Re: [PATCH 00/36] Support building GDB as a C++ program

[Pedro Alves <palves at redhat dot com>]

A quick status update on this series.  All patches but one
up to the TRY_CATCH part are now pushed.  That is,

    > [PATCH 01/36] Create libiberty/libiberty.m4, have GDB and GDBserver use it
    > [PATCH 02/36] Add --enable-build-with-cxx configure switch
    > [PATCH 03/36] C++ keyword cleanliness, mostly auto-generated
    > [PATCH 04/36] Fix struct, union, and enum nesting in C++
    > [PATCH 05/36] Fix redefinition errors in C++ mode
    > [PATCH 06/36] record-btrace.c: Remove redefinitions
    > [PATCH 07/36] Make array object extern
    > [PATCH 08/36] elf-bfd.h: Wrap in extern "C".
    > [PATCH 09/36] floatformat.h: Wrap in extern "C".
    > [PATCH 10/36] Add extern "C" to declarations of C symbols
    > [PATCH 11/36] Make functions and variables exported by the IPA be extern "C"
    > [PATCH 12/36] proc-service, extern "C"
    > [PATCH 13/36] target.h: Include infrun.h
    > [PATCH 14/36] Do not do arithmetic on enum types
    > [PATCH 15/36] Don't forward declare enum target_hw_bp_type
    > [PATCH 16/36] x86 Linux/ptrace: fix offsetof usage in C++ mode
    > [PATCH 17/36] mi/mi-cmd-stack.c|frame filters: print_values <-> ext_lang_frame_args
    > [PATCH 18/36] Rename struct lzma_stream to avoid clash with system header
    > [PATCH 19/36] Exported const objects
    > [PATCH 21/36] opcodes/microblaze: Rename 'or', 'and', 'xor' to avoid C++ conflict
    > [PATCH 22/36] Remove duplicate const
    > [PATCH 23/36] gdbarch.h: include regcache.h
    > [PATCH 24/36] breakpoint.h: move enum Ãprint_stop_actionÃ
    > [PATCH 25/36] python/python-internal.h: enum Ãext_lang_rcà not defined
    > [PATCH 26/36] Adjust self tests to cope with GDB built as a C++ program
    > [PATCH 27/36] catch_command_errors: Remove 'mask' parameter
    > [PATCH 28/36] Move exception_none to common code, and use it

All the above are now pushed.

I didn't push this one, as it introduces an aliasing violation:

    > [PATCH 20/36] gdbserver/tracepoint: Add cast sockaddr_un/sockaddr cast

These are still pending:

    > [PATCH 29/36] Normalize TRY_CATCH exception handling block
    > [PATCH 30/36] quit_force: Replace TRY_CATCH wrapper macros
    > [PATCH 31/36] Split TRY_CATCH into TRY + CATCH
    > [PATCH 32/36] TRY_CATCH -> TRY+CATCH+END_CATCH everywhere
    > [PATCH 33/36] TRY_CATCH -> TRY+CATCH+END_CATCH, the manual conversions
    > [PATCH 34/36] more making TRY/CATCH callers look more like real C++ try/catch blocks
    > [PATCH 35/36] kill volatile struct gdb_exception
    > [PATCH 36/36] Make TRY/CATCH use real C++ try/catch in C++ mode

I don't have time right now to rebase this part, but I'd like
to move ahead with it sometime soon.  If anyone has comments on
this, now's the time to send them out.

> Known problems:
> 
> . Other hosts/native targets will naturally stumble on more
>   host-specific code that needs converting, which I can't easily test.
>   Help very much welcome!

This is still true.  :-)

> 
> . '--enable-targets=all' doesn't link yet in C++ mode.  This just more
>   extern "C" problems.

This is fixed, or will be, once the TRY_CATCH patches land,
at least on x86_64 GNU/Linux.  It was handled with
a couple bfd and opcodes patches.

>   
> I wrote a second series a few months ago that built on an older
> version of this one and fixes all the '-fpermissive' errors/warnings,
> until GDB and GDBserver build cleanly with -Werror, on x86_64 Fedora.
> It's naturally rotten in a few places by now, and in need of some
> further cleaning up but it shouldn't be that far off.  You can find it
> here:
> 
>   https://github.com/palves/gdb/commits/palves/cxx-conversion-attempt-part-2-no-fpermissive
>   git@github.com:palves/gdb.git palves/cxx-conversion-attempt-part-2-no-fpermissive

A couple weeks ago I rebased this second part and cleaned it up
substantially.  It's not fully ready, and now needs
another rebase though...

Thanks,
Pedro Alves