This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: [PATCH] (Partial) Implementation of simplificaiton of CSHIFT
- From: Steve Kargl <sgk at troutmask dot apl dot washington dot edu>
- To: Gerald Pfeifer <gerald at pfeifer dot com>
- Cc: "H.J. Lu" <hjl dot tools at gmail dot com>, Paul Richard Thomas <paul dot richard dot thomas at gmail dot com>, fortran at gcc dot gnu dot org, gcc-patches at gcc dot gnu dot org
- Date: Thu, 31 Dec 2015 08:13:34 -0800
- Subject: Re: [PATCH] (Partial) Implementation of simplificaiton of CSHIFT
- Authentication-results: sourceware.org; auth=none
- References: <20151120005836 dot GA53763 at troutmask dot apl dot washington dot edu> <20151120200947 dot GA61350 at troutmask dot apl dot washington dot edu> <CAGkQGi+bvMEzo_xk_=Ugu5GdrtV28peFvOQHnAAqAUoSTJaVxw at mail dot gmail dot com> <20151121162657 dot GA23304 at troutmask dot apl dot washington dot edu> <CAMe9rOqOJKti4JRKbRMB-QRRJ1HMCdNTL8F14GYo9uNNgOQBSQ at mail dot gmail dot com> <20151121182014 dot GA23753 at troutmask dot apl dot washington dot edu> <CAMe9rOrv0tj2r6c1W482gcqj+BGdkYrPVS04AQGiVnrpdb8RcA at mail dot gmail dot com> <20151121192622 dot GA81197 at troutmask dot apl dot washington dot edu> <20151121200735 dot GA55464 at troutmask dot apl dot washington dot edu> <alpine dot LSU dot 2 dot 20 dot 1512291725490 dot 5144 at anthias>
On Thu, Dec 31, 2015 at 09:57:10PM +0800, Gerald Pfeifer wrote:
> On Sat, 21 Nov 2015, Steve Kargl wrote:
> > 2015-11-21 Steven G. Kargl <kargl@gcc.gnu.org>
> >
> > * simplify.c (gfc_simplify_cshift): Work around bootstrap issues
> > due to inappropriate warning options.
>
> > Index: simplify.c
> > ===================================================================
> > + /* GCC bootstrap is too stupid to realize that the above code for dm
> > + is correct. First, dim can be specified for a rank 1 array. It is
> > + not needed in this nor used here. Second, the code is simply waiting
> > + for someone to implement rank > 1 simplification. For now, add a
> > + pessimization to the code that has a zero valid reason to be here. */
> > + if (dm > array->rank)
> > + gcc_unreachable ();
>
> I'm not sure this comment is appropriate as is. At a minimum, it
> should list the version of GCC this was introduced for/with. So,
> something like
>
I have no intention to change the comment.
--
Steve