[RFC/DWARF2] Fallback unknown language to C?

[RFC/DWARF2] Fallback unknown language to C?

[Joel Brobecker]

This is a followup on a concern reported by Stephane Carrez on the
GDB mailing list:

    http://sources.redhat.com/ml/gdb/2003-02/msg00533.html

> In the past, a gdb without Ada support could be used to, at least, see the
> source.  However, we are now in this sad situation where this is no longer
> possible because now gdb complains at unknown languages and... stops.

I think we shouldn't fail like this because we have some symbols for an
unsupported language. Could we for instance fallback to C for unknown
language symbols.

I looked at set_cu_language in dwarf2read.c. What would be the
consequences of setting cu_language to language_c in these cases:

    case DW_LANG_Ada83:
    case DW_LANG_Ada95:
    case DW_LANG_Cobol74:
    case DW_LANG_Cobol85:
    case DW_LANG_Pascal83:
    case DW_LANG_Modula2:
    default:
      cu_language = language_unknown;
      break;

-- 
Joel

Re: [RFC/DWARF2] Fallback unknown language to C?

[Daniel Jacobowitz]

On Tue, Feb 25, 2003 at 03:02:04PM -0800, Joel Brobecker wrote:
> This is a followup on a concern reported by Stephane Carrez on the
> GDB mailing list:
> 
>     http://sources.redhat.com/ml/gdb/2003-02/msg00533.html
> 
> > In the past, a gdb without Ada support could be used to, at least, see the
> > source.  However, we are now in this sad situation where this is no longer
> > possible because now gdb complains at unknown languages and... stops.
> 
> I think we shouldn't fail like this because we have some symbols for an
> unsupported language. Could we for instance fallback to C for unknown
> language symbols.
> 
> I looked at set_cu_language in dwarf2read.c. What would be the
> consequences of setting cu_language to language_c in these cases:
> 
>     case DW_LANG_Ada83:
>     case DW_LANG_Ada95:
>     case DW_LANG_Cobol74:
>     case DW_LANG_Cobol85:
>     case DW_LANG_Pascal83:
>     case DW_LANG_Modula2:
>     default:
>       cu_language = language_unknown;
>       break;

I am inclined to agree.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer