[FLASH-USERS] Bounds-checking fails on IBM BG/L

Aaron Jackson Aaron.Jackson at stonybrook.edu
Tue Jul 28 13:19:43 EDT 2009


Hello FLASH users,

 

I'm getting an error running my simulation in which the error output says
things like:

Glibc detected

: 0xmalloc(): memory corruption0f749020: 0x ***

: 0xfree(): invalid pointer10780c70: 0x ***

: 0xdouble free or corruption (out)0e11b930: 0x ***

 

I'm attempting to compile the code with bounds-checking enabled; however, I
get the following error message in a subroutine which I'm not concerned
about.  The code will not continue compiling past this point where I need it
to analyze my subroutines.

 

/bgl/BlueLight/ppcfloor/bglsys/bin/mpixlf90 -g -C -qintsize=4 -qrealsize=8
-qfixed -qnosave -c -qsuffix=cpp=F -qarch=440 -qtune=440 -qmaxmem=131072
-qstrict -qsuffix=f=F90:cpp=F90 -qfree=f90 -WF,-DMAXBLOCKS=500 -WF,-DNXB=16
-WF,-DNYB=16 -WF,-DNZB=1 -WF,-DN_DIM=2 Grid_getBlkData.F90

"Grid_getBlkData.F90", line 483.28: 1516-152 (S) Zero-sized arrays must not
be subscripted.

1501-511  Compilation failed for file Grid_getBlkData.F90.

make: *** [Grid_getBlkData.o] Error 1

 

Does anyone know of a way to get around this or fix this problem so that I
can compile with bounds-checking?

 

I'm using the IBM BlueGene/L with the XL Fortran 90 compilers.

 

Thanks,

Aaron Jackson

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://flash.rochester.edu/pipermail/flash-users/attachments/20090728/68f91ef2/attachment.htm>


More information about the flash-users mailing list