Re: [tree-ssa] mainline->branch merge status

[Dale Johannesen <dalej at apple dot com>]

On Jan 29, 2004, at 3:12 AM, Andreas Schwab wrote:
> Dale Johannesen <dalej@apple.com> writes:
> > On Jan 28, 2004, at 10:22 AM, Diego Novillo wrote:
> > > A heads-up for those of us following SPEC results.  The new merge 
> > > brings
> > > in the removal of assignments to non-lvalues.  This produces a 
> > > compile
> > > error in 176.gcc:
> > >
> > > reorg.c: In function `find_end_label':
> > > reorg.c:831: error: invalid lvalue in increment
> > > reorg.c: In function `delete_from_delay_slot':
> > > reorg.c:1033: error: invalid lvalue in increment
> > > reorg.c: In function `make_return_insns':
> > > reorg.c:4157: error: invalid lvalue in increment
> > > reorg.c: In function `dbr_schedule':
> > > reorg.c:4237: error: invalid lvalue in increment
> > > specmake: *** [reorg.o] Error 1
> > >
> > > This comes from the old obstack.h file.  I fixed the file, but cc1
> > > miscompares at runtime.  Maybe my fix is wrong, or maybe we've 
> > > inherited
> > > a bug from mainline.  I still don't know.
> >
> > Well, how did you fix it?  The offending code is
> >
> >        *((void **)__o->next_free)++ = ((void *)datum);  \
> >
> > If you just break this into two in the obvious fashion
> >
> >    { \
> >        *((void **)__o->next_tree) = ((void*)datum); \
> >        ((void **)__o->next_tree)++; \
> >    } \
>
> This has the same problem.
>
>    { \
>        *(void **) __o->next_free = (void *) datum; \
>        __o->next_free += sizeof (void *); \
>    } \

True, cut and paste error.  But it still has the problems I mentioned 
before.

> > can happen.  I believe the following is correct, performance neutral, 
> > and
> > standard-conforming:
> >
> >        {
> >        \
> >         void ***t = (void ***)&(__o->next_free);
> >         \
> >          **t = ((void *)datum);                                       
> >   \
> >        (*t)++;
> >        \
> >       }                                            \
>
> I think this will break strict aliasing.

So does the other way;  *(void **)__o->next_free = ... is exactly what 
you can't do.
The above works for me, but let's see if we can't figure out a way to 
do this completely right...