This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Unable to build today, calls.c?
> On Thu, Mar 16, 2000 at 07:52:47AM -0800, Jason Eckhardt wrote:
> >
> > I'm unable to build today:
> >
> > gcc -c -DIN_GCC -g -O2 -DHAVE_CONFIG_H -I. -I../../egcs/gcc -I../../egcs/gcc/config -I../../egcs/gcc/../include ../../egcs/gcc/calls.c
> > ../../egcs/gcc/calls.c: In function `emit_library_call_value':
> > ../../egcs/gcc/calls.c:3736: `unadjusted_args_size' undeclared (first use in this function)
> > ../../egcs/gcc/calls.c:3736: (Each undeclared identifier is reported only once
> > ../../egcs/gcc/calls.c:3736: for each function it appears in.)
> > make[1]: *** [calls.o] Error 1
> > make: *** [all-gcc] Error 2
> >
> >
> > It seems to be related to:
> > Thu Mar 16 16:01:30 MET 2000 Jan Hubicka <jh@suse.cz>
> >
> > * calls.c (expand_call): Do sanity checking on arg_space_so_far.
> > Update arg_space_so_far on stack adjustments.
> > (emit_library_call, emit_library_call_value): Likewise; take into
> > account arg_space_so_far and pending_stack_adjust when calculcating
> > the boundary.
>
> I've reverted both of your calls.c changes today, Jan.
Oops, the problem is found only on in the code:
#ifndef PUSH_ARGS_REVERSED
#ifdef PREFERRED_STACK_BOUNDARY
/* If we pushed args in forward order, perform stack alignment
after pushing the last arg. */
if (argblock == 0)
{
anti_adjust_stack (GEN_INT (args_size.constant
- original_args_size.constant));
arg_space_so_far += args_size.constant - unadjusted_args_size;
}
#endif
#endif
the unadjusted_args_size ought to be original_args_size.constant. I've blindly
copied the code from expand_call and the values are renamed. We really ought
to kill the code duplication here.
The problem didn't exhibited himeslf on i386 bootstrap, only on the machines
with PUSH_ARGS_REVERSED.
May I re-try to install both patches with the variable renamed?
or was there some other problems?
Honza
>
>
> r~