Re: misunderstanding -s behavior

[balducci]

dear paul

I appreciate really very much your promptness and kindness

> It's not that -s is handled differently at different recursion levels.
> It's that -s does not automatically disable the directory entry/exit
> statements.  That's why you need to also include the
> --no-print-directory flag (which does disable the directory entry/exit
> statements).

so, I was erroneously assuming that -s means "suppress everything",
while it only suppresses line recipes (which is actually clearly stated
in the manual...)

> 
> The directory entry/exit statements are not printed at the top-most
> level of make, because it's assumed that you know what directory you
> were in and what makefile you were using when you typed "make" and so
> everything that happens in that directory doesn't need further
> commentary.
> 
> Once make invokes a sub-make, however, the output you see is possibly
> from a different makefile and/or directory, so make prints the
> enter/exit statements around those commands so you know that.

OK, now things are much clearer

I now think that the problem is in the emacs Makefile:

The definition:

     info_misc=`cd doc/misc; ${MAKE} -s echo-info`

assumes that the Makefile will never be run by a sub-make, which is
true most of the time, but is false in my case. A more general version
would thus be:

     info_misc=`cd doc/misc; ${MAKE} --no-print-directory -s echo-info`

I will drop a note to the emacs developers

I just want to thank you very much again for your work and patience


ciao
gabriele