Re: [Patch, Fortran, F03] PR 39630: Procedure Pointer Components with PASS

[Daniel Kraft <d at domob dot eu>]

Hi Janus,

Janus Weil wrote:
> The PASS attribute for PPCs is quite similar to the PASS attribute for
> Type-Bound Procedures (which has already been implemented by Daniel
> Kraft some time ago). Consequently my patch stays quite close to
> Daniel's TBP implementation, sharing with it the data structure
> (gfc_typebound_proc) and quite a bit of code (such as
> match_binding_attributes, mio_typebound_proc and update_arglist_pass).
> Of course some parts had to be adjusted specifically to PPCs
> (extract_ppc_passed_object, update_ppc_arglist, etc). Btw the patch
> also contains a small bugfix for TBPs with optional arguments (cf.
> typebound_call_10.f03).

it's nice to see you were really able to reuse some of my work!

> I have successfully tested for regressions on
> x86_64-unknown-linux-gnu. Ok for trunk?

Looks good to me, so ok.  Just one comment:

+	  /* TODO: Make this an error once CLASS is implemented.  */
+	  if (!sym->attr.sequence)
+	    gfc_warning ("Polymorphic entities are not yet implemented,"
+			 " non-polymorphic passed-object dummy argument of '%s'"
+			 " at %L accepted", c->name, &c->loc);
+
+	}

I may mistake here someting, but what has sym->attr.sequence to do with 
polymorphic entities?

Thanks for the patch and your work on PPCs!

Yours,
Daniel

--
Done:  Arc-Bar-Cav-Ran-Rog-Sam-Tou-Val-Wiz
To go: Hea-Kni-Mon-Pri