Re: gEDA-user: Problem with PCB makefile

[Dan McMahill <dan@xxxxxxxxxxxx>]

Karel Kulhavy wrote:
> Isn't it possible to alleviate the problem with the makefile calling
> recursively non-gnu make by using $(MAKE) instead of make in the makefile?
>
> CL<

It should already be doing this.

At the top of the source tree where you had the failure, what does

  grep SET_MAKE config.status

show?

Also, near line 57 or so of config.log, you should see something like:

configure:1571: checking whether make sets $(MAKE)
configure:1591: result: yes

what do you have there?

Here's my current theory.  'configure' tries to see if the make program 
sets the $(MAKE) variable automatically.  Many do.  For example I see on 
a NetBSD-2.0 box:

dan 134 % cat Makefile

all:
        @echo "MAKE is set to ${MAKE}"

dan 135 % make
MAKE is set to make

dan@ 136 % gmake
MAKE is set to gmake

dan@bondage 137 % alias foomake gmake
dan@bondage 138 % foomake
MAKE is set to gmake

so you can see that the MAKE variable is always set to the right thing 
depending on which make program was invoked.

If the make program does not set this variable, then the configure 
script is supposed to detect this and do something specifically to set 
it.  The catch is ./configure checks the program called 'make', not 
'gmake' or what ever t/csh has aliased 'make' to be.  So my theory 
(which could be wrong) is that your 'make' does not set MAKE.

If you set MAKE in your configuration environment, then configure will 
use the specified make program.  In other words

env MAKE=gmake ./configure

will check to see if gmake sets MAKE.


The good news is that the thing which caused the OpenBSD make to fail in 
pcb/src has been fixed in pcb's cvs.

-Dan


_______________________________________________
geda-user mailing list
geda-user@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user