This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: java/Makefile.in patch for missing dependencies (was: Re: Patches for libf2c/lib[FIU]77/Makefile.in (will help ...))
- To: bothner at cygnus dot com
- Subject: Re: java/Makefile.in patch for missing dependencies (was: Re: Patches for libf2c/lib[FIU]77/Makefile.in (will help ...))
- From: Hans-Peter Nilsson <hans-peter dot nilsson at axis dot com>
- Date: Thu, 28 Jan 1999 21:08:21 +0100
- CC: hans-peter dot nilsson at axis dot com, egcs-patches at cygnus dot com
> Date: Thu, 28 Jan 1999 10:49:56 -0800
> From: Per Bothner <bothner@cygnus.com>
> But these patches seems to be dependencies which are *not* needed
> for a working Posix make! At least, I *assume* that it is
> redundant to specify that jv-scan.o depende on jv-scan.c etc.
To say buffer.o depends on buffer.c on one line, and saying
jcf-dump.o depends on jcf-reader.c on the other is not
consistent; rather confusing. I do not believe that has
anything to do with Posix make in particular.
Also dependencies were missing totally for check-init.o.
> However, the main gcc Makefile.in does have these dependencies,
> so I'm confused. Can someone clarify what is going on?
Given that consistency among the dependencies in
java/Makefile.in is wanted, this patch, by adding a few .o to .c
dependences, is less than taking out the .o to .c dependencies
that already were there. As you noted, it also uses the
existing convention of explicitly saying that foo.o depends on
foo.c without defaulting to implicit suffix dependencies.
Trying carefully to avoid waving red blankets,
brgds, H-P