[Bug c++/14998] GDB cannot handle pointer to member functions being encoded with DW_TAG_ptr_to_member_type

["dblaikie at gmail dot com" <sourceware-bugzilla@sourceware.org>]

http://sourceware.org/bugzilla/show_bug.cgi?id=14998

--- Comment #9 from David Blaikie <dblaikie at gmail dot com> 2013-01-07 19:41:03 UTC ---
(In reply to comment #8)
> (In reply to comment #7)
> > The DWARF standard is a bit vague on this:
> > 
> > "The pointer to member entry has a DW_AT_type attribute to describe the type of
> > the class or structure member to which objects of this type may point."
> > 
> > but it seems like if the DW_AT_type of a pointer-to-member variable "int
> > foo::*x;" is simply "int", then the DW_AT_type of a pointer-to-member function
> > "int (foo::*x)(float)" would be "int(float)".
> 
> I searched the DWARF standard and didn't see anything requiring this.
> However, it seems strange to emit the artificial parameter for a method
> but not for a pointer-to-method.

That's a fair point & at least convinces me to spend the time trying to get the
extra artificial parameter from Clang.

> Certainly doable in gdb, but I think it is nicer to agree on a single
> standard interpretation and try to avoid this kind of adaptive behavior.

Fair enough/agreed. Since I'm going to try to at least experiment with this (to
verify that GDB will play nicely once it has the implicit first parameter)
there's no real reason for GDB to be adaptive - once I've done the work I might
as well just commit it to Clang & thus be backwards compatible with GDB
versions & there's no real reason to rip it out again later.

> 
> > I guess LINKAGE_NAME is DW_AT_MIPS_linkage_name? Though in any case a function
> > pointer won't have a known/fixed name for the function, so I guess it's always
> > 'zero'/always displayed?
> 
> It seems to be a boolean flag (to my surprise, I didn't actually read
> it closely the first time).
> There's probably some way to arrange for this to work properly.
> 
> Offhand I don't know why it is this way.  It might just be so that people
> can tell "what is really going on" -- but a flag would suffice for this.

OK, sounds good.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.