[prev in list] [next in list] [prev in thread] [next in thread] 

List:       busybox
Subject:    Re: In Rules.mak, "check_ldflags" appears to be useless.
From:       Rob Landley <rob () landley ! net>
Date:       2006-07-26 18:37:49
Message-ID: 200607261437.49426.rob () landley ! net
[Download RAW message or body]

On Wednesday 26 July 2006 12:26 pm, David Daney wrote:
> Rob Landley wrote:
> > Ok, here's a snippet:
> > 
> > --- Rules.mak   (revision 15742)
> > +++ Rules.mak   (working copy)
> > @@ -166,6 +166,7 @@
> >  endif
> > 
> >  CHECKED_LDFLAGS := $(call check_ld,$(LD),--warn-common,)
> > +CHECKED_LDFLAGS := $(call check_ld,$(LD),--static-libgcc,)
> > 
> > 
> > Now would someone explain to me why check_ld even _exists_ if the build 
still 
> > breaks in the linking phase if you add that line?
> > 
> > I'm confused...
> 
> '--static-libgcc' (note the two leading '-') is not a valid option for 
> any known tool chain component, so adding that line to a makefile has no 
> possible use.

Yeah, but the point of check_ld is that if it doesn't recognize the option, it 
shouldn't add the option.  Yet it is.  I know why --static-libgcc is wrong 
here, but I dunno why our filter is allowing it to break the build.

> I would say that the check_ld function is broken if that is 
> making the build fail.

Yup.

> Note that '-static-libgcc' (note the single leading '-') *is* a valid 
> option for gcc.  Since it is not a ld option there is no reason to try 
> to pass it to ld.

I caught up with that part. ;)

> David Daney

Rob
-- 
Never bet against the cheap plastic solution.
_______________________________________________
busybox mailing list
busybox@busybox.net
http://busybox.net/cgi-bin/mailman/listinfo/busybox
[prev in list] [next in list] [prev in thread] [next in thread]