Re: [RFA]: LOC_COMPUTED + abstracted dwarf2 evaluator, again

[Daniel Berlin <dberlin at dberlin dot org>]

On 9 Jul 2002, Jim Blandy wrote:

> 
> Daniel Berlin <dberlin@dberlin.org> writes:
> > + /* Evaluate a location description, given in THEBLOCK, in the
> > +    context of frame FRAME. */
> > + static struct value *
> > + evaluate_loc_desc (struct symbol *var, struct frame_info *frame,
> > + 		   struct dwarf_block *theblock, struct type *type)
> > + {
> > +   CORE_ADDR result;
> > +   struct value * retval;
> > +   struct dwarf_expr_baton *baton = xmalloc (sizeof (struct dwarf_expr_baton));
> > +   struct dwarf_expr_context *ctx;
> > +   
> > +   retval = allocate_value(type);
> > +   baton->var = var;
> > +   baton->frame = frame;
> > +   
> > +   VALUE_LVAL (retval) = lval_memory;
> > +   VALUE_BFD_SECTION (retval) = SYMBOL_BFD_SECTION (var);
> > +   
> > +   ctx = new_dwarf_expr_context ();
> > +   ctx->read_reg_baton = baton;
> > +   ctx->read_reg = dwarf_expr_read_reg;
> > +   ctx->read_mem_baton = baton;
> > +   ctx->read_mem = dwarf_expr_read_mem;
> > +   ctx->get_frame_base_baton = baton;
> > +   ctx->get_frame_base = dwarf_expr_frame_base;
> > +   ctx->error = error;
> > +   
> > +   dwarf_expr_eval (ctx, theblock->data, theblock->size);
> > +   
> > +   if (ctx->in_reg)
> > +     {
> > +       store_typed_address (VALUE_CONTENTS_RAW (retval), 
> > + 			   SYMBOL_TYPE (var), dwarf_expr_fetch (ctx, 0));
> > +       VALUE_LVAL (retval) = not_lval;
> > +     }
> > +   else
> > +     {
> > +       result = dwarf_expr_fetch (ctx, 0);
> > +       VALUE_LAZY (retval) = 1;
> > +       VALUE_ADDRESS (retval) = result;
> > +     }
> 
> This looks wrong.  If evaluating the Dwarf location expression yields
> an `in_reg' result, shouldn't the value be an lval_register, with its
> the address set to the register number?

This doesn't work, in reality.

IIRC, you run into problems dereferencing the value, because it 
thinks it *only* lives in a register (IE it's not really a memory 
location), regardless of what you tell it.

I may be misremembering the exact reason why, but I know things don't work 
if you do this, even though it would seem to make sense (I had originally 
written it to do what you suggest).


> 
> I think the evaluator is confused in this regard, too; see my upcoming
> post about that.
> 
>