Re: [PATCH] (Partial) Implementation of simplificaiton of CSHIFT

[Steve Kargl <sgk@troutmask.apl.washington.edu>]

On Sat, Nov 21, 2015 at 11:19:22AM -0800, H.J. Lu wrote:
> On Sat, Nov 21, 2015 at 10:20 AM, Steve Kargl
> <sgk@troutmask.apl.washington.edu> wrote:
> > On Sat, Nov 21, 2015 at 10:07:35AM -0800, H.J. Lu wrote:
> >> On Sat, Nov 21, 2015 at 8:26 AM, Steve Kargl
> >> <sgk@troutmask.apl.washington.edu> wrote:
> >> > On Sat, Nov 21, 2015 at 11:41:51AM +0100, Paul Richard Thomas wrote:
> >> >>
> >> >> Just a couple of small typos:
> >> >> "Unexpected expr_type cause an ICE" ;  causes?
> >> >> "! An array of derived types workd too." ; works?
> >> >>
> >> >> Apart from that it's OK for trunk.
> >> >>
> >> >> Thanks for the patch
> >> >>
> >> >
> >> > Thanks for the the review.  I don't have a clue as
> >> > to how to do simplification for rank > 2. :(
> >> >
> >>
> >> It breaks bootstrap:
> >>
> >>   int dm;
> >>
> >>   /* DIM is only useful for rank > 1, but deal with it here as one can
> >>      set DIM = 1 for rank = 1.  */
> >>   if (dim)
> >>     {
> >>       if (!gfc_is_constant_expr (dim))
> >> return NULL;
> >>       dm = mpz_get_si (dim->value.integer);
> >>     }
> >>   else
> >>     dm = 1;
> >>
> >> dm is set, but never used.
> >>
> >
> > Perhaps, bootstrap needs to set appropriate warning levels.
> 
> https://gcc.gnu.org/ml/gcc-regression/2015-11/msg00648.html
> 

See 5 lines up.

-- 
Steve