[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

970520: building on a Paragon

Wilbur,

Is "Engineering Manufacturing Mechanics (9117)" part of Sandia National
Laboratory?

> To: address@hidden
> From: "Wilbur R. Johnson" <address@hidden>
> Subject: Problems with 3.3 compiling
> Organization: Engineering Manufacturing Mechanics (9117)
> Keywords: 199705191832.MAA14512

In the above message, you wrote:

> > I'm trying to compile on a variant of an intel paragon. Basicly it's
> > a paragon with Pentium Pros.... Lots of them.
> > 
> > I'm using gcc instead of cicc because configure core dumps with cicc.

Cute.

> > 
> > I probably could coerce this to go, libnetcdf.a even gets built. But
> > I would like to have this fire up right out of the ditribution.

I'd like to see that happen too.

> > 
> > Any help would be apprieciated....

Russ wrote that I might ask you to try netCDF version 3.3.1, which was
just released.  Well, I'm asking.  The build mechanism in 3.3.1 should
have fewer bugs than 3.3 and also be more robust in its testing.

> > Script started on Mon May 19 11:58:53 1997
> > janus 101 > uname -a
> > Paragon OSF/1 janus 1.0.4 TF97_WW17 i386

According to the order of the output from the above "uname -a":

    Operating System (uname -s):        Paragon
    Hostname (uname -n):                OSF/1
    Release (uname -r):                 janus
    Version (uname -v):                 1.0.4
    Machine (uname -m):                 TF97_WW17

I suspect that the uname(1) utility is swapping the hostname and
release.  Is this true?

This shouldn't cause problems, however, with building netCDF 3.3.1.

> > janus[src][255]_cat VERSION 
> > 3.3
> > janus 102 > ./configure --prefix=/Net/usr/home/wrjohns
> > creating cache ./config.cache
> > checking for m4... m4
> > checking user-defined C compiler "gcc"
> > checking C compiler... works
> > checking for how to make dependencies... false
> > checking for CC... CC
> > checking C++ compiler "CC"... works
> > checking for xlf... no
> > checking for fort77... no
> > checking for ghf77... no
> > checking for f77... f77
> > checking for math library
> > checking for -lc... no
> > checking for -lm... yes
> > checking FORTRAN-type equivalent to netCDF "byte"... byte
> > checking FORTRAN-type equivalent to C "signed char"... byte
> > checking FORTRAN fill-value for netCDF version 2 byte... -127
> > checking FORTRAN-type equivalent to netCDF "short"... integer*2
> > checking FORTRAN-type equivalent to C "short"... integer*2
> > checking FORTRAN-type equivalent to netCDF "long"... integer*4
> > checking FORTRAN-type equivalent to C "int"... integer
> > checking FORTRAN-type equivalent to netCDF "float"... real*4
> > checking FORTRAN-type equivalent to C "float"... no equivalent type

Interesting.  The f77(1) compiler supports a "real*4" datatype but
doesn't have a datatype that's equivalent to a C "float".  Is this true?

Version 3.3.1 has considerably better checks on Fortran/C type
equivalency, BTW.

> > checking FORTRAN-type equivalent to netCDF "double"... real*8
> > checking FORTRAN-type equivalent to C "double"... real*8
> > checking for ar... ar
> > checking for ranlib... ranlib
> > checking how to run the C preprocessor... gcc -E
> > checking for stdlib.h... yes
> > checking for sys/types.h... yes
> > checking for strerror... yes
> > checking for ftruncate... yes
> > checking for st_blksize in struct stat... yes
> > checking whether cross-compiling... no
> > checking for IEEE floating point format... yes
> > checking for ANSI C header files... yes
> > checking for size_t... yes
> > checking for off_t... yes
> > checking for ssize_t... yes
> > checking for ptrdiff_t... yes
> > checking for uchar... yes
> > checking whether char is unsigned... no
> > checking whether byte ordering is bigendian... no
> > checking size of short... 2
> > checking size of int... 4
> > checking size of long... 4
> > checking size of float... 4
> > checking size of double... 8
> > checking size of off_t... 4
> > checking size of size_t... 4
> > checking for command to create manual-page index... 
> > checking binary distribution directory... /home/ftp/pub/binary/system: not 
> > found

I'm not surprised that the configure script didn't find the system(1)
command (it's a private one of ours).  Version 3.3.1 doesn't print out
the error message, however.

> > dummy_system
> > updating cache ./config.cache
> > creating ./config.status
> > creating macros.make
> > sed: command garbled: s%@FTPBINDIR@%/home/ftp/pub/binary/system: not found

Hmm...  I wonder if the failure of the sed(1) command caused the
problems in your build.

> > creating libsrc/ncconfig.h
> > creating fortran/netcdf-cpp.inc
> > janus 104 > 
> > janus 104 > 
> > janus 104 > cat config.log
> > This file contains any messages produced by compilers while
> > running configure, to aid debugging if configure makes a mistake.
> > 
> > configure:989: gcc -c -O -DNDEBUG conftest.c 1>&5
> > configure:1117: CC -c  -DNDEBUG conftest.C 1>&5
> > conftest.C:
> > configure:1272: f77 -c -O  conftest.f
> > configure:1313: f77 -c -O  conftest.F
> > configure:1345: gcc -o conftest -O -DNDEBUG  conftest.c -lc   1>&5
> > collect2: ld returned 1 exit status
> > /tmp/cc7359671.o: In function `t':
> > /tmp/cc7359671.o(.text+0x14): undefined reference to `tanh'
> > configure:1379: gcc -o conftest -O -DNDEBUG  conftest.c -lm   1>&5
> > configure:1421: f77 -c -O  conftest.f
> > configure:1448: f77 -c -O  conftest.f
> > configure:1465: gcc -o conftest -DNDEBUG -O  conftest.c conftestf.o 
> > configure:1475: ./conftest
> > configure:1512: f77 -c -O  conftest.f
> > configure:1539: f77 -c -O  conftest.f
> > configure:1556: gcc -o conftest -DNDEBUG -O  conftest.c conftestf.o 
> > configure:1566: ./conftest
> > configure:1591: f77 -c -O  conftest.f
> > configure:1618: f77 -c -O  conftest.f
> > configure:1635: gcc -o conftest -DNDEBUG -O  conftest.c conftestf.o 
> > configure:1645: ./conftest
> > configure:1670: f77 -c -O  conftest.f
> > configure:1697: f77 -c -O  conftest.f
> > configure:1714: gcc -o conftest -DNDEBUG -O  conftest.c conftestf.o 
> > configure:1724: ./conftest
> > configure:1749: f77 -c -O  conftest.f
> > configure:1776: f77 -c -O  conftest.f
> > configure:1793: gcc -o conftest -DNDEBUG -O  conftest.c conftestf.o 
> > configure:1803: ./conftest
> > configure:1898: gcc -E -DNDEBUG conftest.c >/dev/null 2>conftest.out
> > configure:1944: gcc -E -DNDEBUG conftest.c >/dev/null 2>conftest.out
> > configure:1978: gcc -E -DNDEBUG conftest.c >/dev/null 2>conftest.out
> > configure:2028: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > configure:2076: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > configure:2110: gcc -c -O -DNDEBUG conftest.c 1>&5
> > configure:2142: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > configure:2194: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > configure:2223: gcc -E -DNDEBUG conftest.c >/dev/null 2>conftest.out
> > configure:2288: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > configure:2544: gcc -c -O -DNDEBUG conftest.c 1>&5
> > configure:2560: gcc -c -O -DNDEBUG conftest.c 1>&5
> > configure: In function `t':
> > configure:2555: `not' undeclared (first use this function)
> > configure:2555: (Each undeclared identifier is reported only once
> > configure:2555: for each function it appears in.)
> > configure:2555: parse error before `big'
> > configure:2628: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > configure:2662: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > configure:2696: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > configure:2730: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > configure:2764: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > configure:2802: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > configure:2840: gcc -o conftest -O -DNDEBUG  conftest.c  1>&5
> > janus 104 > 
> > janus 104 > 

I didn't see anything Earth shattering in the config.log file.

> > janus 101 > gmake
> > Makefile:190: warning: overriding commands for target `/netcdf-.tar.Z'
> > Makefile:176: warning: ignoring old commands for target `/netcdf-.tar.Z'
> > rules.make:116: warning: overriding commands for target `/'
> > rules.make:114: warning: ignoring old commands for target `/'
> > rules.make:118: warning: overriding commands for target `/'
> > rules.make:116: warning: ignoring old commands for target `/'
> > rules.make:121: warning: overriding commands for target `/'
> > rules.make:118: warning: ignoring old commands for target `/'
> > rules.make:124: warning: overriding commands for target `/'
> > rules.make:121: warning: ignoring old commands for target `/'
> > rules.make:138: warning: overriding commands for target `/man1'
> > rules.make:135: warning: ignoring old commands for target `/man1'
> > rules.make:140: warning: overriding commands for target `/man3'
> > rules.make:135: warning: ignoring old commands for target `/man3'
> > rules.make:143: warning: overriding commands for target `/man3f'
> > rules.make:135: warning: ignoring old commands for target `/man3f'
> > rules.make:148: warning: overriding commands for target `/'
> > rules.make:124: warning: ignoring old commands for target `/'
> > gmake[1]: Entering directory `/Net/usr/home/wrjohns/src/netcdf-3.3/src'
> > Makefile:190: warning: overriding commands for target `/netcdf-.tar.Z'
> > Makefile:176: warning: ignoring old commands for target `/netcdf-.tar.Z'
> > rules.make:116: warning: overriding commands for target `/'
> > rules.make:114: warning: ignoring old commands for target `/'
> > rules.make:118: warning: overriding commands for target `/'
> > rules.make:116: warning: ignoring old commands for target `/'
> > rules.make:121: warning: overriding commands for target `/'
> > rules.make:118: warning: ignoring old commands for target `/'
> > rules.make:124: warning: overriding commands for target `/'
> > rules.make:121: warning: ignoring old commands for target `/'
> > rules.make:138: warning: overriding commands for target `/man1'
> > rules.make:135: warning: ignoring old commands for target `/man1'
> > rules.make:140: warning: overriding commands for target `/man3'
> > rules.make:135: warning: ignoring old commands for target `/man3'
> > rules.make:143: warning: overriding commands for target `/man3f'
> > rules.make:135: warning: ignoring old commands for target `/man3f'
> > rules.make:148: warning: overriding commands for target `/'
> > rules.make:124: warning: ignoring old commands for target `/'

All of the above warnings make me think that something went wrong in
setting the makefile macros in file "macros.make".

> > 
> > Making `all' in directory /Net/usr/home/wrjohns/src/netcdf-3.3/src/libsrc
> > 
> > gmake[2]: Entering directory 
> > `/Net/usr/home/wrjohns/src/netcdf-3.3/src/libsrc'
> > ../rules.make:118: warning: overriding commands for target `/'
> > ../rules.make:116: warning: ignoring old commands for target `/'
> > ../rules.make:124: warning: overriding commands for target `/'
> > ../rules.make:118: warning: ignoring old commands for target `/'
> > ../rules.make:148: warning: overriding commands for target `/'
> > ../rules.make:124: warning: ignoring old commands for target `/'
> > gcc    -c attr.c
> > gcc    -c dim.c
> > gcc    -c error.c
> > gcc    -c -DVERSION=`cat ../VERSION` libvers.c
> > gcc    -c nc.c
> > gcc    -c ncio.c
> > posixio.c: In function `ncio_close':
> > In file included from ncio.c:8:
> > posixio.c:926: warning: passing arg 1 of `unlink' discards `const' from 
> > pointer target type
> > gcc    -c ncx.c
> > gcc    -c putget.c
> > gcc    -c string.c
> > gcc    -c v1hpg.c
> > gcc    -c v2i.c
> > gcc    -c var.c
> > ar rv libnetcdf.a attr.o dim.o error.o libvers.o nc.o ncio.o ncx.o putget.o 
> > string.o v1hpg.o v2i.o var.o
> > a - attr.o
> > a - dim.o
> > a - error.o
> > a - libvers.o
> > a - nc.o
> > a - ncio.o
> > a - ncx.o
> > a - putget.o
> > a - string.o
> > a - v1hpg.o
> > a - v2i.o
> > a - var.o
> > libnetcdf.a
> > gmake[2]: execvp: libnetcdf.a: Permission denied
> > gmake[2]: *** [libnetcdf.a] Error 127
> > gmake[2]: Leaving directory 
> > `/Net/usr/home/wrjohns/src/netcdf-3.3/src/libsrc'
> > gmake[1]: *** [subdir_target] Error 1
> > gmake[1]: Leaving directory `/Net/usr/home/wrjohns/src/netcdf-3.3/src'
> > gmake: *** [libsrc/all] Error 2

The above suggests that the macro RANLIB in the file "macros.make" was
set by the configure script to the empty string.  Is this true?

Try building version 3.3.1 and let me know what you find.

--------
Steve Emmerson   <address@hidden>