Re: [WIP]: LOC_COMPUTED and LOC_COMPUTED_ARG

[Daniel Berlin <dan@dberlin.org>]

On Sat, 6 Apr 2002, Andrew Cagney wrote:

> > and
> >> 
> > 
> >> > - the frame base address (for DW_OP_fbreg)
> >> > 
> >> > 
> >> > Not possible.
> >> > the frame base can be a location list.
> >> > That's why it pulls it out of the frame function on the fly. 
> > 
> >> 
> >> Something I've never understood.  Shouldn't frame_base be stored in 
> >> frame->base as part of the initial frame creation.  Hence avoiding this 
> >> recursion?
> > 
> > 
> > Theoretically you could, but it's sort of pointless.
> > You still end up doing the same lookup.
> > It also hides the code, since the code to create frames is nowhere 
> > near the dwarf2 reader.
> 
> I'm still missing something.  If dwarf2 info isn't used to compute 
> frame->base (and frame->prev->pc) where are the values comming from?

I'm confused now.
I grepped for frame->base, and find no occurrences of it.

Anyway, DW_OP_fbreg has a special meaning.
It specifically means "Use the value you get from evaluating the 
DW_AT_frame_base attribute of the current function".  So that's what we do.  
Even more specifically, "The DW_OP_fbreg operation provides a signed 
LEB128 offset from the address specified by the location description in 
the DW_AT_frame_base attribute of the current function."

I'm just doing what the spec says, nothing more, nothing less.
:)


> > Andrew
>