This is the mail archive of the gdb-patches@sources.redhat.com mailing list for the GDB 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: Start Fortran support for variable objects.


On Fri, Jul 01, 2005 at 09:24:17AM +1200, Nick Roberts wrote:
>  > Well, that's what was SUPPOSED to happen, anyway.
>  > 
>  >  <1><a52>: Abbrev Number: 4 (DW_TAG_array_type)
>  >      DW_AT_sibling     : <a62>  
>  >      DW_AT_type        : <a62>  
>  >  <2><a5b>: Abbrev Number: 5 (DW_TAG_subrange_type)
>  >      DW_AT_type        : <a62>  
>  >      DW_AT_upper_bound : 4      
> 
> This is Dwarf output? Using something like readelf? (I'm just guessing).

That's right.

>  > I would have expected there to be a lower bound also... but there just
>  > isn't.  Your patch is more or less OK.  Let me reply to it to pick up
>  > some formatting comments.
> 
> For future reference, are you saying that there should be some internal
> representation for arrays that GDB uses which should describe the offset;
> that this could be used to print the variable object without reference
> to the language; and that this method should work with other unsupported
> (by varobj.c) languages like Pascal, for example?

I'd like that.  Varobj shouldn't need to know this; the rest of GDB
should be able to provide it.  But that's just my design instinct
speaking.  It may not be worth the trouble putting all the pieces
together.

-- 
Daniel Jacobowitz
CodeSourcery, LLC


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