This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: g++ 3.1 EH broken for irix6.5
- To: Richard Kenner <kenner at vlsi1 dot ultra dot nyu dot edu>
- Subject: Re: g++ 3.1 EH broken for irix6.5
- From: Richard Henderson <rth at redhat dot com>
- Date: Fri, 29 Jun 2001 09:16:02 -0700
- Cc: fjh at cs dot mu dot oz dot au, gcc-patches at gcc dot gnu dot org
- References: <10106291202.AA03894@vlsi1.ultra.nyu.edu>
On Fri, Jun 29, 2001 at 08:02:37AM -0400, Richard Kenner wrote:
> We're not changing their "representation", but their *alignment*. That only
> has an effect if these structs are used, in turn, within a struct that
> *is not* packed and where the field isn't already at its alignment boundary.
> That's a *exceptionally* rare set of circumstances.
You're completely ignoring the set of problems you'll encounter when
linking together objects created with different versions of the compiler.
Under those circumstances, a change in alignment is _easily_ visible,
and not rare at all.
And, CLEARLY, it doesn't even take that to break real programs, since
during this entire conversation, libgcc ITSELF has been broken on
Sparc, MIPS, and other strict-alignment machines.
Please quit arguing that "no one should see this" -- that's false.
r~