Re: V3 patch causes EH failures in ILP32 mode

["B. Kosnik" <bkoz at nabi dot net>]

>Ugh.  Do you really think this kind of change is worth it?  I'm all for 
>Scott Meyers, but now every time we add a data member we've got to 
>remember to update these functions...

Yes, I think this change (-Weffc++ support) was worth it. This
diagnostic, when run over the libstdc++ sources, was/is quite useful,
catching several bugs and some ABI-impacting issues before release.
Personally, I regret not having this diagnostic working for 3.2.0, it
would have caught some obvious bugs for that release series. Cutting
down on the generated noise makes it more useful.

As for the maintenance burden, I consider it quite minor. Presumably,
people hacking this file know the basics of C++, and can do the updates
themselves. If not, I'll catch it myself at release time.

>> I wasn't expecting any issues like this to surface.
>
>It seems to be an interaction between the explicit test in the "operator 
>=" (which, as you note, is the major difference from what the compiler 
>does) and the use of "__restrict" in several places in that file. 
>(Removing either one makes the problem go away.)

Perhaps you should just remove the explicit test for self in operator= then....

best,
benjamin