gmail.com
Tue Dec 23 08:44:39 CST 2008


I'm very sorry... I posted this to the wrong address :-(

---------- Forwarded message ----------
From: Lisandro Dalcin <dalcinl at gmail.com>
Date: Tue, Dec 23, 2008 at 12:42 PM
Subject: about Ticket #337 increment MAX_FINALIZE_FUNC
To: mpich2-maint <mpich2-maint at mcs.anl.gov>


Rajeev, look at this, I hope you will get convinced ;-)

I've built mpich2-dev like this:

$ mpich2version
MPICH2 Version:         1.1a3
MPICH2 Release date:    Unknown, built on Tue Dec 23 10:33:28 ARST 2008
MPICH2 Device:          ch3:nemesis
MPICH2 configure:       --prefix=/usr/local/mpich2/dev
--enable-sharedlibs=gcc --enable-g=all
MPICH2 CC:      gcc  -g -O2
MPICH2 CXX:     c++  -g -O2
MPICH2 F77:     gfortran  -g -O2
MPICH2 F90:     f95  -g -O2


As you can see, I'm using '--enable-g=all'. Then this surely #define's
MPICH_DEBUG_HANDLEALLOC. Then the routine MPIU_CheckHandlesOnFinalize
is being registered many times... To confirm this, see the attached
file (obtained by running the mpi4py testsuite using the 'ddd'
front-end to gdb)

Regards,

--
Lisandro Dalcín
---------------
Centro Internacional de Métodos Computacionales en Ingeniería (CIMEC)
Instituto de Desarrollo Tecnológico para la Industria Química (INTEC)
Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
PTLC - Güemes 3450, (3000) Santa Fe, Argentina
Tel/Fax: +54-(0)342-451.1594



-- 
Lisandro Dalcín
---------------
Centro Internacional de Métodos Computacionales en Ingeniería (CIMEC)
Instituto de Desarrollo Tecnológico para la Industria Química (INTEC)
Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
PTLC - Güemes 3450, (3000) Santa Fe, Argentina
Tel/Fax: +54-(0)342-451.1594
-------------- next part --------------
A non-text attachment was scrubbed...
Name: trace.log
Type: application/octet-stream
Size: 3050 bytes
Desc: not available
URL: <https://lists.mcs.anl.gov/mailman/private/mpich2-dev/attachments/20081223/fdbb321b/attachment.obj>


More information about the mpich2-dev mailing list