This is the mail archive of the
elfutils-devel@sourceware.org
mailing list for the elfutils project.
Re: dwarflint versus linkage_name attributes
- From: Roland McGrath <roland at hack dot frob dot com>
- To: elfutils-devel at lists dot fedorahosted dot org
- Date: Mon, 04 Apr 2011 07:10:00 -0700
- Subject: Re: dwarflint versus linkage_name attributes
> - DW_TAG_variables, which do have a linkage_name, can be found in symtab
> (not checked by the code, just my manual spot checking), but don't have
> DW_AT_external set. I think this is just a gcc bug.
If the symbol is STB_LOCAL, then the bug is emitting linkage_name in the
DWARF where it's not needed. If the symbol is global, then the bug is
failing to emit linkage_name where it is needed.
> - DW_TAG_union_type and DW_TAG_structure_type with DW_AT_linkage_name
> set (but not DW_AT_external, which on itself makes sense). These are
> unexpected because the dwarf spec doesn't explain why or how these would
> be necessary.
I can imagine these give the mangled name of the type, i.e. the prefix of
the mangled names of its methods et al. That could conceivably be of use
to something composing new names. But you would think that a debugger
would only be considering existing things (methods, member variables,
etc.), which should have their own DIEs with linkage_name anyway.
Thanks,
Roland