This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: BINFO_OFFSET
- To: gcc at gcc dot gnu dot org, kenner at vlsi1 dot ultra dot nyu dot edu
- Subject: Re: BINFO_OFFSET
- From: Mike Stump <mrs at windriver dot com>
- Date: Tue, 22 Feb 2000 17:38:31 -0800 (PST)
> Date: Sun, 20 Feb 00 08:45:20 EST
> From: kenner@vlsi1.ultra.nyu.edu (Richard Kenner)
> To: gcc@gcc.gnu.org
> Sometimes it's passed to ssize_binop,
I think these calls to ssize_binop are wrong. I think they should all
be size_binop. They all happen in cp/class.c as far as I can tell.
> I'm looking at all calls to size_binop and it appears there are
> whole bunch of typing errors in the cp directory: I want to have
> size_binop abort if it is called with invalid types of operands.
Sounds good, but before doing it, the cp code will have to be fixed.
Some of the uses are + (base + offset, 0 - offset) to get base, some
of them are - (base+offset, offset) to get base and so on. In the
generated code, it should never matter that they are signed or
unsigned, so the error is a relatively harmless one.