This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: 3.0 bogus file information in debugging output
- From: Zack Weinberg <zack at codesourcery dot com>
- To: Mark Mitchell <mark at codesourcery dot com>
- Cc: "gcc-bugs at gcc dot gnu dot org" <gcc-bugs at gcc dot gnu dot org>,"neil at daikokuya dot demon dot co dot uk" <neil at daikokuya dot demon dot co dot uk>
- Date: Wed, 30 Jan 2002 13:38:16 -0800
- Subject: Re: 3.0 bogus file information in debugging output
- References: <20020130205038.GJ754@codesourcery.com> <25830000.1012424200@gandalf.codesourcery.com>
On Wed, Jan 30, 2002 at 12:56:40PM -0800, Mark Mitchell wrote:
> Zack Weinberg <zack@codesourcery.com> wrote:
>
> >3.0.3 seems to have the same bug with functions with #line in the
> >middle, that Neil fixed for the mainline not so long ago. I think
> >this is worth addressing in 3.0.4.
> >
> >(There's still a bug with this stuff in 3.1, see next message.)
>
> Assuming this is a regression from GCC 2.95,
It is.
> and assuming that you
> think Neil's patch is not likely to introduce new problems,
My bad - it's not Neil who fixed it in 3.1, it was Jason Merrill.
There are two patches that might be worth considering for 3.0:
http://gcc.gnu.org/ml/gcc-patches/2002-01/msg01082.html
(Jason's patch to deal with a change of filename in the middle
of a function)
http://gcc.gnu.org/ml/gcc-patches/2002-01/msg00278.html
(Jeff Law's patch to get line numbers correct in multi-line
conditionals)
And then, having seen what they both did, I think I know what's wrong
with the filename info for the beginning of a function, and will try
to patch it for 3.1 - then we might want to consider *that* for 3.0.
zw