[mpich2-dev] (1.1b1) Fortran attributes

Bob Cernohous bobc at us.ibm.com
Mon Apr 13 09:17:53 CDT 2009


We have a problem with fortran attributes in the beta.  I believe it 
passes the address of a void* with MPIR_ATTR_INT.  So when casting back, 
shouldn't it be using a pointer sized integer (MPIR_Pint) not MPI_Aint? 
This fixes BG/P.

---
 mpich2/src/mpi/attr/comm_get_attr.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/mpich2/src/mpi/attr/comm_get_attr.c 
b/mpich2/src/mpi/attr/comm_get_attr.c
index 8e996e3..b3253c6 100644
--- a/mpich2/src/mpi/attr/comm_get_attr.c
+++ b/mpich2/src/mpi/attr/comm_get_attr.c
@@ -264,7 +264,7 @@ int MPIR_CommGetAttr( MPI_Comm comm, int comm_keyval, 
void *attribute_val,
                     if (outAttrType == MPIR_ATTR_AINT)
                                 *(MPI_Aint*)attr_val_p = 
*(MPI_Aint*)*(void **)attr_val_p;
                     else if (outAttrType == MPIR_ATTR_INT)
-                                *(MPI_Aint*)attr_val_p = *(int *)*(void 
**)attr_val_p;
+                                *(MPIR_Pint*)attr_val_p = *(int *)*(void 
**)attr_val_p;
                 }
     }
     else {
-- 
1.5.3.7

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

Bob Cernohous:  (T/L 553) 507-253-6093

BobC at us.ibm.com
IBM Rochester, Building 030-2(C335), Department 61L
3605 Hwy 52 North, Rochester,  MN 55901-7829

> Chaos reigns within.
> Reflect, repent, and reboot.
> Order shall return.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mcs.anl.gov/mailman/private/mpich2-dev/attachments/20090413/84b5f545/attachment.htm>


More information about the mpich2-dev mailing list