On Fri, Mar 24, 2017 at 09:29:00AM +0100, Richard Biener wrote:
> Yeah, the thing BLOCK_NONLOCALIZED_VARS wants to do is optimize generated
> dwarf by adding a DW_AT_abstract_origin (just to refer to the
> subprogram DIE) but
Well, for FUNCTION_DECLs in BLOCK_VARS/BLOCK_NONLOCALIZED_VARS we actually don't
emit any further DIE and so there is no DW_AT_abstract_origin.
E.g. gen_subprogram_die has:
/* Detect and ignore this case, where we are trying to output
something we have already output. */
if (get_AT (old_die, DW_AT_low_pc)
|| get_AT (old_die, DW_AT_ranges))
return;
That is why the posted testcase doesn't ICE without -fno-toplevel-reorder,
normally the body is emitted earlier and so we don't do anything at all.
Otherwise we just want to make sure we have a DIE and, if it is
inline/clone, have also DW_AT_inline set, and if the DIE is without parent
that we put it into proper place in the DIE tree. And when we actually
see the body of the function we fill locations and all other details.
Jakub
