Re: [PATCH][match.pd] PR middle-end/66915 Restrict A - B -> A + (-B) to non-fixed-point types

[Kyrill Tkachov <kyrylo dot tkachov at arm dot com>]

On 23/07/15 12:16, Richard Biener wrote:
> On Thu, 23 Jul 2015, Kyrill Tkachov wrote:
>
> > On 21/07/15 11:11, Richard Biener wrote:
> > > On Tue, 21 Jul 2015, Kyrill Tkachov wrote:
> > >
> > > > On 21/07/15 08:24, Richard Biener wrote:
> > > > > On Mon, 20 Jul 2015, Kyrill Tkachov wrote:
> > > > >
> > > > > > Hi all,
> > > > > >
> > > > > > This patch fixes the PR in question which is a miscompilation of
> > > > > > gcc.dg/fixed-point/unary.c on arm.
> > > > > > It just restricts the A - B -> A + (-B) transformation when the type
> > > > > > is
> > > > > > fixed-point.
> > > > > >
> > > > > > This fixes the testcase for me.
> > > > > > Is this the right approach?
> > > > > >
> > > > > > Bootstrap and test on arm and x86 running.
> > > > > >
> > > > > > Ok if testing is clean?
> > > > > Ok, but I think the fold-const.c code has the same issue, no:
> > > > >
> > > > >          /* A - B -> A + (-B) if B is easily negatable.  */
> > > > >          if (negate_expr_p (arg1)
> > > > >              && !TYPE_OVERFLOW_SANITIZED (type)
> > > > >              && ((FLOAT_TYPE_P (type)
> > > > >                   /* Avoid this transformation if B is a positive
> > > > > REAL_CST.
> > > > > */
> > > > >                   && (TREE_CODE (arg1) != REAL_CST
> > > > >                       ||  REAL_VALUE_NEGATIVE (TREE_REAL_CST (arg1))))
> > > > >                  || INTEGRAL_TYPE_P (type)))
> > > > >            return fold_build2_loc (loc, PLUS_EXPR, type,
> > > > >                                fold_convert_loc (loc, type, arg0),
> > > > >                                fold_convert_loc (loc, type,
> > > > >                                                  negate_expr (arg1)));
> > > > >
> > > > > ah, no.  The above only applies to float-type and integral-types.
> > > > >
> > > > > Thus yes, your patch is ok.  Can you double-check the other pattern,
> > > > >
> > > > > /* -(A + B) -> (-B) - A.  */
> > > > > (simplify
> > > > >     (negate (plus:c @0 negate_expr_p@1))
> > > > >     (if (!HONOR_SIGN_DEPENDENT_ROUNDING (element_mode (type))
> > > > >          && !HONOR_SIGNED_ZEROS (element_mode (type)))
> > > > >      (minus (negate @1) @0)))
> > > > >
> > > > > ?
> > > > Thanks, committed with r226028.
> > > > I can add (FLOAT_TYPE_P (type) || INTEGRAL_TYPE_P (type)) to the
> > > > condition.
> > > > That would more closely mirror the original logic, right?
> > > > That passes x86_64 bootstrap and aarch64 testing looks ok.
> > > Yeah, that works for me, too.
> > How about this patch then?
> > Bootstrapped and tested on x86_64 and aarch64.
> Hmm.  The code already pretty much matches the one in fold-const.c.
>
> So what's the actual issue with fixed-point types and
> -(A + B) -> -B - A iff negate_expr_p says that B can be
> safely negated?
>
> That is, can you add a testcase that fails without the patch?

I don't have such a testcase.
If negate_expr_p does what we want here, then I suppose it's redundant
and I withdraw the patch.
I'm not very familiar with the fold-const.c code...

Kyrill

> Thanks
> Richard.