This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH, Fortran, OpenACC] Fix PR70598, Fortran host_data ICE


On Mon, May 09, 2016 at 10:26:50PM +0800, Chung-Lin Tang wrote:
> 2015-05-09  Chung-Lin Tang  <cltang@codesourcery.com>
> 
> 	gcc/
> 	* fortran/openmp.c (resolve_omp_clauses): Adjust use_device clause
> 	handling to only allow pointers and arrays.

As has been mentioned earlier, this should go into gcc/fortran/ChangeLog
without fortran/ prefix.

> --- gcc/fortran/openmp.c	(revision 236020)
> +++ gcc/fortran/openmp.c	(working copy)
> @@ -3743,11 +3743,18 @@ resolve_omp_clauses (gfc_code *code, gfc_omp_claus
>  			      && CLASS_DATA (n->sym)->attr.allocatable))
>  			gfc_error ("ALLOCATABLE object %qs in %s clause at %L",
>  				   n->sym->name, name, &n->where);
> -		      if (n->sym->attr.pointer
> -			  || (n->sym->ts.type == BT_CLASS && CLASS_DATA (n->sym)
> -			      && CLASS_DATA (n->sym)->attr.class_pointer))
> -			gfc_error ("POINTER object %qs in %s clause at %L",
> -				   n->sym->name, name, &n->where);
> +		      if (n->sym->attr.flavor == FL_VARIABLE
> +			  && !n->sym->as && !n->sym->attr.pointer

Better put every && on a separate line if the whole if (...) doesn't fit
on a single line.

> +			  && !n->sym->attr.cray_pointer
> +			  && !n->sym->attr.cray_pointee)

This is too ugly.  I'd instead move the if after the cray pointer/pointee
tests, i.e.
if (n->sym->attr.cray_pointer)
  gfc_error (...);
else if (n->sym->attr.cray_pointee)
  gfc_error (...);
else if (n->sym->attr.flavor == FL_VARIABLE
	 && !n->sym->as
	 && !n->sym->attr.pointer)
  gfc_error (...);

> +			gfc_error ("%s clause variable %qs at %L is neither "
> +				   "a pointer nor an array", name,
> +				   n->sym->name, &n->where);

Why pointer rather than POINTER ?

> +		      if (n->sym->ts.type == BT_CLASS && CLASS_DATA (n->sym)
> +			  && CLASS_DATA (n->sym)->attr.class_pointer)
> +			gfc_error ("POINTER object %qs of polymorphic type in "
> +				   "%s clause at %L", n->sym->name, name,
> +				   &n->where);

And, doesn't this mean you emit 2 errors, the first one because it is not
a pointer nor array, and the second one because it is a class with
attr.class_pointer?  Also put && on the next line.

I think best would be to make it else if after the cray pointer/pointee
checks.

	Jakub


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]