Re: Submake question

[Paul D. Smith]

%% address@hidden writes:

  j> distall:
  j>       @for dir in $(GEN_MAKES_DIR); do \
  j>             make -C $$dir distall; \
  j>       done

First, _never_ use "make" to invoke submakes.  You should _always_ use
the variable $(MAKE).

Second, this is not a good way to do sub-makes; better is to create a
target for each directory and have the rule for it build that one
subdirectory.  Then you won't need a loop at all and if the sub-make
fails the build will fail, unless you specify -k.

There's an example of this in the GNU make manual section on "Phony
Targets".

  j> My question is : How can i stop the process if i get some error in
  j> a make ?  I have read in 'make' manual, chapter 9, that the exit
  j> status is not equal to 0 if any error is found, but how can i
  j> access this value?

This is a shell question, not a make question.  The contents of the
command script are shell commands.

  j> distall:
  j>       @for dir in $(GEN_MAKES_DIR); do \
  j>             result = make -C $$dir distall; \
  j>             ifneq $result  0
  j>                   $(error Subproject not made! )
  j>             endif
  j>       done

Well, you cannot use the make $(error ...) function because that will
always be evaluated before the shell script is even invoked.

You can do this two ways:

 distall:
       @set -e; \
        for dir in $(GEN_MAKES_DIR); do \
             $(MAKE) -C $$dir distall; \
       done

or:

 distall:
       @for dir in $(GEN_MAKES_DIR); do \
             $(MAKE) -C $$dir distall || exit 1; \
       done

See your documentation for the Bourne shell (sh) for more info.

Note neither of these is optimal because now you have the opposite
problem: even if you run "make -k" it will still stop after the first
failure.

-- 
-------------------------------------------------------------------------------
 Paul D. Smith <address@hidden>          Find some GNU make tips at:
 http://www.gnu.org                      http://make.paulandlesley.org
 "Please remain calm...I may be mad, but I am a professional." --Mad Scientist