From mboxrd@z Thu Jan  1 00:00:00 1970
From: Diego Novillo <dnovillo@redhat.com>
To: Jason Merrill <jason@redhat.com>
Cc: Gabriel Dos Reis <gdr@integrable-solutions.net>, Neil Booth <neil@daikokuya.co.uk>, "gcc@gcc.gnu.org" <gcc@gcc.gnu.org>
Subject: Re: Language-independent functions-as-trees representation
Date: Wed, 28 Aug 2002 08:02:00 -0000
Message-id: <1030546945.27089.5.camel@frodo>
References: <wvllm86k1m1.fsf@prospero.cambridge.redhat.com> <wvl65y1px1j.fsf@prospero.cambridge.redhat.com> <20020823200122.GF7880@daikokuya.co.uk> <wvl4rdf439u.fsf@prospero.cambridge.redhat.com> <m37kibqizr.fsf@soliton.integrable-solutions.net> <wvln0r72hxe.fsf@prospero.cambridge.redhat.com>
X-SW-Source: 2002-08/msg01840.html

On Wed, 2002-08-28 at 10:10, Jason Merrill wrote:
> On 28 Aug 2002 14:13:12 +0200, Gabriel Dos Reis <gdr@integrable-solutions.net> wrote:
> 
> > Jason Merrill <jason@redhat.com> writes:
> >
> > | Does anyone object to using TREE_COMPLEXITY for source position in all
> > | expressions (i.e. things which have TREE_COMPLEXITY, not *_DECL, *_CST)?
> >
> > Sorry, I'm unclear here.  We already have DECL_SOURCE_LOCATION to
> > indicate source location for _DECLs.  Would you mind elaborating a bit
> > on your idea?
> 
> Yes, but that only applies to _DECLs.  We need source location info for
> other expressions, too, if we're going to have useful debugging info.  I'm
> suggesting that we use TREE_COMPLEXITY for this purpose.
> 
Minor nit.  We would be renaming the accessor to something more
intuitive than TREE_COMPLEXITY, right?


Diego.