Fwd: Problems with building NCL using the Intel compiler suite

From: Mary Haley <haley_at_nyahnyahspammersnyahnyah>
Date: Thu Jun 30 2011 - 12:49:34 MDT

Hi David,

Wei Huang, one of our developers, said that there are only a few places where "compound" appears as a variable name, so if you could try changing these by hand to "nclcompound" and recompiling, it might work.

Here's Wei's email that includes the location of the changes and which files to change. Please let us know if you have any success with this.

--Mary

>
> holywell:/Users/huangwei/ncl/src/trunk/ni/src/ncl 720 > svn diff NclDataDefs.h
> Index: NclDataDefs.h
> ===================================================================
> --- NclDataDefs.h (revision 12429)
> +++ NclDataDefs.h (working copy)
> @@ -119,7 +119,7 @@
> typedef int obj;
> typedef int list;
> typedef int group;
> -typedef int compound;
> +typedef int nclcompound;
>
> typedef union _NclScalar {
> double doubleval;
> @@ -140,7 +140,7 @@
> logical logicalval;
> obj objval;
> group groupval;
> - compound compoundval;
> + nclcompound compoundval;
> }NclScalar;
>
> typedef struct _NclRefList{
> holywell:/Users/huangwei/ncl/src/trunk/ni/src/ncl 721 > svn diff DataSupport.c
> Index: DataSupport.c
> ===================================================================
> --- DataSupport.c (revision 12429)
> +++ DataSupport.c (working copy)
> @@ -820,7 +820,7 @@
> case NCL_group:
> return(sizeof(group));
> case NCL_compound:
> - return(sizeof(compound));
> + return(sizeof(nclcompound));
> case NCL_logical:
> return(sizeof(int));
> case NCL_list:
>
>
>>>
>>>
>>> Begin forwarded message:
>>>
>>>> From: David Mechem <dmechem@ku.edu>
>>>> Date: June 27, 2011 9:50:32 AM MDT
>>>> To: <ncl-install@ucar.edu>
>>>> Subject: Problems with building NCL using the Intel compiler suite
>>>>
>>>> Greetings,
>>>>
>>>> I'm having issues building NCL from source using the Intel compiler
>>>> suite. I'm building a vanilla version without OPeNDAP and netCDF-4
>>>> support. The NCARGraphics libraries and many of the utilities seem to be
>>>> building OK, but NCL does not. I'm using version 11.0 of ifort and icc,
>>>> 64-bit CENTos kernel 2.6.18-194.17.1.el5.
>>>>
>>>> I'm getting compilation errors on three files. The errors are all of a
>>>> similar type and appear to be related to the variable "compound" being
>>>> multiply defined.
>>>>
>>>> 1.
>>>> -----------------
>>>> make[5]: Entering directory `/usr/local/ncl_ncarg-6.0.0/ni/src/lib/nfp'
>>>> icc -ansi -O2 -I../../ncl -I../../../.././include
>>>> -I/usr/include/X11 -DSYS
>>>> V -D_POSIX_SOURCE -D_XOPEN_SOURCE -DByteSwapped -DNeedFuncProto
>>>> -DNDEBUG -c -o
>>>> areavolW.o areavolW.c
>>>> ../../ncl/NclDataDefs.h(122): error: "compound" has already been
>>>> declared in the
>>>> current scope
>>>> typedef int compound;
>>>> ^
>>>>
>>>> areavolW.c(258): warning #188: enumerated type mixed with another type
>>>> return(ret);
>>>> ^
>>>>
>>>> areavolW.c(447): warning #188: enumerated type mixed with another type
>>>> return(ret);
>>>> ^
>>>>
>>>> areavolW.c(636): warning #188: enumerated type mixed with another type
>>>> return(ret);
>>>> ^
>>>>
>>>> areavolW.c(903): warning #188: enumerated type mixed with another type
>>>> return(ret);
>>>> ^
>>>>
>>>> areavolW.c(1141): warning #188: enumerated type mixed with another type
>>>> return(ret);
>>>> ^
>>>>
>>>> areavolW.c(1379): warning #188: enumerated type mixed with another type
>>>> return(ret);
>>>> ^
>>>>
>>>> areavolW.c(1643): warning #188: enumerated type mixed with another type
>>>> return(ret);
>>>> ^
>>>>
>>>> areavolW.c(1927): warning #188: enumerated type mixed with another type
>>>> return(ret);
>>>> ^
>>>>
>>>> areavolW.c(2262): warning #188: enumerated type mixed with another type
>>>> return(ret);
>>>> ^
>>>>
>>>> compilation aborted for areavolW.c (code 2)
>>>> make[5]: *** [areavolW.o] Error 2
>>>> -------------------------------
>>>>
>>>> 2.
>>>> ------------------------------
>>>> make[5]: Entering directory `/usr/local/ncl_ncarg-6.0.0/ni/src/lib/nio'
>>>> icc -ansi -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -O2
>>>> -I../../../.././include -I/usr/include/X11 -DLinux -DMAJOR=2
>>>> -DNIO_LIB_ONLY -DSYSV -D_POSIX_SOURCE -D_XOPEN_SOURCE -DByteSwapped
>>>> -DNeedFuncProto -DNDEBUG -c -o NclTypedouble.o NclTypedouble.c
>>>> /usr/local/intel/11.0/084/include/math.h(796): error: "compound" has
>>>> already been declared in the current scope
>>>> _LIBIMF_EXT _LIBIMF_DOUBLE _LIBIMF_PUBAPI compound( _LIBIMF_DOUBLE
>>>> __x, _LIBIMF_DOUBLE __y );
>>>> ^
>>>>
>>>> compilation aborted for NclTypedouble.c (code 2)
>>>> make[5]: *** [NclTypedouble.o] Error 2
>>>> ---------------------------------
>>>>
>>>> 3.
>>>> -----------------------------
>>>> icc -ansi -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
>>>> -I/usr/local/include -O2 -I../../.././include -I/usr/include/X11
>>>> -DLinux -DMAJOR=2 -DSYSV -D_POSIX_SOURCE -D_XOPEN_SOURCE
>>>> -DByteSwapped -DNeedFuncProto -DNDEBUG -c -o MathFuncs.o MathFuncs.c
>>>> NclDataDefs.h(122): error: "compound" has already been declared in the
>>>> current scope
>>>> typedef int compound;
>>>> ^
>>>>
>>>> compilation aborted for MathFuncs.c (code 2)
>>>> make[4]: *** [MathFuncs.o] Error 2
>>>> -----------------------------
>>>>
>>>>
>>>> Any advice will be appreciated!
>>>>
>>>> Dave
>>>>
>>>> --
>>>> --------------------------------------
>>>> David B. Mechem dmechem@ku.edu
>>>> Assistant Professor
>>>> Atmospheric Science Program
>>>> Department of Geography
>>>> University of Kansas
>>>> ph: (785) 864-5707 fax: (785) 864-5378
>>>> --------------------------------------
>>>>
>>>>
>>>> _______________________________________________
>>>> ncl-install mailing list
>>>> List instructions, subscriber options, unsubscribe:
>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-install
>>>
>>
>

_______________________________________________
ncl-install mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-install
Received on Thu Jun 30 12:49:40 2011

This archive was generated by hypermail 2.1.8 : Thu Jun 30 2011 - 12:54:57 MDT