This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Suggestion for option
- To: nix at esperi dot demon dot co dot uk, pedwards at disaster dot jaj dot com
- Subject: Re: Suggestion for option
- From: "RenE J.V. Bertin" <rjvbertin at hotmail dot com>
- Date: Fri, 27 Jul 2001 16:32:28 +0200
- Cc: mark at codesourcery dot com, gcc at gcc dot gnu dot org
>
>On 26 Jul 2001, Phil Edwards gibbered:
> > I don't suppose it would be as easy as noticing an error_node result and
> > setting flag_syntax_only to true for subsequent RTL generation to
>notice?
>
>It should be. Note that setting flag_syntax_only doesn't quash all code
>generation, but it stops most of it.
>
>What should we do about the output file? It'll be even more garbaged
>than it was before, after all, but it might be stdout or a FIFO or
>something that we can't safely delete. -fsyntax-only has driver help to
>avoid this question, but if we're switching states in the middle of a
>compilation we must handle it in the compiler proper...
>
>(And yes, I do know people who run compilers and stuff their output down
What happens to those people when/if they use -fsyntax-only? :) I guess one
can't guarantee anyway that an inproperly caused (output-related) otion
won't wreak havoc downstream... So is it really a problem if an option like
the one under discussion is not compatible with all potential ways people
use the compiler?
RB
_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp