ajudge@maths.tcd.ie (Alan Judge) (12/11/89)
I have just obtained GNU Make V3.57, having been using V3.27 up to now,
and there would appear to be a new bug in the handling of chained rules.
The following script illustrates the problem. The first make works OK, but
the second make, which should do nothing, screws up.
The makefile works fine with gmake 3.27.
If you have a fix for this, I would be very grateful if you would mail it to
me.
(Note: I am running make on a VAX, and the bug appears with both cc and gcc.)
--
Thanks,
Alan
SysAdmin, Dept. of Maths, Trinity College, Dublin, Ireland.
<ajudge@maths.tcd.ie>
Script started on Sun Dec 10 16:16:47 1989
csaran 1% cat > Makefile
t: t.o
cc -o t t.o
csaran 2% mkdir RCS
csaran 3% cat > t.c
main(){}
csaran 4% ci -u t.c
RCS/t.c,v <-- t.c
initial revision: 1.1
enter description, terminated with ^D or '.':
NOTE: This is NOT the log message!
>>
done
csaran 5% ls
Makefile RCS t.c
csaran 6% rm t.c
rm: override protection 444 for t.c? y
csaran 7% /bofin/src/gnu/make-3.57/make -v
GNU Make version 3.57, by Richard Stallman and Roland McGrath.
Copyright (C) 1988, 1989 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE.
test -f t.c || co RCS/t.c,v t.c
RCS/t.c,v --> t.c
revision 1.1
done
cc -c t.c
cc -o t t.o
rm t.c
csaran 8% !!
/bofin/src/gnu/make-3.57/make -v
GNU Make version 3.57, by Richard Stallman and Roland McGrath.
Copyright (C) 1988, 1989 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE.
cc -c t.c
0: No source file t.c
make: *** [t.o] Error 1
csaran 8%
script done on Sun Dec 10 16:18:14 1989moss@takahe.cs.umass.edu (Eliot &) (12/16/89)
We have had a similar though slightly different problem. We have rules
something like this:
%.dvi: %.tex
latex ....
%.ps: %.dvi
dvips
%.print: %.ps
printing program
The first make runs fine. then I update foo.tex and invoke make again. It
*does* run latex, but *not* dvips. This happens (apparently) because I am
running make with -n and it is not updating its internal concept of the time
on foo.dvi, so it thinks foo.dvi is up to date. (I use make to build a script
that I run separately because it asks interactive questions as it goes, etc.)
For my usage, this is losing behavior, and it *worked* in the past. the
current release we have installed is 3.57. I seem to recall some recent
messages about this point; is it being fixed? Thanks for any info you can
offer! Eliot
--
J. Eliot B. Moss, Assistant Professor
Department of Computer and Information Science
Lederle Graduate Research Center
University of Massachusetts
Amherst, MA 01003
(413) 545-4206; Moss@cs.umass.edu