This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
Dean, >Date: Tue, 07 Dec 1999 17:08:19 -0800 >From: "Dean N. Williams" <address@hidden> >Organization: PCMDI >To: Steve Emmerson <address@hidden> >Subject: Re: 19991207: "make nf_test/test" failure: Red Hat Linux 6.1: fort77 >Keywords: 199912071402.HAA12184 In the above message, you wrote: > I sent two files config.log and configure.log. Check out the > configure.log. I did -- which is why I wondered how and why your fort77 differs from ours. > %uname -a > Linux chico 2.2.12-20 #1 Mon Sep 27 10:40:35 EDT 1999 i686 unknown > %fort77 -v > /usr/bin/fort77: fort77 Version 1.14a > /usr/bin/fort77: No input files specified Odd. Your fort77 compiler appears to be the same version (1.14a) as ours. Yet our build using fort77 worked and yours didn't. If you wouldn't mind, what does the eqivalent of the following do on your system? (I'm using standard UNIX shell syntax.) $ cat >conftest.c <<'EOF' void sub_(values) int values[4]; { exit(values[1] != -2 || values[2] != -3); } EOF $ cc -c -DNDEBUG -Df2cFortran conftest.c $ cat >conftestf.f <<'EOF' integer values(4) data values /-1, -2, -3, -4/ call sub(values) end EOF $ fort77 -c -g -Nx400 -w conftestf.f MAIN: $ fort77 -o conftest -g -Nx400 -w conftestf.o conftest.o $ if ./conftest; then echo success; else echo failure; fi success This is what the configure script does to determine if a Fortran "integer" is equivalent to a C "int". If it fails, then something is seriously wrong and I'd love to know what. -------- Steve Emmerson <http://www.unidata.ucar.edu>