This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Converting GCC to compilation with C++
- From: Robert Dewar <dewar at gnat dot com>
- To: gdr at acm dot org
- Cc: Paolo Bonzini <bonzini at gnu dot org>, gcc at gcc dot gnu dot org,Joe Buck <joe dot buck at synopsys dot com>,Mark Mitchell <mark at codesourcery dot com>, nathan at codesourcery dot com
- Date: Tue, 13 Jul 2004 17:25:44 -0400
- Subject: Re: Converting GCC to compilation with C++
- References: <40F2C9D5.20703@codesourcery.com> <32805.::ffff:128.194.146.35.1089667109.squirrel@webmail.nerim.net> <20040712142843.A7657@synopsys.com> <40F34A8D.2010601@gnat.com> <32794.::ffff:24.250.169.187.1089704266.squirrel@webmail.nerim.net> <40F3F25F.4070507@gnat.com> <40F3F75A.409@gnu.org> <33027.::ffff:24.250.169.187.1089738114.squirrel@webmail.nerim.net>
I'm amazed at seeing how people prefer brittle hacks to working
codes. Virtual functions are guranteed to be set properly,
once for while. Currently, we do not have such invariants and
we have to do everything hands with more opportunities to make
mistakes.
Well it is interesting to note that the use of dynamic dispatching
in safety-critical programs is extremetly problematic, and it is
not at all clear that FAA certification could be achieved for
programs using such techniques. Why? Because MCDC testing is
very difficult for such programs. In Ada, people often entirely
reject the use of the USE clause since they want to know immediately
where any external reference, including most importantly calls,
is going, so they want fully qualified names designating the
declaration target. Virtual functions really conflict with that
goal. So far from seeming un-brittle, to me dynamic dispatching
can be very troublesome.
Sure, really good coders can do things clearly in any language,
I have delivered perhaps 600-700,000 lines of assembly language
programs for commercial environments, including one complete
operating system, that I consider to be better organized and
better documented than 99% of the code I see in any other
language, but that does not mean I recommend that things
be coded in assembly language :-)