This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: x86_64 related expand_movcc fix
> On Thu, Oct 18, 2001 at 04:20:59PM +0200, Jan Hubicka wrote:
> > ! if (ct && !expand_simple_binop (mode, PLUS,
> > ! tmp, GEN_INT (ct),
> > ! tmp, 1, OPTAB_DIRECT))
> > ! abort ();
>
> expand_binop is not guaranteed to use the destination provided.
> You need to reassign tmp.
I was looking at the returns in builtin_binop if there is such a case and
didn't find any (I know that in general expanders behave that way).
But it is probably cleaner to add the assignment. Only it will require
emit_move_insns calls in some cases.
Honza
>
>
>
> r~