Meeting tomorrow (31 August)

[Jim Dehnert]

NOTE:  we have a new location, in SGI building 41 (see directions on
the contact page and give yourself a few extra minutes to find it).
The phone number has also changed, to (650) 933-7976.

Following is an updated agenda.  The status pages are updated and on
the web (HTML and PDF) at:

      http://reality.sgi.com/dehnert_engr/cxx/cxx-summary.html

The contact info page is now also there, in password-protected PDF only
for privacy.  The password is "not an orc".

As usual, most of the updates are in the ABI layout document, in red.
The significant changes from last meeting are:

  -- Added Level II Exception specification, abi-eh.html Chapter 2.

Please take a careful look at the colored text, and raise any issues
you observe.

Take a look at the material associated with the following issues.
If you see something, please send it by email.  #7 (exceptions) is most
important, and I expect to devote at least half of the meeting to it
unless noone has issues with it.  Please look it over carefully.

  1) Mangling grammar modifications.  I think we match Alex's now.  Is it OK?
     Should we suppress result types for constructors in mangling?
     What should we do about Alex's local-name modification proposal?

  2) RTTI -- should member names be normative or not?
     Must NTBS names be COMDAT?
     Should std::type_info::__type_name be const?
     How about std::type_info::__virtual_table?

  3) Array constructors (__cxa_vec_new2/3):  If dealloc() throws, what
     should the result be?  Undefined, as when the destructor throws?

  4) __cxa_demangle -- Should we specify what this does with
     ambiguities between "C" variable names and builtin type manglings,
     e.g. for "i"?

  5) C-19:  Should we uniformly allow NULL ctor/dtor parameters to
     the runtime create/destroy utility routines?  Mark sent a list of
     potential changes.

  6) G-4:  Thread-safe data initialization.  See 3.3.2.  Christophe
     had some ideas for optimization here.  I've updated 3.3.2 a bit,
     but the discussion of potential changes for Christophe's concerns
     is in the open issues page for issue G-4.  Christophe, can you
     take a look at it?

  7) Exception handling (D-14):  I believe I've captured, in the Level
     II chapter of the ABI exception handling spec, what needs to be
     there from the HP document.  Please look it over carefully from
     that standpoint -- does it include what is necessary to achieve a
     consistent exception runtime library interface, but exclude what
     needs only be agreed between the personality routine, the LSDA,
     and the generated code (which all come from the same
     implementation)?  I will next tackle a description of the LSDA in
     the Level III chapter, but may or may not attempt to transfer the
     rest of the HP document there, as opposed to just referencing it.
     Does anyone think that would be significantly better?

     We'll go over any comments anyone has at the meeting, but given
     how close that is, and how major the subject, I expect that we
     will end up doing it more carefully at the August 31 meeting.
     Since I hope that will be the only significant business at that
     meeting, please let us know if you care about it and won't be
     available that day, so we can plan to deal with it the next time
     and perhaps cancel the 31 August meeting.

Note that it doesn't usually work to try to contact me by email
or phone the morning of the meeting, as I come directly from home.  If
you come in physically to the meeting, and the receptionist won't let
you come back to the conference room on your own, have her call the
conference room (it has a phone other than the one we use for the
meeting).

Jim

-		Jim Dehnert  x3-4272