Subject: Re: sun-lamp CVS commits
To: Greg Hudson <email@example.com>
From: Chris G Demetriou <Chris_G_Demetriou@BALVENIE.PDL.CS.CMU.EDU>
Date: 12/18/1995 00:02:03
> After looking over pr 1488, it's not at all clear to me why you
> introduced a violation of the POSIX.2 standard (section 18.104.22.168) in
> order to make a broken makefile work.
> foo: foo.o ../bfd/libbfd.a
> cc -o foo foo.o ../bfd/libbfd.a
> /bin/rm -f foo *.o
> make is functioning correctly
I found this bug building the cygnus GNU toolchain sources. at one
point (not too long ago), i pointed it out to J.T. and he said that
the makefiles in question _were_ correctly formed, and that make was
to blame (though not for the .c.a rule explicitly; it's correct to
have a .c.a rule, just, at the time, according to him, make was doing
the wrong thing with it)....
If it's a problem with the GNU toolchain makefiles, then they should
fix them. However, it's apparently not a problem, because they work
on (i would assume) EVERY OTHER SYSTEM THEY'VE BEEN TRIED ON.
What that says to me is "it's our problem," and since i want my
non-kernel source tree to match the master sources (and want to be
able to _build_ my port), i made this change which _only_ affects the