Heikki,
You don't say exactly what happens when you change the constants
and recompile cilink.p but it is possible that the compilation is going
quite happily, but the resulting binary is not replacing the original.
On my setup (on a Sun 4) I have all my binaries in /usr/local/bin, which
is on my $PATH. The tree for the Linkage package itself looks a bit like
this:-
/home
|
linkage51
|
+----------+-------------+
| | |
gen thg (the rest)
|
+---------+---------+
| | |
src obj bin
| | |
cilink.p cilink.o cilink
Doing a cd to /home/linkage51/thg/src and then a make cilink results in
a new cilink.o in the obj subdirectory and a new cilink in the bin sub-
directory. However, I then have to do a mv ../bin/cilink /usr/local/bin
to get the new binary where it will be found when it is called. Unless it
is moved, overwriting the original binary, I still get the same run-time
error as it is the original binary that is being executed.
If you check, you may well find that this is happening to you, too, and
is the cause of your problems. If not, please post a summary of the
messages produced when you do the make and I will try to diagnose the
problem.
Note, you can try out a newly made binary whilst it is still in the sub-
directory where make puts it. All you do is edit your pedin script file and
replace line(s) containing call(s) to cilink with (in this case)
/home/linkage51/thg/bin/cilink.
Hope this helps
John Attwood,
MRC Human Biochemical Genetics Unit,
London,
England.
