[Bug tree-optimization/63986] [5 Regression][SH] gcc.target/sh/pr51244-15.c failures

["rguenther at suse dot de" <gcc-bugzilla at gcc dot gnu dot org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63986

--- Comment #6 from rguenther at suse dot de <rguenther at suse dot de> ---
On Thu, 20 Nov 2014, ppalka at gcc dot gnu.org wrote:

> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63986
> 
> --- Comment #5 from ppalka at gcc dot gnu.org ---
> (In reply to Richard Biener from comment #3)
> > Ok, now already existing forwprop code gets fed with
> > 
> >   <bb 2>:
> >   _3 = a_2(D) == 0;
> >   x_4 = (char) _3;
> >   _7 = ~_3;
> >   _8 = (int) _7;
> >   MEM[(int *)d_5(D) + 8B] = _8;
> >   if (x_4 != 0)
> > 
> > where we now in the first forwprop pass identified the opportunity
> > to use ~_3 instead of x_4 == 0 thus x_4 is now no longer multi-use.
> > This makes us optimize if (x_4 != 0) to if (_3 != 0) which we
> > re-optimize in fold_gimple_cond now to '_3' and then of course
> > if (_3 != 0) (err, and we return "changed"....) which means we
> > now propagate _again_ via forward_propagate_into_gimple_cond
> > which now specifically allows aggressive forwarding of compares,
> > bypassing single-use restrictions.  See
> > 
> > 2014-11-16  Patrick Palka  <ppalka@gcc.gnu.org>
> > 
> >         PR middle-end/63790
> >         * tree-ssa-forwprop.c (forward_propagate_into_comparison_1):
> >         Always combine comparisons or conversions from booleans.
> > 
> > thus me fixing my "mistake" does not help anymore.
> 
> I worried that such an issue might pop up...
> 
> If you think my patch should be reverted, note that there is a more benign
> version of the patch here
> https://gcc.gnu.org/ml/gcc-patches/2014-11/msg00963.html (which still fixes the
> test case I was originally interested in fixing).

No, I think that conceptually aggressive forwarding into if conditions
is wanted.  I believe it's not impossible to fix the very special
cases that might pop up - usually it will help code generation.

Richard.