RE: [PATCH 1/1] gdb/dwarf, fortran: Use the function scope for DW_TAG_imported_declaration

["Ijaz, Abdul B" <abdul.b.ijaz@intel.com>]

Thanks Tom for the feedback.

>> +      if (cu->lang () == language_fortran)
>> +	list_to_add = cu->list_in_scope;
>> +      else
>> +	list_to_add = &cu->get_builder ()->get_global_symbols ();

Tom>That is, do we know for sure that there's some language requiring that a local import be stuffed into the global scope?

Abdul> Also not sure about the others languages.

Tom>I tend to think such DWARF wouldn't exist and the code would be easier to maintain in the long run without a fortran-specific check.
Tom>Would you mind giving this a try?

Tom>	      list_to_add
Tom>		= ((cu->list_in_scope
Tom>		    == &cu->get_builder ()->get_file_symbols ())
Tom>		   ? &cu->get_builder ()->get_global_symbols ()
Tom>		   : cu->list_in_scope);

Abdul>Should be working fine. I will try this change and if it works fine then I will push the patch 2 with this change.

Best Regards,
Abdul Basit

-----Original Message-----
From: Tom Tromey <tom@tromey.com> 
Sent: Friday, May 1, 2026 6:49 PM
To: Ijaz, Abdul B <abdul.b.ijaz@intel.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [PATCH 1/1] gdb/dwarf, fortran: Use the function scope for DW_TAG_imported_declaration

>>>>> Abdul Basit Ijaz <abdul.b.ijaz@intel.com> writes:

> All Fortran imported variable aliases (`use module, alias => var`) 
> were being added to "global scope", regardless of whether they appeared in:
> - Program/module scope (should be global)
> - Function scope (should be local)

Thanks for the patch.

>      case DW_TAG_imported_declaration:
> +      sym->set_domain (TYPE_DOMAIN);
> +      sym->set_loc_class_index (LOC_TYPEDEF);
> +      /* For Fortran imported declarations, use current scope context
> +	 rather than forcing global scope.  This ensures variable aliases
> +	 are scoped correctly (function-level, module-level, or
> +	 program-level).  */
> +      if (cu->lang () == language_fortran)
> +	list_to_add = cu->list_in_scope;
> +      else
> +	list_to_add = &cu->get_builder ()->get_global_symbols ();

I wonder if this could simply omit the 'if' and do the "usual"
file->global transform like:

	      list_to_add
		= ((cu->list_in_scope
		    == &cu->get_builder ()->get_file_symbols ())
		   ? &cu->get_builder ()->get_global_symbols ()
		   : cu->list_in_scope);

That is, do we know for sure that there's some language requiring that a local import be stuffed into the global scope?

I tend to think such DWARF wouldn't exist and the code would be easier to maintain in the long run without a fortran-specific check.

Would you mind giving this a try?

thanks,
Tom
Intel Deutschland GmbH
Registered Address: Dornacher Strasse 1, 85622 Feldkirchen, Germany
Tel: +49 89 991 430, www.intel.de
Managing Directors: Harry Demas, Jeffrey Schneiderman, Yin Chong Sorrell
Chairperson of the Supervisory Board: Nicole Lau
Registered Seat: Munich
Commercial Register: Amtsgericht Muenchen HRB 186928