This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Request for Obsoletion: DWARF
- From: mike stump <mrs at windriver dot com>
- To: jason at redhat dot com, mark at codesourcery dot com
- Cc: gcc at gcc dot gnu dot org
- Date: Sat, 6 Jul 2002 11:32:23 -0700 (PDT)
- Subject: Re: Request for Obsoletion: DWARF
- References: <wvl3cuxbwi8.fsf@prospero.cambridge.redhat.com>
> To: Mark Mitchell <mark@codesourcery.com>
> From: Jason Merrill <jason@redhat.com>
> Date: Sat, 06 Jul 2002 00:11:43 +0100
> >> I think there are still a number of compilers and debuggers in the
> >> embedded space that use DWARF 1.1 with some C++ extensions that we
> >> haven't bothered to implement.
> > Why do we care? (Not meaning to be flip; a serious question.)
> Well, currently, people using such tools can build their code with
> gcc and debug it with their random debugger. If we remove dwarf1
> support, that won't work anymore, which would make gcc less useful
> to developers using other tools. I don't know if such tools are
> still being actively developed, or how widely used they are, but my
> guess would be yes, and pretty widely in the embedded space.
> Why don't we deprecate it and see if people complain?
My take, some people do use Dwarf 1 in the embedded space. We (Wind
River) switched from stabs to Dwarf 2 enmass, and it went fairly
smooth despite thrid party tools and such. We have yet to feel some
potential downsides from it, so maybe the jury is still out, but I
think it would be safe to announce/disable/remove dwarf 1 support. I
would not be in favor of removing stabs for a while.