This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: new alias sets from the backend
- To: Richard Henderson <rth at cygnus dot com>
- Subject: Re: new alias sets from the backend
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Mon, 21 Sep 1998 00:32:30 -0600
- cc: mark at markmitchell dot com, egcs-patches at cygnus dot com
- Reply-To: law at cygnus dot com
In message <19980920232410.A12317@dot.cygnus.com>you write:
> On Sun, Sep 20, 1998 at 10:55:04PM -0600, Jeffrey A Law wrote:
> > Right. varargs/stdargs by their nature tend to create aliased memory
> > accesses which C claims can't/shouldn't happen.
> >
> > Consider the register flush and the loads via va_arg.
>
> Exactly. My thought being to force both of these to use the
> same alias set, and so resolve these problems, as well as
> indicating that they don't interfere with anything else.
But don't you also have to deal with pointers within the va list?
ie, one arg could be an address of another arg in the va_list which
was flushed to the stack. So you have to go beyond just loads from
va_list, you have to track how they're used later.
jeff